Podcasts about eu ai act

What is the state of AI and videogames? Who is considering it? What are the big fails so far? This and much more is covered in this 1st episode of AI and videogames.   Buy me a coffee https://ko-fi.com/datascience Discord Channel: https://discord.gg/4UNKGf3   ✨ Connect with us! Personal newsletter: https://defragzone.substack.com

What is "wrong" with the EU AI Act? Watch the full episode and get in touch.   Buy me a coffee https://ko-fi.com/datascience Discord Channel: https://discord.gg/4UNKGf3   ✨ Connect with us! Personal newsletter: https://defragzone.substack.com

Safe, reliable, production-ready AI agents don't come from the model. They come from everything built around it.Rasmus Hauch is CTO of Boost.ai, one of the enterprise platforms banks, insurers and public sector organisations use to build customer-facing AI agents. His work sits one layer below every deployment: the orchestration, the guardrails, the testing infrastructure and the model choices that decide whether an agent holds up in front of millions of customers.In this episode, Rasmus opens that layer, explaining why conversational AI is shifting away from traditional conversation flows towards orchestration, testing and continuous evaluation. He walks through how Boost approaches it in practice: persona-based simulation across thousands of digital and voice conversations, predefined test suites for jailbreaking attempts and financial advice, layered guardrails inside and outside the agent, and a trust loop that feeds production results back into the build.Rasmus is direct about the hard parts. We touch on why the big LLM providers can't yet deliver the consistent response times voice demands, which is why Boost runs fine-tuned smaller models on its own infrastructure for latency-critical voice work. Rasmus explains why he advises clients against plugging in their own models. We also discuss where liability sits under the EU AI Act when an agent gets it wrong, and why a volcano eruption in Iceland is a good example of what no model can be trained for.Towards the end, Rasmus shares what's exciting him right now, including adaptive voice, real-time language switching and where agent-to-agent protocols are heading.Show notes
Discover more about boost.aiFollow Rasmus on LinkedIn:https://www.linkedin.com/in/rasmushauchFollow Kane on LinkedIn:⁠⁠⁠https://www.linkedin.com/in/kanesimms⁠⁠⁠Subscribe to VUX World: ⁠⁠⁠⁠https://vuxworld.typeform.com/to/Qlo5aaeW?utm_source=podcast&utm_medium=audio⁠⁠Subscribe to The AI Ultimatum Substack: ⁠⁠⁠⁠https://open.substack.com/pub/kanesimms⁠

In de tweede aflevering van seizoen twee van Succesvol Werven met AI, de podcast van RecruitmentTech.nl in samenwerking met Ubeeo, schuift Inge Brattinga aan bij hosts Martijn Hemminga en Leon Buijsman. De jurist en onderzoeker aan de Juridische Hogeschool Avans-Fontys legt uit waarom ‘kan niet, mag niet' de verkeerde reflex is, wat je vooral niet in ChatGPT moet gooien, en waar in het recruitmentproces de EU AI Act écht om de hoek komt kijken.

An airline once argued in a tribunal that its chatbot was a separate legal entity, responsible for its own actions. It lost. Now the machines don't just talk — they act, execute multi-step tasks, and move money. Dr. Vivian Atud, economist and AI governance thinker, unpacks Yale CELI's landmark cross-industry research showing 2026 as the year agentic AI shifted from capability to execution — while governance lagged behind. Inside: what the EU AI Act's August 2026 enforcement phase actually demands, why penalties reach 7% of global turnover, and The Accountability Chain™ — a five-link framework (Name It, Bound It, Log It, Kill It, Explain It) that every board should run this quarter. Evidence sources: Yale CELI / Fortune, EU AI Act Articles 9, 13, 50, 99; Moffatt v. Air Canada 2024 BCCRT 149.agentic AI, AI governance, EU AI Act, AI liability, corporate governance, Jeffrey Sonnenfeld, Yale CELI, AI agents, Article 99 penalties, human oversight, AI audit trail, Moffatt v Air Canada, board risk, AI accountability, Dr. Vivian Atud, The Clarity Mandate0:00 — Cold open: the chatbot that claimed to be a person1:20 — Act I: 2026, the year the machine started doing (Yale CELI)4:30 — Act II: The Frankenstein problem and the empty chair7:30 — Act III: The regulatory clock — Article 99 and August 202610:30 — Act IV: The Accountability Chain™13:30 — The faith lens: the parable of the talents14:20 — Close and mandate“Accountability cannot be delegated to an algorithm.”“Autonomy is a technical property. Agency is a moral and legal one. Do not confuse them.”“An off switch nobody has ever pulled is a rumor, not a control.”“The steward who says the instrument acted on its own has not offered a defense. He has offered a confession.”“If you cannot list your agents in ten minutes, you don't have governance. You have hope.”Fortune / Yale CELI (May 2, 2026), Sonnenfeld, Henriques et al., part three of a four-part series: 2026 marks the shift from capability to execution; agents interact with external tools, execute multi-step tasks, learn and iterate; governance and regulatory policy move far more slowly. Six months of research, hundreds of company materials, dozens of interviews across financial services, healthcare, insurance, retail, supply chain, telecom, public sector.Yale Insights / Fortune (final part of series): the Cursor “Sam” support-agent incident, April 2025 — invented licensing policy, canceled subscriptions.Artificial Intelligence News (April 9, 2026): agents can act without a clear record of what, when, and why; requirements for agent identity, centralized verbose logs, policy checks, human oversight, seconds-fast revocation, vendor documentation; EU AI Act Articles 9 and 13; multi-agent chains hardest to trace.EU AI Act, Regulation 2024/1689, Article 99: €35M / 7% worldwide turnover (Art. 5 prohibited practices); €15M / 3% (operator obligations incl. Arts. 16, 26, 50); €7.5M / 1% (misleading information). SME cap inverts to the lower figure.Digital Omnibus provisional agreement (May 7, 2026): Annex III high-risk deadline deferred to 2 December 2027, pending formal adoption; Article 99 fine provisions and market-surveillance enforcement powers apply from 2 August 2026.Moffatt v. Air Canada, 2024 BCCRT 149: negligent misrepresentation; tribunal rejected the argument that the chatbot was a separate legal entity; damages of C$812.02.note: all figures verified against primary regulatory text (Article 99) plus at least three independent legal commentaries. The Digital Omnibus deferral is a provisional agreement pending formal adoption — the script states this explicitly on-air rather than asserting a settled deadline. Institutional audiences reward that precision; a competitor claiming a hard August 2026 high-risk deadline is overstating the record.

What happens when the brightest minds in legal, technology, and innovation come together to shape the future of our profession? Welcome to the Legally Speaking Podcast Legal Tech Talk 2026 miniseries. As the UK's number one legal podcast and show ranked in the top 1% of podcasts globally, we're delighted to be returning as the official media partner for Legal Tech Talk for the third time. Having supported this event since the beginning. We've witnessed firsthand how it has become Europe's largest legal technology conference, bringing together the people, ideas, and innovations transforming the legal industry. Across this special series, we'll be speaking with some of the world's leading legal professionals, founders, technologists, change makers to explore what's next for law. Because our mission remains the same. Inspiring legal minds, transforming legal futures. Let's get into it. Jeffrey Shapiro is a legal tech and future-of-law thinker who talks about how AI is changing the legal industry, but not just as a tool problem — more as a mindset, culture, and integration challenge. He focuses on breaking down silos so firms can use data intelligently, staying ahead of regulation like the EU AI Act, and keeping human trust and authenticity central as technology advances. He also emphasises curiosity, continuous learning, and community as key traits for future legal leaders, often explaining complex ideas in memorable, down-to-earth ways. So why should you be listening in? You can hear Rob and Jeffrey discussing:- Cross-silo data integration matters- Human connection builds trust- Regulatory agility is critical- Curiosity shapes future leadership- Openness improves collaboration.Connect with Jeffrey Shapiro here - https://www.linkedin.com/in/jtshap/

AI takes over (kinda). Regulators look for a new job. Watch the full episode and get in touch.   Buy me a coffee https://ko-fi.com/datascience Discord Channel: https://discord.gg/4UNKGf3   ✨ Connect with us! Personal newsletter: https://defragzone.substack.com

Một năm trước, trong diễn văn khai mạc Tuần lễ Công nghệ Luân Đôn (London Tech Week 09-13/06/2025), thủ tướng Anh Keir Starmer đã nêu ra chiến lược lấy việc phát triển trí tuệ nhân tạo AI làm bàn đạp thúc đẩy tăng trưởng kinh tế giai đoạn 2025-2030. Sau một năm lên kế hoạch, với khoản đầu tư gần 100 tỷ bảng vào các trung tâm dữ liệu, vào y tế và giáo dục, ngành AI của Anh đã đạt được tiến bộ gì và giúp cho tăng trưởng kinh tế tới đâu? Thông tín viên Nguyễn Giang từ Luân Đôn giải thích: Anh Quốc đánh cược vào AI RFI : Đầu tiên, xin Nguyễn Giang nói về việc triển khai chiến lược AI của Vương quốc Anh?   Nguyễn Giang : Quan sát những gì Anh quốc đã và đang làm hơn một năm qua, tôi thấy họ đặt trọng tâm của kế hoạch phát triển AI vào các Khu vực Tăng trưởng Ai, tức AI Growth Zones, gọi tắt là AHGZ, với ba mục tiêu: hỗ trợ tăng trưởng kinh tế, tạo đột phá công nghệ và cân bằng, kết nối các vùng miền của Vương quốc Anh. Mô hình AI Growth Zones (Khu vực Tăng trưởng AI) tạo ra các cụm siêu trung tâm dữ liệu (hyperscale data centres) và điện toán hiệu năng cao. Chính phủ Anh đã phê duyệt hàng loạt khu vực trọng điểm ở hạt Oxfordshire (Culham - trung tâm đầu tiên), Đông Bắc Anh (Blyth & North Tyneside), xứ Wales (Wylfa) và gần đây nhất là Scotland (khu vực Lanarkshire - tháng 1/2026). Họ chia ra để mọi khu vực địa lý chính của Vương quốc Anh đều có cơ hội tiếp cận AI.  Khu quan trọng nhất là hành lang công nghệ Oxford-Cambridge (thực tế là cùng Luân Đôn lập ra một Tam giác High-Tech). Xin nhắc là từ Luân Đôn tới Cambridge và Oxford khoảng cách chỉ chừng hơn 100 km. Các đại học lớn, kể cả phòng thí nghiệm nguyên tử quan trọng nhất của Anh (Arbington), cũng nằm trong tam giác này. Điều này giảm được chi phí giao thông, hút được nguồn nhân lực chất lượng cao và tạo điều kiện cho việc kết nối băng thông cho hệ thống điện toán. Nhưng Anh cũng lập các vùng AI Growth Zone ở cả Wales, miền bắc Anh và Scotland là có ý duy trì nhiều động cơ kinh tế cho các vùng đất kia. RFI: Cho tới nay thì các khoản đầu tư được khu vực tư nhân đăng ký vào các khu vực tăng trưởng AI này là bao nhiêu? Nguyễn Giang : Mục tiêu chung là phải đạt tới 100 tỷ bảng Anh đầu tư từ khối tư nhân vào hạ tầng AI thông qua cơ chế này. Ở vùng Đông Bắc (North East AIGZ), quỹ đầu tư Blackstone đã cam kết bỏ ra10 tỷ bảng Anh để xây dựng trung tâm dữ liệu tại Blyth, đồng thời mở đường cho 20 tỷ bảng bổ sung từ các đối tác khác. Chính phủ Anh cũng trực tiếp chi ngân sách tài trợ ban đầu hơn 2,5 tỷ bảng để kích hoạt hệ thống điện toán quốc gia. Điều đáng chú ý là phần cam kết từ cả hai phía. Về phía mình, chính phủ Anh cam kết giá điện đặc biệt cho data centre tại các vùng có thặng dư năng lượng (ví dụ: giảm từ 14 đến 24 bảng/MWh tùy khu vực, tiết kiệm tới 80 triệu bảng/năm cho một trung tâm 500MW). Để đẩy nhanh tiến độ thiết kế các trung tâm dữ liệu, họ ưu tiên đưa chúng vào danh mục Hạ tầng Quốc gia Trọng yếu (Critical National Infrastructure - CNI), cho phép rút ngắn thời gian chờ đấu nối vào lưới điện quốc gia lên đến 5 năm và cắt giảm quy trình phê duyệt quy hoạch từ 18 tháng xuống còn 12 tháng. Còn về phía các nhà đầu tư cùng đối tác ở địa phương thì họ tìm cách bố trí quỹ đất tối thiểu 100 mẫu Anh (acres) đã được dọn sẵn hoặc đủ tiêu chuẩn sẵn sàng xây dựng trước 2028, cam kết tạo ra hàng nghìn việc làm chất lượng cao (như vùng North East cam kết tạo 5.000 việc làm), và phối hợp với các đại học địa phương (Newcastle, Durham...) xây dựng chương trình đào tạo nhân lực AI. Họ cam kết sẽ kết nối với các đại học địa phương. Hiện đang làm nghiên cứu trong một đại học ở Luân Đôn và đã dự nhiều khóa học, hội thảo, tôi thấy một xu hướng rõ rệt: các đại học Anh đang xóa bỏ, thậm chí đóng nhiều khoa ở các trường khắp cả nước trong các bộ môn khoa học xã hội, nghệ thuật, ngôn ngữ, nhưng lại đang mở rất nhiều khóa, ngành liên quan đến trí tuệ nhân tạo, từ mảng nghiên cứu, phát triển công nghệ tới ứng dụng Ai trong đủ mọi lĩnh vực. Xem ra đây là tương lai mới cho giáo dục đại học. RFI: Một câu chuyện nhiều nước nói tới là các trung tâm dữ liệu tiêu thụ điện năng rất lớn. Vậy ở Anh, nguồn năng lượng điện đến từ đâu? Có nằm cạnh nhà máy điện hạt nhân? Nguyễn Giang : Đúng thế, năng lượng là "nút thắt" lớn nhất của hạ tầng AI, và chiến lược của Anh là áp sát nguồn phát năng lượng sạch quy mô lớn, nhưng không chỉ giới hạn ở điện hạt nhân. Điện nguyên tử không phải tất cả, nhưng đây là một trục xương sống cho AI. Ví dụ điển hình nhất là North Wales AI Growth Zone, nơi được thiết kế cặp đôi chiến lược (strategically paired) nằm ngay cạnh dự án Lò phản ứng mô-đun nhỏ (SMR) đầu tiên của Anh tại Wylfa. Sự kết hợp này nhằm cung cấp nguồn điện tải nền (baseload) cực kỳ ổn định và khổng lồ cho các cụm chip AI mà không làm quá tải lưới điện dân dụng. Ngoài ra thì tại vùng North East (Blyth), các trung tâm dữ liệu dùng nguồn điện đến từ các trạm tiếp bờ của điện gió ngoài khơi lớn nhất thế giới ở Biển Bắc (North Sea) và mua cả năng lượng qua cáp ngầm kết nối thủy điện từ Na Uy. Tại Scotland, họ tận dụng nguồn năng lượng tái tạo thặng dư (gió và thủy điện) vốn thường bị nghẽn, không thể truyền tải về miền Nam. Các báo Anh gần đây hay đăng các câu chuyện thú vị như nhà nông nào đó ở vùng quê bán trang trại, lò sản xuất bánh mì, trạm xay bột cho công ty Ai để xây data centre. Theo một điều tra của nhà báo Henry Mawhood cho trang báo nghề cá ở Anh thì thực ra câu chuyện không đơn giản như vậy nhưng đúng là các vùng ven biển phía Tây đảo Anh, từ Weston-Super-Mare tới Cumbria (England), rồi miền Nam xứ Wales và duyên hải Đông Bắc Scotland, người ta đang có kế hoạch xây hai thứ: các nhà máy điện nguyên tử mini, các trạm nhận điện qua cáp ngầm từ Bắc Âu (Na Uy) và các trạm đón tuyến cáp truyền dữ liệu xuyên Đại Tây Dương, nối Anh với Hoa Kỳ. Vì thế, đúng là có những nơi nhà nông được mời chào bán lại các miếng đất, nhà xưởng, ít nhất với diện tích 1-2 trăm mét vuông, để chứa data centre. Trong tháng 6 năm nay, có câu chuyện mang tính biểu tượng ở Anh. Thủ tướng Kier Starmer đã thăm một nhà máy xà phòng cũ có 130 năm tuổi ở Warrington, hạt Cheshire, để dự lễ khai trương dự án trung tâm dữ  liệu. Tức là họ biến một điểm công nghiệp rất là cũ thành trung tâm dữ liệu, dùng nguồn điện, nước có sẵn.  RFI: Theo anh, công suất cho điện dùng vào trí tuệ nhân tạo ở Anh có đủ không? Nguyễn Giang : Theo các báo chuyên ngành đánh giá thì điện ở Anh, gồm cả điện nguyên tử, hiện tại là đủ cho lộ trình giai đoạn 1 nhờ chiến lược phân bổ vùng. Ví dụ, riêng trung tâm dữ liệu AI tại Blyth đang tăng tốc để đạt công suất 1,1 GW (Gigawatt) trong vòng 6 năm tới, biến nó thành một trong những cụm dữ liệu lớn nhất châu Âu. Bằng cách đặt data centre tại nơi "nguồn phát" dư thừa, Anh giải quyết được bài toán tổn thất truyền tải và giá điện cao. RFI: Cho tới nay, việc sử dụng AI trong y tế, tài chính và giáo dục tại Anh ra sao? Nguyễn Giang : Theo chỉ số tâm lý AI (EY AI Sentiment Index) tính đến giữa năm 2026, bức tranh ứng dụng của Anh như sau: Trong Dịch vụ Sức khỏe Quốc gia (NHS- tức Y tế công) thì kế hoạch 10 năm của Anh đã đưa AI từ giai đoạn thí điểm sang tích dợp hệ thống toàn diện. AI hiện là "trợ lý số" bắt buộc ở nhiều khâu: các công cụ "Ambient Scribing" (AI ghi âm cuộc thoại bác sĩ - bệnh nhân tự động chuyển thành hồ sơ bệnh án) phổ cập mạnh mẽ; ứng dụng NHS App tích hợp AI thành cổng sàng lọc bệnh tự động. Đặc biệt, Bộ Y tế Anh vừa tung ra mô hình "MHRA AI Airlock / Sandbox" (tháng 6/2026) để thử nghiệm AI trong việc dự đoán tác dụng phụ của thuốc, giúp NHS giảm bớt gánh nặng chi phí 2 tỷ bảng/năm do phản ứng ngược của thuốc gây ra. Mạnh và sâu hơn là ứng dụng trong tài chính dùng công nghệ số (fintech). Trên thực tế thì tỷ lệ ứng dụng AI ở Anh trong ngành này đã đạt 75%. Các ngân hàng và công ty tài chính Anh đang dùng AI sâu trong phát hiện gian lận (anti-frauds), trong giao dịch tự động và tuân thủ pháp lý (compliance). Vào tháng 1/2026, Cơ quan Quản lý Tài chính (FCA) đã mở cuộc điều tra Mills Review để đánh giá toàn diện tác động dài hạn của AI đối với dịch vụ tài chính tiêu dùng, đồng thời tích hợp sâu công cụ Agentic AI (AI tự chủ) vào hệ thống vận hành. Trong giáo dục đại học, Anh đang chuyển mạnh từ các ngành xã hội sang công nghệ. Hiện Bộ Giáo dục đã chuẩn hóa hơn 400 chương trình thạc sĩ chuyên sâu về AI và 28 trung tâm đào tạo tiến sĩ (CDT). Các trường đại học lớn như Edinburgh (với Bayes Centre), Cambridge, Oxford hoạt động như các lõi R&D chuyển giao công nghệ trực tiếp cho doanh nghiệp trong các AI Growth Zones. RFI:  So với Liên Âu thì Anh Quốc đi trước hay sau về công nghệ AI trong ứng dụng vào kinh tế và các ngành nghề? Nguyễn Giang : Có thể nói về mặt thương mại, hạ tầng và đầu tư thì Anh đi trước. Anh hiện đứng đầu châu Âu và thứ 4 thế giới về năng lực nghiên cứu cũng như thu hút vốn đầu tư mạo hiểm cho AI (riêng năm 2024 đạt gần 3 tỷ bảng). Do lợi thế về nguồn dữ liệu tập trung của NHS và thị trường tài chính London, các tập đoàn Mỹ (Google, Microsoft, OpenAI) chọn Anh làm bãi đáp hạ tầng lớn hơn so với Liên Âu Ưu thế của Anh là có tốc độ tích hợp AI vào dịch vụ công như trong y tế công (NHS) đồng bộ hơn nhờ cơ chế quản lý tập trung, trong khi EU bị phân mảnh theo từng quốc gia thành viên. Nhưng so sánh Anh và Pháp thì bức tranh lại hơi khác RFI:  Anh và Pháp được cho là hai quốc gia châu Âu hành đầu, cạnh tranh với Hoa Kỳ, Trung Quốc về AI nhưng cũng cạnh tranh với nhau? Nguyễn Giang : Các thông tin do hai chính phủ Anh và Pháp công bố dần dần cho thấy sự khác biệt, ưu điểm và nhược điểm của mỗi bên trong chiến lược phát triển AI. Tôi xin điểm ra một số nhỏ: Anh đã triển khai các trung tâm dữ liệu nhanh hơn Pháp. Vì ở Pháp, dù Hội nghị thượng đỉnh Choose France cam kết cắt giảm thủ tục hành chính của Pháp, nhưng Pháp cũng bị ràng buộc bởi các quy định quản lý đất đai và bảo vệ môi trường chặt chẽ của Liên Hiệp Châu Âu nên thủ tục làm tốn thời gian hơn trong việc giải phóng mặt bằng quy mô lớn. Pháp muốn đi nhanh nhưng vẫn phải tuân thủ luật Liên Âu (EU AI Act), còn Anh thì có môi trường pháp lý thoáng hơn, gần với Hoa Kỳ hơn, và thái độ ‘pro-innovation' có vẻ đang tốt hơn Pháp. Trong việc xây dựng hệ sinh thái tài chính và thương mại kỹ thuật số thì Anh có sẵn Luân Đôn là trung tâm Fintech số 1 châu Âu, kết nối truyền thống mạnh mẽ trong thế giới nói tiếng Anh với New York, với Singapore, Hồng Kông nên giúp các AI Growth Zone dễ dàng kết nối trực tiếp giải pháp AI vào thị trường tài chính toàn cầu. Thêm vào đó, việc sở hữu Dữ liệu tập trung của Y tế công NHS là lợi thế độc quyền mà Pháp không có. Anh dễ dàng thu hút các tập đoàn y sinh AI lớn đến thiết lập trung tâm tính toán, vì họ có thể tiếp cận kho dữ liệu y tế khổng lồ hàng đầu thế giới thông qua các cơ chế an toàn số (như MHRA AI Airlock). Thế nhưng, về quy mô vốn đầu tư hạ tầng khổng lồ bằng tiền nước ngoài (FDI Magnitude) thì Pháp đang hơn Anh. Ví dụ, Hội nghị Choose France, tập đoàn SoftBank đã cam kết chi tới 75 tỷ euro (trong đó 45 tỷ euro giải ngân trước 2031) để xây dựng cụm trung tâm dữ liệu AI khổng lồ công suất 5 GW tại Dunkerque. Bên cạnh đó là 10 tỷ euro từ Brookfield và 7,5 tỷ euro từ quỹ MGX (UAE). Hiện chưa rõ cuộc chạy đua đường dài về AI (AI race) giữa Anh và Pháp ra sao, nhưng cả hai nước đều đang đổ tiền vào. Tổng thống Emmanuel Macron công bố trong tháng 6/2026 khoản đầu tư 1,55 tỷ euro tiền công vào công nghệ quantum và bán dẫn (semiconductors). Cùng lúc, chính phủ Anh cũng trực tiếp chi ngân sách tài trợ ban đầu hơn 2,5 tỷ bảng để kích hoạt hệ thống điện toán quốc gia kết nối AI. RFI: Nếu làm tốt, các kế hoạch này có giúp kinh tế Anh tăng trưởng giai đoạn 2027-2030 không? Nguyễn Giang : Anh Quốc đang hy vọng kế hoạch AI Growth Zones và chuyển đổi số này được thực thi trơn tru, Nhờ đó mà tác động lên giai đoạn 2027-2030 sẽ có thể phát huy tác dụng qua ba kênh: Một là người ta hy vọng có dòng vốn dịch chuyển mạnh từ Mỹ sang Anh, không chỉ với con số 100 tỷ bảng đầu tư hạ tầng cam kết đến năm 2030 và còn có các cam kết mới. Các đầu tư này sẽ trực tiếp đóng góp vào GDP thông qua xây dựng và mua sắm thiết bị công nghệ cao ở Anh. Hai là AI sẽ làm tăng hiệu suất của y tế và tài chính, dịch vụ công. Lấy ví dụ ứng dụng AI giúp rút ngắn thời gian chẩn đoán của bác sĩ, cắt ngắn thời gian chờ được khám bệnh, và gián tiếp giải phóng hàng tỷ bảng Anh áp lực ngân sách (thời gian, giờ công) cho phép tái đầu tư tiền vào các khu vực kinh tế tạo ra thặng dư cao hơn. Và thứ ba là tác động lan tỏa (Spillover effects). Đây là điều cần chờ xem. Thế nhưng theo các mô hình dự báo, khi các AI Growth Zones đi vào hoạt động ổn định từ năm 2027, các ngành kế toán, tài chính, năng lực tính toán trong sản xuất v.v. sẽ nhờ giải pháp AI mà giảm được chi phí. Qua đó, giá trị chung của toàn chuỗi cung ứng ở Anh sẽ tăng lên, cùng sức cạnh tranh của hàng hóa Anh. Chúng ta sẽ không thấy ngay lợi ích này, nhưng đây là quá trình chuyển đổi, có người nói là cuộc cách mạng cơ sở hạ tầng lớn nhất từ thế kỷ 16 nên cần phải có  thời gian để “hưởng lợi”, nhìn từ góc độ của Anh Quốc.

AI in hiring isn't a yes-or-no question, it's a risk-management one. Johnny talks with Heather Bussing, an employment attorney who's spent almost 40 years at the intersection of employment law and technology, about what's actually required under the EU AI Act and US state laws. They dig into why risk rises the closer AI gets to a hiring decision, why pre-interview sorting is the hardest area to get right, and how bias audits can protect you.Download Heather Bussing's Top Tips Here

Brando Benifei MEP, Co-rapporteur on the EU AI Act and Chair of the European Parliament Delegation for Relations with the US, talks to Paul Adamson about AI regulation, technological sovereignty and the state of transatlantic relations.

PMI schießt an der KI-Front aus allen Rohren: Jüngst einen 260-seitigen Standard, letztes Jahr einen „Practice Guide", dazu eine neue Zertifizierung und seit 2024 auch das eigene KI-System Infinity. Thomas Wuttke ordnet ein, was im Standard und im Guide drinsteht, wie die zusammenhängen und wo Unterschiede liegen. Außerdem im Fokus: Wie gut passt der neue Standard zum EU AI Act, der ab August 2026 für Hochrisiko-Systeme scharf geschaltet wird? Und die große Frage zum Schluss: Machen Standards bei einer Technologie, die sich alle sechs Monate neu erfindet, überhaupt noch Sinn? Shownotes: https://swiy.co/PTP239

Wir haben diese Revision mit Florian Sowade über den EU AI Act gesprochen. Florian hat sich in die Regulierung eingelesen und erklärt uns, was die EU damit eigentlich regulieren will, welche Risikokat…

KI diskriminiert nicht von selbst – sie wiederholt unsere Muster. Alex Ulbricht von Fujitsu hat über 400 Keynotes zu AI-Fairness gehalten. In dieser Folge geht es um den Amazon-Recruiting-Fall, um Crashtest-Dummies und Medikamentenforschung, um Fairness by Design und um den Disparate Impact Ratio, mit dem du Bias in einer einfachen Formel testen kannst. In dieser Folge erfährst du: → Warum KI keine Vorurteile erfindet, sondern historische Muster reproduziert → Was der Amazon-Recruiting-Fall von 2014 bis heute lehrt → Wie ein Bias-Audit abläuft – und was Intersectional Bias bedeutet → Warum Fairness von Anfang an ins Design gehört (wie Barrierefreiheit) → Wie du mit dem Disparate Impact Ratio in einer Formel auf Bias testest Über den Gast: Alex Ulbricht ist seit über zwölf Jahren bei Fujitsu und war dort die jüngste Vertriebsleiterin im Datacenter-Sales. Heute ist sie Keynote-Speakerin zum Thema AI-Fairness und hat über 400 Vorträge gehalten. MY DATA IS BETTER THAN YOURS ist ein Projekt von BETTER THAN YOURS, der Marke für richtig gute Podcasts.

„Was passiert, wenn es nicht klappt?“ Die Gründerrealität hinter AuditrIm Videocast PITCH & PEOPLE spricht Auditr-Gründer Elias Müller offen über die Schattenseiten des Gründens. Zwischen Vollzeitjob, schwindenden Ersparnissen und schlaflosen Nächten baut er sein Startup rund um den EU AI Act auf. Warum er trotzdem an seine Idee glaubt, wie er die ersten Kunden gewinnt und weshalb sich der erste Euro Umsatz für ihn besser anfühlen wird als jedes Gehalt.

Fintech is swimming in capital, yet the deal count collapses and the message is blunt: the market still funds big ideas, but it no longer rewards sloppy execution. We call this moment the “summer compression” a quiet reset where operational discipline, transparency, and provable trust replace the old era of growth-at-any-cost storytelling. If you build in payments, banking, compliance, or financial technology strategy, this is the shift that will decide who survives the next cycle. We dig into why Europe is forcing maturity with real deadlines: the EU AI Act makes explainability a live requirement for credit, risk, and fraud models, and instant payments push verification of payee as a baseline safety check. From there, we use Romania's fintech ecosystem to show how consolidation really looks on the ground: revenue concentration, a profitability gap, and a new hybrid consensus where banks bring scale and trust while fintechs provide the infrastructure plumbing. Then we zoom out to the modern growth playbook: a profitable neobank expands across markets by staying lean with an AI assistant embedded in operations and by focusing hard on one segment digital nomads and cross-border professionals. That focus ties directly to the next infrastructure upgrade, EPC's OCTINST and ISO 20022, which aim to make cross-border instant payments feel as seamless as domestic transfers. But speed creates a “speed trap” for fraud teams, so we explore agentic AI, “human plus agent” oversight, and what audit-ready governance must look like when money moves in seconds. If this helped you see where fintech is headed, subscribe, share the episode with a builder or investor, and leave a review. What part of the trust reset feels most urgent to you right now?Thank you for tuning into our podcast about global trends in the FinTech industry.Check out our podcast channel.Learn more about The Connector. Follow us on LinkedIn.CheersKoen Vanderhoydonkkoen.vanderhoydonk@jointheconnector.com#FinTech #RegTech #Scaleup #WealthTech

My co-host Ken Suzan and I are welcoming you the episode 176 of the IP Fridays Podcast. Today's interview guest is returning guest Franklin Graves, who is a senior counsel at Linkedin and teaching IP law at Emerson College. With my co-host Ken Suzan he is discussing how the law for creators has dramatically changed in the past years. Franklin Graves is expressing his personal views and not the views of Linkedin or Microsoft. He is talking about the paper “Upload Complete” before he joined Linkedin. Bio: https://www.linkedin.com/in/franklingraves/ Paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5271442 Website: https://creatoreconomylaw.com/ But before we jump into this interview, I have news for you! Richard Meade, a judge on the UK High Court and one of the most prominent figures in European patent law, was appointed Lord Justice of Appeal at the British Court of Appeal on June 12, 2026. Meade played a key role in numerous landmark British patent decisions, particularly in the area of standard-essential patents (SEPs) and FRAND licenses. In Insulet Corp. v. EOFlow Co., No. 2025-1807, the U.S. Court of Appeals for the Federal Circuit completely overturned the original $452 million judgment (which had already been reduced by the District Court to $59.4 million) in favor of Insulet. In its decision of June 2, 2026, in the case of Fujifilm v. Kodak, the UPC Board of Appeal provided comprehensive clarifications regarding so-called “long-arm jurisdiction”—that is, the question of whether the UPC can also rule on national patent claims outside the UPC territory (such as in the United Kingdom). In 14 guiding principles, the judges established specific procedural rules for various categories of cases. There is no automatic UPC jurisdiction over national patent claims outside the UPC territory. The Munich Regional Court has issued an arrest warrant against the managing director of Polytech Health & Aesthetics GmbH because he is alleged to have continued to exploit the Brazilian company Silimed's patent for breast implants despite a preliminary injunction. A number of IT and automotive industry associations—which are among the most frequent users of Inter Partes Reviews (IPR) at the U.S. Patent and Trademark Office—have filed an amicus brief with the Supreme Court, urging the Court to grant Google's certiorari petition. An attorney for a Las Vegas performer has asked a California federal judge to temporarily prohibit Taylor Swift from using “The Life of a Showgirl” as a trademark while the trademark lawsuit is pending. Swift's attorney called the lawsuit baseless. And now let's hear Ken discuss creator law with Franklin! AI, Platform Law, and the Creator Economy: What Businesses Need to Know Now Franklin Graves has spent his entire career watching digital content move through systems that most people never see. He started in marketing at a major music label right out of law school, then represented individual creators on YouTube in a pro bono capacity, then moved to the platform side at Eventbrite, and today works as Senior Product Counsel at LinkedIn, where he focuses on AI, data, and the regulatory questions that come with both. His recently published law review article, Upload Complete: An Introduction to Creator Economy Law, is the first academic paper to address the creator economy as a distinct legal field. In a recent episode of the IP Fridays podcast, he spoke with host Kenneth Suzan about responsible AI development, platform regulation, and what it actually means to own your audience in a world where the rules keep changing overnight. From Content Creator to Platform Lawyer The through-line in Graves’ career is a genuine understanding of how content moves from an idea in someone’s head to an audience on a screen. That experience, he argues, is precisely what in-house counsel needs right now. Lawyers working on AI and product development cannot afford to sit at a distance from the technology they are advising on. They need to use the tools, experience them as a creator or end user would, and understand the nuances of how a product actually operates before it reaches the public. Understanding the product first is the precondition for everything else. That philosophy translates directly into how he approaches responsible AI implementation. The landscape of AI standards is crowded: NIST frameworks, the EU AI Act, sector-specific guidance, and a growing body of industry-adopted best practices. The challenge for in-house counsel is not knowing that these standards exist. It is making them actionable for the engineering and product teams they support. Abstract principles need to become concrete controls and workflows. Graves offers one practical shortcut: most companies already have open source software review processes that involve the right stakeholders, the right sign-off levels, and the right security checks. Layering the specifics of generative AI or large language models onto those existing processes is far more efficient than building something new from scratch. A Fragmented Regulatory World The geopolitical dimension of AI regulation is something Graves thinks about constantly in his role at LinkedIn. The EU AI Act, shifting US executive orders, and country-specific approaches to data privacy have created a regulatory environment that can change the rules of the game without warning. His analogy is instructive: creators have long understood what it means to build a community on a platform they do not own. An algorithm change, a policy update, or a government ban can wipe out years of audience-building overnight. Businesses deploying AI tools globally now face a structurally similar problem. The response, for creators and for platforms alike, is to build resilience rather than rely on stability that may not last. TikTok is the clearest recent example. When the platform faced the prospect of being shut down in the United States on national security grounds, it triggered a broader conversation about platform dependence that had been building for years. Creators who had invested their entire business in one platform suddenly confronted the possibility that their audience could simply disappear. The lesson is not that platforms are bad. It is that concentration of any kind, whether it is your audience, your data pipeline, or your regulatory compliance strategy, creates fragility. What Is a Creator, Legally Speaking? One of the central contributions of Graves’ law review article is definitional. The terminology matters more than it might seem. When courts and regulators talk about creators without a shared understanding of what that word means, the resulting legal analysis tends to miss the mark. Graves draws a distinction between users who post content, creators who post with the intent to build an audience and eventually monetize it, and influencers, a subset of creators who are actively running a small business through their content. The difference is intent. A parent posting family photos on Facebook is a user. Someone building a subscription community around their professional expertise is running a business, and the legal framework that applies to them should reflect that. That distinction matters practically when it comes to liability. As more creators build their own platforms, whether through custom membership sites, open source tools like Ghost, or federated social networks, they take on obligations that previously fell to large platforms: content moderation policies, privacy notices, terms of service, and compliance with data regulations across multiple jurisdictions. A creator in Tennessee running a membership platform with subscribers in Germany is operating a global business, whether they think of themselves that way or not. Protecting Children Online: A Question Without a Clean Answer The tension between age verification and privacy is one of the more difficult problems in platform law right now. Australia, several European countries, and a growing number of US states have introduced or passed minimum age requirements for social media accounts. The technical challenge is real: verifying age online requires collecting identifying information, and collecting identifying information creates privacy risk, particularly for the young people the laws are designed to protect. Who should bear the responsibility for that verification is also unresolved. Is it the platform? The app store? The mobile operating system? Graves does not pretend there is a clean answer, but he points to the mobile layer as an underexplored option. The Apple App Store and Google Play Store already have significant leverage over which apps reach users on their devices. Whether that leverage should extend to age verification is a question that deserves more attention than it currently receives. The Right of Publicity in the Age of AI Voice cloning, digital replicas, and AI-generated synthetic media have pushed the right of publicity into territory that traditional IP law was not designed to cover. Trademark law, copyright law, and existing publicity rights each capture part of the problem but none of them covers it completely. The result, as Graves describes it, is a period of experimentation: lawyers filing trademarks on vocal sounds and phrases, states updating their publicity statutes to explicitly mention artificial intelligence, and entertainment unions negotiating over who controls a performance and any AI-generated iterations of it. Tennessee’s Elvis Act is a concrete example of the legislative response: the state updated its right of publicity law to include voice and to reference AI directly. Similar efforts are underway elsewhere. The underlying challenge is calibrating protection so that it gives creators and performers meaningful control over their likeness and voice without foreclosing the development of generative AI systems that depend on broad rights to process and learn from content. Somewhere between those two interests, a workable legal framework needs to emerge. The brand deal context may be where the issue becomes most immediately practical. When a brand partners with an influencer and the campaign involves generative AI in any form, the contract needs to address control explicitly. Who has final approval over how the influencer’s likeness or voice is used in AI-generated deliverables? What happens to those assets after the campaign ends? These are not hypothetical questions. They are contract drafting problems that any brand counsel or creator attorney should be addressing today. What Comes Next Graves is cautious about predictions, but his sense of direction is clear. The regulatory environment will continue to fragment before it converges. The right of publicity will be updated, imperfectly, in more jurisdictions. Creators will continue to move toward owning more of their infrastructure. And the lawyers who do this work best will be the ones who understand the technology well enough to translate it into practical, defensible decisions for the people they advise. Full Transcript: Ken Suzan: Thank you, Rolf. Our returning guest today is Franklin Graves. Franklin is the founder and editor of Creator Economy Law, a website and newsletter that educates creator economy professionals on the intersection of law and policy with the world of creators, brands, and platforms. Franklin also published the first law review article focused on the creator economy, Upload Complete, an introduction to creator economy law. He regularly appears across news and media outlets as a commentator and contributor with a focus on educating creators and raising awareness of all legal aspects of the creator economy. Franklin is based in Nashville, Tennessee. Ken Suzan: Franklin was invited to participate as one of the creators and creator economy professionals in the first ever White House creator economy conference. Franklin works full time as a product counsel at LinkedIn Corporation. As a member of the product and data team, he focuses on emerging issues in AI and data. Franklin previously held roles on the technology law group at HCA Healthcare, the commercial legal team at Eventbrite, and the business and legal affairs team at Naxos Music Group. Welcome back Franklin to the IP Fridays podcast. Franklin Graves: Thank you so much for having me. It is exciting to be back and reflecting over the last decade since I last joined and also the paper that I wrote that dives into this in more detail. So I really appreciate it. And yes, full disclosure, I currently work for LinkedIn, which is a subsidiary of Microsoft. I’m here in my personal capacity to talk about this, the paper I wrote before joining LinkedIn and all of that. So thank you so much for having me back. Ken Suzan: Excellent. So Franklin, since your last appearance on IP Fridays in 2017, your career has evolved significantly. You are now senior product counsel at LinkedIn focusing on AI and data. How has working inside a major tech platform changed your perspective on the legal frameworks governing digital content compared to when you were viewing it purely from the creator side? Franklin Graves: I appreciate that question because when I wrote the article, I did not work for LinkedIn. And I had been coming from a history in my career where I, right out of law school, worked for a record label like we talked about almost 10 years ago. And I was on the content creation side. I’ve represented a major distributor of classical music digitally at the time. And that was my first exposure to understanding how content was taken from the initial inception stage from creators and routed through all the various digital platforms that were at the time still evolving and even arguably still today continue to evolve. The early days of YouTube Music launching and then Apple Music launching, and then going through all the phases of high-res audio and everything that came after that. So that was an interesting perspective to start my career with. And then I went to Eventbrite, which is a ticketing platform, but was also focused on elevating event creators. They kind of took on that moniker of “Hey, we are event creators that we support.” And that was arguably my first exposure to the platform side, the tech platform side of it, because Eventbrite is a platform. And so then I evolved from there in my personal capacity, in a pro bono capacity representing individual creators across the YouTube space. And that’s what we talked about a little bit back when I first came on the podcast. Franklin Graves: Over the last decade, it’s been a chance to grow my own understanding of the creator economy. The terminology “creator economy” came around. And then now on the other side of it, having written the article and all that, and now being fully in-house at LinkedIn, I truly am experiencing a social media platform. LinkedIn is of course arguably way more than just the platform itself. There are so many different avenues to it, but it is a chance for me to understand what it is like working for a company that is operating the platform that people are distributing content on. There’s a user journey to content and all of that. So it’s definitely enhanced and given me a different perspective from a major tech platform side. And part of my role at LinkedIn is really heavily focused on understanding regulation and how that from an AI and data perspective impacts the company. And so I’ve been really leveling up my game over the last year and a half that I’ve been here, understanding mostly EU regulations, but also US regulations that are still in their infancy when it comes to AI. But really when it comes to privacy and data, those are pretty well established across the board. It’s been kind of a combination of what I learned at Eventbrite, because I went to Eventbrite when GDPR was going into effect. And so that was an eyes-wide-open moment of getting in the weeds with negotiating data processing agreements, understanding data transfers and cross-border data transfers and the like. So it’s been kind of an evolution as the laws and regulations have evolved. So has my career, so has my own understanding, so have the platforms’ responses to those laws and regulations. And I’m sure that probably resonates with a lot of your listeners who have also been growing their practice and their understanding as the laws and regulations in this realm have been evolving too. Ken Suzan: Yes, indeed. Now let’s switch gears and talk about AI. You advise on AI and data daily. As platforms integrate generative AI tools into their tech stacks, what are the most critical best practices in-house counsel should be adopting right now to embed responsible AI principles into product development? Franklin Graves: So as an attorney, one of my key roles is to understand the technology. Even representing creators and working for creator platforms, that’s something I’m constantly trying to do: put myself in the shoes of being a creator. And I think I talked about this last time I was on, but I come from a background where I was working for a major label doing marketing, video editing, social media work. And I was creating content. I understood the whole life cycle from the inception point of an idea to execution and then to the final delivery and distribution of that content to an audience within a major music label. And so part of that is the same thing that I think attorneys, especially in-house, should be doing: using the tools that the product and engineering teams are either developing in-house or partnering with third parties to develop, or a combination of the two. Using them, understanding them, using them as a creator would, using them as an end user or a client or customer would. And making sure that if you understand the product and understand the nuances of how it operates, and being a part of the iterations of that internally before it fully ramps, that really gives you a chance to understand: okay, we have a lot of responsible AI principles and standards and protocols that are in existence right now, whether it’s NIST, whether it’s based on the EU AI Act or anything and everything in between. It’s understanding how to apply those and bring those into a product and an engineering environment in a way that is practical and actionable for the people that you’re supporting, the stakeholders you’re supporting. So I think one of the critical best practices is, number one, understand the product or features that you’re supporting. Franklin Graves: And then understand how you as an attorney can use your expertise and understanding of responsible AI practices, whether it’s a regulatory standard or an industry-adopted standard or a hybrid of the two, to leverage those and implement those, break those down and make them into actionable controls and processes and flows that work within your existing infrastructure. That’s a lot of high-level talk, but that’s the general idea. One concrete example we talk about frequently is with open source AI. If you’re working with a product team or an engineering team that is taking an off-the-shelf open source model and bringing that in-house, a lot of times companies have pre-existing open source processes that cover the use of open source software or code. Piggyback on that. That’s the easiest quick win for attorneys: leveraging your existing open source processes to just build on top of that the AI flavor and layering. It’s not very much that you have to do, but the underlying process of the key stakeholders that need to be involved in the review, whether it’s security, whether it’s executive sign-off if it gets to that point, even export control considerations should already be part of your existing open source software process. So layering in on those existing processes the specifics of generative AI or large language models that you’re trying to bring in is a great way to put this into practice. Ken Suzan: Now looking at the geopolitical landscape that we currently have, we have the EU AI Act setting strict standards and shifting US executive orders. How should platforms and brands prepare for this fragmented regulatory environment when deploying AI tools to a global user base? Franklin Graves: It’s a great question. It’s something that is still evolving, I think is fair to say. I would equate it, as I do in the paper that I wrote, to how creators and arguably brands don’t own the platforms that they’re building their communities on. That spawned this concept of de-platforming or going into building your own platform, a decentralized platform of sorts, and owning your community. That gives you that control and takes away the level of instability that can come for creators trying to build a business on a platform they don’t own, they don’t control when certain updates happen, when algorithms change, when tools and functionalities either become available or go away completely. So it’s very similar to what we’ve been experiencing in a regulatory environment where we have geopolitical complexities, for lack of a better term, that can overnight seemingly disrupt the way in which a platform or even a multinational brand is able to connect and reach an audience or continue to leverage the user base that they’ve built. I think TikTok is a great example of that, where it became a national security concern and suddenly it was facing an executive order that required it to be effectively disabled in the US or completely owned and operated by a US entity. All the mechanics and technicalities of whether it’s actually possible and still have a global platform with a global user base is a whole different discussion. But that’s an example of very similar considerations that are now not just a discussion point at the creator level or the individual brand level, but also in a much broader context at a platform level as well. Ken Suzan: Franklin, let’s now shift gears and talk about your article. In your recently published journal article, Upload Complete, which we will have linked in our show notes, you advocate for a shift in terminology from internet creator law, a term used during our first podcast almost a decade ago, to creator economy law. Why is this distinction important and how does it change the way legal practitioners should view the ecosystem of creators, brands, and platforms? Franklin Graves: Oh yes, this is part of the reason why I wanted to write the article: to lay this foundation of understanding. Because at the time I’d written the article, the term creator economy and creator had really not appeared but for maybe once in an actual court decision. And it was kind of focused on influencers and this concept, and it was just not getting it right. And so it was also, as you mentioned, when we first spoke I was even using the term internet creators. And I think that was something that was common at the time. The “internet” portion as a qualifier has since dropped off. And now for purposes of the creator economy, the term creators refers to individuals, it can be small businesses, which is what we’ve seen from a regulatory standpoint, how these small businesses are being impacted by regulations. But essentially creators in the article I pin in the context of intent. What is the intent behind the person or the small business that is posting content, trying to build a community and form a community in a virtual environment? And then that can even spill over into real physical world environments. And so the intent is kind of what I look at. Franklin Graves: And I have a chart in the article that has a diagram showcasing the overlap of what I refer to as “users generating content.” It’s a play on the concept of user-generated content, UGC. Users generating content is that large bucket of anyone posting on a platform of some kind. And within that large bucket, that large circle, are smaller subsets. You have creators, you have brands. Those are really the two buckets you can put people into. Otherwise it’s like your grandmother or your parents posting content on Facebook or Instagram, and those are everyday users of a platform. The distinction to get into that subcategory of being a creator more so has been analyzing the intent behind the posting. Are you posting content to build an audience, to build a community, to eventually have a chance to monetize the following that you’re bringing in or sell services or something like that? Brands are posting for that reason. Creators are maybe posting for that same reason. But even within the creator category, there’s a subcategory of influencers that are trying to sell something, that are trying to build more than just an awareness of who they are, their influence. They are trying to do brand deals, partnership deals, upsells and all that, and start an actual small business aside from just the content itself that they’re creating. So that’s kind of the distinctions that I make in the paper. And that’s why it’s important to understand and lay that foundation, that anyone can post content online, but the intent, the why behind their posting that content, really does ultimately matter, especially when you’re looking at it from a court case or from a regulatory standpoint. Ken Suzan: Now, Franklin, we’re seeing unprecedented geopolitical activity around platform ownership. For example, the US legislation targeting TikTok and Brazil’s recent temporary ban of X. How do these macro-level battles impact the day-to-day livelihood of creators? And how can they legally and operationally protect themselves? Franklin Graves: So the shift that we’re seeing, and I alluded to this earlier in our conversation, is this concept of Web 3. And that term may or may not be really popular anymore, but that’s essentially what we’re looking at: a shift into a federated, decentralized operation of a platform. So instead of one owner, one company, one entity owning and operating the platform, it’s decentralized. Anyone can start up a server, and it’s interoperable, meaning anyone can plug and play and connect to that larger network. And it creates this unified social network experience. Within each operating node of that network, there can be your own decisions around content moderation, your own decisions around the hosting providers you use, where you’re operating out of, the terms and conditions that apply to that. But the flip side is that instead of creators posting and sharing in a closed environment run and controlled by a singular entity, you’re now experiencing a peer-to-peer type operation where your experience can change based on which server, which node, which user you’re engaging with. You might have content that’s acceptable in one area but not acceptable in another, and maybe it just doesn’t even show up in that other area. Franklin Graves: But from a liability standpoint, as creators start to build their own networks and communities, even outside of a concept like the fediverse, it’s even down to creators building their own communities through online courses, subscription membership-based platforms that they run on their own website. There’s open source software out there, even something called Ghost, where you have memberships. And that is a creator or a small business in the creator economy that is now taking on the obligations that would typically fall upon a platform. They need to take into consideration terms and conditions, privacy policies, legal aspects, and regulatory considerations for running a platform, especially in a global world. So it’s a lot of liability that then shifts over to those small businesses and even brands sometimes that are doing the same thing. Whether it is something as simple or complex as content moderation or all the way up to monetizing an audience, this new world where creators can spin up and run a platform all dovetails back to the concept of creators not feeling like they have control in reaching the audience and the community that they’re building on an individual platform. And so this really became more mainstream conversation with TikTok and the issues around it potentially being shut down in the US. That was kind of the mindset shift and eyes opening for many creators, especially within the influencer subset, of realizing: we need to make sure that we have a way to reach the audience we’ve built if the individual platform that we’ve committed to over the last year or three years or so is no longer available. We need a way to continue that relationship outside of that one platform controlling it. Ken Suzan: Franklin, we have a few minutes left and a number of topics. So I’m going to switch gears and talk about a few issues. First, a major emerging topic in your paper is the evolution of protecting kids online. With state-level age-gating laws like the CAADCA and the recent FTC updates to COPPA, how should platforms navigate the significant tension between strict age verification mandates and the privacy and First Amendment rights of their users? Franklin Graves: Man, that is a whole discussion to unravel. It is a consideration that we’re seeing happen again, going back to the geopolitical nature of everything. Countries like Australia and certain countries in Europe and now even individual states in the US are trying to look at ways, and some of them have already put into place minimum age requirements before you can even sign up for an account with a social media platform. One of the things I’d just highlight quickly here is that one of the tensions is around how you verify someone’s age online and still maintain the ability to be at least pseudonymous. How do you still have a level of privacy, autonomy, and protection when it comes to having to provide something like a driver’s license or have parental consent tied and connected to an account managed by a parent in a situation where maybe it’s not appropriate or not beneficial to the child in that manner? But then maybe there are counterbalancing factors that outweigh that. All of that comes down to the technicalities of how it’s actually implemented and maintaining the sense of openness and freedom that we’ve had on the internet to date. And then the other element there is, since a lot of the internet that we think of today is more so through mobile applications, is it something that the mobile operating system providers and app store providers should be thinking about? So whether that’s the Google Play Store or the Apple App Store, where does that initial age verification need to fall? Is it at the platform level? Is it the app store or mobile device management level or something else? Yeah, there’s a lot to discuss there. And a lot of the issues we’re seeing with how the internet is changing in terms of being able to browse a website without disclosing personal information that might not have been required before is largely stemming from a focus on protecting children online. Ken Suzan: It sounds like, Franklin, we could have another episode covering lots of issues connected with that one topic alone. Franklin Graves: I would absolutely agree with that. There’s a lot going on there. And again, it’s different across the world. And so I know you all have a global listener base. And so there’s a lot of nuances to that whole discussion too, that are worth exploring. Ken Suzan: Last question for today’s episode is regarding the right of publicity. With the explosion of AI-generated synthetic media, digital replicas, and voice cloning, the right of publicity is taking center stage. What are the biggest legal risks for brands partnering with influencers right now? And how can creators protect their most valuable asset, their likeness? Franklin Graves: That’s a great question. I think we’re seeing kind of a throwing-spaghetti-against-the-wall-to-see-what-sticks approach right now by a lot of different parties, whether it’s trademark attorneys, whether it’s general entertainment attorneys or whoever. For example, we’ve seen Taylor Swift filing trademarks to protect certain sounds of her voice and phrasing that she uses. It’s a difficult area because in the realm of generative AI with deep fakes and virtual avatars, that is where it gets tricky, because traditional IP laws are just not able to fully cover that spectrum. It’s a piecemeal approach, but even then it doesn’t fully cover it. So for example, I’m based in Tennessee and a couple of years ago we had the Elvis Act that updated our right of publicity law to add voice and to explicitly reference artificial intelligence. And so that’s the kind of effort we’re probably going to continue to see: efforts to develop some framework around protecting what is essentially a privacy right, in a manner that doesn’t restrict generative AI systems from continuing to develop and operate the way they’re operating now, while layering in those protections so that in the US at least a First Amendment right doesn’t necessarily get squashed, and those traditional well-recognized efforts to not overregulate a technology in its early stages are respected. Franklin Graves: And so I think a lot of what we’re seeing is just a need to update laws. The SAG-AFTRA debate and the strikes that happened around maintaining control of your performance and any iterations of that, or building upon that by a media company that might come later, it’s all on the table right now and still being discussed, still being worked out. I think in the short run, a lot of times if it’s in a brand deal, the key question is: if you are using generative AI to enhance in some way the final deliverable for the campaign, who has control over that? Who has final say and sign-off on how that likeness or that digital replica or that person’s voice is represented? And even outside of the brand space, we’ve seen actors like James Earl Jones signing over certain aspects like their voice and allowing it to continue to be used in these manners powered by generative AI as Darth Vader. And I think I saw something that Boy George was even starting up an AI company that allows musicians, the original recording artist, to rerecord new versions of their masters so that they don’t miss out on that revenue. It’s powered by generative AI, by taking their voice now, which is significantly different than it was back in the 80s, and using generative AI to make it sound closer to the original, but all based on their current performance. So I think it’s still an evolving area. And what’s interesting too is on the platform side, we’re seeing the early stages of platforms like Google starting to acknowledge and rely on the license grant contained in their terms of service for YouTube, which grants them broad rights to use the content to run their platform. So all that to be said, it’s still early stages. I’m very interested to see where we go from here in the future, especially from a global perspective as well. Ken Suzan: Franklin, I could spend hours talking to you about this. You’re such a knowledgeable person on these topics. Maybe in a few years, will we connect again and talk further on AI and all the things that are yet to be developed? Franklin Graves: Thank you. Yeah, it doesn’t have to be another decade. Maybe we can cut it to half a decade, given the pace at which technology is going now. Ken Suzan: Sounds good, Franklin. Thanks again for being on the IP Fridays podcast.

  KI und Compliance: Warum dein DSGVO-konformes Tool noch lange nicht ausreicht   Viele Unternehmen glauben, mit einem DSGVO-konformen KI-Tool auf der sicheren Seite zu sein… das ist ein Irrtum. In dieser Episode spricht Tanja mit Carsten Wittmann, TÜV-Rheinland-zertifiziertem AI-Consultant und Experte für AI, Compliance und Data Privacy am Koerting Institute, darüber, warum Compliance kein Verhinderer, sondern ein Enabler ist.   Tanja Förster auf LinkedIn: LinkedIn - https://www.linkedin.com/in/foerster-tanja/ Carsten Wittmann auf LinkedIn: LinkedIn - https://www.linkedin.com/in/tumaki/     Die größte Compliance-Lüge: "Unser Tool ist DSGVO-konform, also sind wir es auch" Ein Tool kann DSGVO-konform sein und trotzdem völlig falsch eingesetzt werden. Wer die falsche Rechtsgrundlage wählt oder Datenlöschungen nicht nachweisen kann, handelt non-compliant, egal wie seriös der Anbieter ist. Compliance ist kein einmaliger Status, sondern ein laufender Prozess. Shadow-KI: Was passiert, wenn Verbote ohne Alternativen kommen Wer KI verbietet, ohne eine offizielle Lösung anzubieten, treibt die Nutzung in den Untergrund. Carstens Empfehlung: Frag deine Mitarbeitenden, welche Tools sie heute schon nutzen — nicht um sie zu bestrafen, sondern um daraus eine compliant geprüfte Lösung zu entwickeln. Was der EU-AI-Act wirklich von dir verlangt Für die meisten Unternehmensanwendungen gilt vor allem Transparenzpflicht: KI-generierte Inhalte kennzeichnen, Chatbots als solche erkennbar machen. Dazu kommt Artikel 4 — KI-Kompetenz ist Pflicht, aber rollenspezifisch: nicht jeder braucht die gleiche Schulung. Fazit Die meisten KI-Projekte scheitern nicht an der Technologie, sondern daran, dass Compliance und Kultur zu spät mitgedacht werden. Der erste Schritt ist einfach: eine ehrliche Bestandsaufnahme, welche KI-Tools im Unternehmen heute schon im Einsatz sind.     Noch mehr von den Koertings ... Das KI-Café ... jede Woche Mittwoch (>350 Teilnehmer) von 08:30 bis 10:00 Uhr ... online via Zoom .. kostenlos und nicht umsonstJede Woche Mittwoch um 08:30 Uhr öffnet das KI-Café seine Online-Pforten ... wir lösen KI-Anwendungsfälle live auf der Bühne ... moderieren Expertenpanel zu speziellen Themen (bspw. KI im Recruiting ... KI in der Qualitätssicherung ... KI im Projektmanagement ... und vieles mehr) ... ordnen die neuen Entwicklungen in der KI-Welt ein und geben einen Ausblick ... und laden Experten ein für spezielle Themen ... und gehen auch mal in die Tiefe und durchdringen bestimmte Bereiche ganz konkret ... alles für dein Weiterkommen. Melde dich kostenfrei an ... www.koerting-institute.com/ki-cafe/   Mit jedem Prompt ein WOW! ... für Selbstständige und Unternehmer Ein klarer Leitfaden für Unternehmer, Selbstständige und Entscheider, die Künstliche Intelligenz nicht nur verstehen, sondern wirksam einsetzen wollen. Dieses Buch zeigt dir, wie du relevante KI-Anwendungsfälle erkennst und die KI als echten Sparringspartner nutzt, um diese Realität werden zu lassen. Praxisnah, mit echten Beispielen und vollständig umsetzungsorientiert. Das Buch ist ein Geschenk, nur Versandkosten von 9,95 € fallen an. Perfekt für Anfänger und Fortgeschrittene, die mit KI ihr Potenzial ausschöpfen möchten. Das Buch in deinen Briefkasten ... https://koerting-institute.com/shop/buch-mit-jedem-prompt-ein-wow/   Die KI-Lounge ... unsere Community für den Einstieg in die KI (>2800 Mitglieder) Die KI-Lounge ist eine Community für alle, die mehr über generative KI erfahren und anwenden möchten. Mitglieder erhalten exklusive monatliche KI-Updates, Experten-Interviews, Vorträge des KI-Speaker-Slams, KI-Café-Aufzeichnungen und einen 3-stündigen ChatGPT-Kurs. Tausche dich mit über 2800 KI-Enthusiasten aus, stelle Fragen und starte durch. Initiiert von Torsten & Birgit Koerting, bietet die KI-Lounge Orientierung und Inspiration für den Einstieg in die KI-Revolution. Hier findet der Austausch statt ... www.koerting-institute.com/ki-lounge/   Starte mit uns in die 1:1 Zusammenarbeit Wenn du direkt mit uns arbeiten und KI in deinem Business integrieren möchtest, buche dir einen Termin für ein persönliches Gespräch. Gemeinsam finden wir Antworten auf deine Fragen und finden heraus, wie wir dich unterstützen können. Klicke hier, um einen Termin zu buchen und deine Fragen zu klären. Buche dir jetzt deinen Termin mit uns ... www.koerting-institute.com/termin/   Weitere Impulse im Netflix Stil ... Wenn du auf der Suche nach weiteren spannenden Impulsen für deine Selbstständigkeit bist, dann gehe jetzt auf unsere Impulseseite und lass die zahlreichen spannenden Impulse auf dich wirken. Inspiration pur ... www.koerting-institute.com/impulse/   Die Koertings auf die Ohren ... Wenn dir diese Podcastfolge gefallen hat, dann höre dir jetzt noch weitere informative und spannende Folgen an ... über 500 Folgen findest du hier ... www.koerting-institute.com/podcast/   Wir freuen uns darauf, dich auf deinem Weg zu begleiten!

2026 entscheidet nicht das beste KI-Modell über Gewinner und Verlierer – sondern wer seine Daten und Inhalte wirklich gut aufbereitet hat. Im Tim-&-Jonas-Stammformat diskutieren Jonas Rashedi und Dr. Tim Wiegels, warum über 95 % der GenAI-Piloten keinen messbaren Mehrwert liefern, was hinter dem Hype um den Agent-Ready Data Layer steckt und welche Pflichten der EU AI Act ab dem 2. August bringt. Eine Folge über Datenstrategie, GenAI und die Frage, warum die größten Datensilos in den Köpfen sitzen. Key Takeaways: → Warum über 95 % der GenAI-Piloten scheitern – und es fast nie am Modell liegt → Das Gewinner-Verlierer-Framework: Daten- und Content-Verfügbarkeit gegen Prozess-Integration → Agent-Ready Data Layer: warum oft keine neue Plattform, sondern eine schlanke Verknüpfungsschicht reicht → EU AI Act ab 2. August: Governance und Erklärbarkeit als Eintrittskarte statt Compliance-Ballast

Du willst dein eigenes KI-Leitbild erstellen? Hier findest du die Schritt-für-Schritt-Anleitung.-----#Conscious Intelligence ist die Solo-Reihe von Vision trifft Business. Hier denke ich laut: über Business, Kommunikation und Verantwortung, über KI, Marketing und Führung, immer mit dem Blickwinkel auf Bewusstsein und Ethik.In einem Workshop fiel der Satz: „Eine KI-Deklaration? Das ist doch vermutlich nur ein Marketingtrick." Anlass war die öffentliche "No-AI"-Haltung von Dove, die als echte Überzeugung oder gezielte Positionierung gelesen werden kann. Genau diese Spannung ist es, die mir in Gesprächen über KI-Transparenz immer wieder begegnet: die Skepsis, dass Offenheit über den eigenen KI-Einsatz vor allem strategisches Kalkül ist.Über den EU AI Act wird ab 2. August 2026 die Kennzeichnungspflicht für KI-generierte Inhalte im geschäftlichen Kontext eingeführt. Das zeigt: Die Richtung ist klar. Doch zwischen gesetzlicher Pflicht und gelebter Transparenz liegt ein entscheidender Unterschied, und genau dieser ist das Thema dieser Folge.In dieser Folge erfährst du:Warum der Unterschied zwischen KI-unterstützt und KI-generiert entscheidend istWas es mit der freiwilligen KI-Deklaration auf sich hat und warum sie weit über Compliance hinausgehtWie die LEAD-Methodik den Weg vom eigenen Werteverständnis zur glaubwürdigen Aussenkommunikation strukturiertWarum viele Menschen bereits eine klare innere Haltung zu KI haben und sie nur noch nicht sichtbar gemacht habenWas eine KI-Charta bzw. ein KI-Leitbild konkret enthalten und wie du damit anfängstWenn dich das Thema beschäftigt, lohnt es sich, diese Folge in Ruhe zu hören. Denn die meisten Antworten auf die KI-Fragen, die viele umtreiben, haben wir bereits in uns.-----Mehr über mich und meine Arbeit findest du hier:

From Responsible Business to Responsible AI: what you need to know and do. In this episode of Frankly Speaking, Richard Howitt is joined by two leading voices on AI and responsibility to cut through the hype and the fear around artificial intelligence and its environmental impact. From staggering data centre energy figures to algorithmic bias in HR tools, this conversation is essential listening for anyone in a sustainability or responsible business role who is trying to make sense of AI, and what their company should actually be doing about it.Kathryn Dovey is founder of Recalibrate, working on responsible AI issues. She spent ten years at the OECD Centre for Responsible Business Conduct, including the period in which the OECD AI Principles were agreed.Alex de Vries-Gao is a data scientist and researcher at the Institute for Environmental Studies at VU Amsterdam, where he has undertaken extensive research on the environmental impact of AI.You will also hear about:How major US tech companies successfully lobbied the European Commission to keep environmental data from individual data centres out of the public domainThe significant positive use cases for AI in sustainability from satellite data analysis tracking climate change and biodiversity, to breakthroughs in healthcareWhy the Global South risks bearing the environmental costs of data centres while the benefits flow predominantly to the Global North, and how this echoes patterns seen with large infrastructure projects throughout historyThe key frameworks companies need to understand: the OECD AI Principles, the OECD Due Diligence Guidance, and the EU AI Act, and why the EU AI Act is a practical benchmark to start working towards now, even as it undergoes its simplification processListen in and follow us on LinkedIn and Youtube!

Recorded live at EIC 2026 in Berlin, Jeff and Jim sit down with Martin Sandren, IAM Product Lead at IKEA, for a wide-ranging conversation covering nearly every corner of modern identity security. Martin shares what has changed since his first IDAC appearance on episode 293, including the rise of AI, growing interest in digital sovereignty, and the maturing shared signals framework. The conversation moves through risk-based defense in depth, tiered MFA rollout strategies, session management, and the real challenge of trusting AI to make security decisions. Martin introduces identity dark matter and explains how IVIP can surface the 95-plus percent of applications that never reach an IGA system. The episode also covers shadow AI, MCP server risks, the SaaSpocalypse debate, and the EU AI Act. It closes on a grounded note: solar panels.Connect with Martin: https://www.linkedin.com/in/martinsandren/Connect with us on LinkedIn:Jim McDonald: https://www.linkedin.com/in/jimmcdonaldpmp/Jeff Steadman: https://www.linkedin.com/in/jeffsteadman/Visit the show on the web at http://idacpodcast.comTIMESTAMPS00:00 Welcome and EIC 2026 intro01:47 What has changed in two years: AI, sovereignty, shared signals03:06 Martin's EIC presentations: AI for IAM and IAM for AI04:46 Can you prioritize one direction over the other?07:13 What would it take to trust AI making identity decisions?09:32 AI-enhanced detection and risk-based session management13:07 Session invalidation and the shared signals framework14:11 Defense in depth and right-sizing privileges18:25 MFA today: any MFA versus phish-resistant MFA19:17 AI chatbots, enterprise LLMs, and shadow AI23:11 MCP servers, NHI risk, and return on risk thinking27:00 AI configuring IAM systems: how close are we?31:30 LLM costs, the SaaSpocalypse, and enterprise AI futures40:10 Identity dark matter and the IVIP concept44:16 CMDB versus IVIP: do you need both?46:18 The EU AI Act and building an AI governance registry49:18 Where to start: get your AI inventory in place first50:00 Closing thoughts and the solar panel tangentKEYWORDSAI for IAM, IAM for AI, identity dark matter, IVIP, IGA, shared signals framework, phish-resistant MFA, defense in depth, session management, MCP servers, NHI, shadow AI, SaaSpocalypse, EU AI Act, AI governance, zero standing privilege, EIC 2026, IKEA, IDAC, Identity at the Center, Jeff Steadman, Jim McDonald, Martin Sandren

today we examine the 2026 landscape of artificial intelligence, specifically comparing proprietary and open-source models regarding privacy, cost, and legal compliance. Organizations must choose between proprietary APIs, hosted open-source solutions, and self-hosting to balance performance with data sovereignty requirements like HIPAA or the EU AI Act. While proprietary models currently lead in complex reasoning, open-source weights offer significant long-term cost savings and transparency for high-volume users. However, true total cost of ownership includes hidden expenses such as specialized talent, hardware infrastructure, and continuous model maintenance. Legal frameworks like the EU AI Act introduce strict obligations for high-risk systems, making explainability and governance essential for enterprise deployment. Ultimately, the transition from experimental pilots to industrialized AI factories requires mastering token economics and navigating the evolving regulatory environment.

Regular a inteligência artificial: até onde, por quem e a que custo? Igor Alcantara recebe a advogada Raquel Maciel para uma conversa sobre riscos, responsabilidade e os limites entre proteger as pessoas e travar a inovação. Do Dilema do Bonde às Leis de Asimov, do EU AI Act ao futuro do trabalho, um papo sobre quem deve pagar a conta quando a máquina erra.A Pauta foi escrita por Igor Alcantara e Raquel Maciel. A edição foi feita por Leo Oliveira e a vitrine do episódio feita por Igor Alcantara em colaboração com a Inteligência Artificial Claude Design da empresa Anthropic. A coordenação de redação e de redes sociais é de Tatiane do Vale. A gerência financeira é de Kézia Nogueira. As vinhetas de todos os episódios foram compostas por Rafael Chino e Leo Oliveira.

Send us Fan MailShow Notes:1:35 Patrick McGranaghan's background 2:45 McGranaghan's work with Pierre Valentin3:05 focus on collision of culture and infrastructure4:45 “evidential fog” around AI in the arts6:00 abstract nature of these AI issues 7:00 his writing on these issues to navigate these issues8:30 EU's AI framework “recognizes the structural nature of the problem” – can't be minor updates to old copyright debates, “AI creates problems of scale, opacity and jurisdictional arbitrage that traditional legal categories do not solve very elegantly.” 10:00 incentive for jurisdiction shopping11:40 Getty v. Stability AI in the UK 14:05 EU AI Act's extraterritorial obligations 15:00 EU AI Act, Article 53: general purpose models brought into EU must comply with EU copyright law, including opt out reservations; and detailed summary of training data17:55 UK's approach is more exposed to loopholes19:25 opt in versus opt out systems21:35 Kadrey v. Meta 22:55 the burden placed on creators by the opt out system 25:45 sporadic licensing deals and unclear remuneration standard27:30 interoperability 28:40 impact of robots.txt31:15 Alan Robertshaw re: impact of AI on the practice of law34:50 AI defamation cases36:20 McGranaghan - need for lawyers regardless of AI37:25 Robertshaw - legal professions' varied approaches to AI38:55 AI and astronomy40:30 moral conflict with not compensating artists43:00 justices/injustices related to AI46:45 market harm created by AI49:25 definition of justice 53:05 protections that artists can use, e.g., robots.txt, metadata, units based protection, Glaze and Nightshade 58:00 mark Patrick hopes to make around AI and art Please share your comments and/or questions at stephanie@warfareofartandlaw.comMusic by Toulme.To hear more episodes, please visit Warfare of Art and Law podcast's website.To leave questions or comments about this or other episodes of the podcast and/or for information about joining the 2ND Saturday discussion on art, culture and justice, please message me at stephanie@warfareofartandlaw.com. Thanks so much for listening!This podcast and its content may not be used for training or developing AI systems without permission.© Stephanie Drawdy [2026]

In der neuen Episode 196 der Turtlezone Tiny Talks, diesmal wieder in Zusammenarbeit mit dem KI Expertenforum, geht um die Transparenzpflichten des Artikel 50 vom EU AI Act. Regeln, die Anfang August in Kraft treten. Das Gesetz wird oft im Kontext mit Deepfakes gelesen und wenn man den Begriff Deepfake hört, hat man sofort Bilder vor Augen: Ein Politiker sagt etwas, das er nie gesagt hat. Ein CEO kündigt einen Börsenschritt an, den es nie gab. Eine Prominente wird in kompromittierende Szenen montiert.  In solchen Fällen haben wir bislang oft eine Diskrepanz zwischen dem gesellschaftlichen Konsens und der rechtlichen Bewertung und Verfolgbarkeit gehabt. Das hat der europäische Gesetzgeber natürlich zu Recht auch vor Augen gehabt. Aber das ist nicht der Kern und Knackpunkt bei den neuen Regelungen des Artikel 50.  Die eigentliche Debatte beginnt bei den tausend alltäglichen KI-unterstützen Anwendungen, die heute schon in Smartphones, Bildbearbeitung, Podcasts, Videos und Social Media stecken. Es geht um die Zukunft von Authentizität und um die Frage, was bedeutet überhaupt noch "echt", wenn KI immer stärker Bestandteil jeder Kommunikation wird? Genügt Transparenz allein, um Vertrauen in Content und Informationen zu erhalten? In Zukunft fragen wir wohl verstärkt nicht nur, on ein Inhalt durch die KI generiert ist, sondern: Ist nachvollziehbar, wie dieser Inhalt entstanden ist? Und entsteht daraus eine relevante Täuschung? 37 spannende Podcast-Minuten.Ergänzende Informationen:Die neue Transparenzpflicht für KI-Inhalte (KI Expertenforum)Entwurf der Leitlinien zur Umsetzung der Transparenzpflichten für bestimmte KI-Systeme gemäß Artikel 50

In this episode, Katherine Forrest and Scott Caravello phone a friend across the pond to discuss the latest developments under the EU AI Act. John Patten, head of the UK and European Intellectual Property & Technology practice for Paul, Weiss, joins the conversation to unpack the digital omnibus package, revised high-risk AI timelines, transparency obligations, and draft guidance on AI system classification. For the sources referenced in this episode, please see the links below: OJEU: The AI Act Explorer   ## Learn More About Paul, Weiss's Artificial Intelligence practice: https://www.paulweiss.com/industries/artificial-intelligence

AI is moving from “helpful assistant” to autonomous actor, and payments leaders are about to feel the difference. I sit down with Russell Moore, Co-Founder and CEO of Amotivv, to get concrete about what breaks when generative AI and agentic AI leave the lab and touch regulated data, customer outcomes, and real money movement.We talk through why so many AI initiatives stall after a promising proof of concept: not because the model is useless, but because teams cannot control the context, prove what happened, or satisfy audit and compliance requirements at scale. Russell explains Amotivv's three-layer view: persistent AI memory you own, a governed workspace for using any model, and a verification layer (including cryptography and append-only records) that produces tamper-resistant, independently verifiable proof of what AI did, which tools it used, and what policies allowed it.We also dig into practical realities that every fintech team runs into fast: model selection and token costs, why caching and routing matter, and how platform lock-in sneaks in when your vendor effectively owns the memory. On the policy side, we discuss the pace of AI regulation, why the EU AI Act is a useful north star for building “bomb-proof” guardrails, and what it means to be able to prove both usage and non-usage of AI as expectations tighten.If you're building AI for fraud, marketing, customer support, underwriting, or agentic commerce, this is a roadmap for making it trustworthy.

We are revisiting the AI-copyright interplay for the first time in nearly three years. Copyright remains very relevant to our sphere of interest, not least because the EU AI Act specifically points at EU copyright law with regards to training data and transparency requirements for AI models.Malcolm Bain is an English solicitor and Spanish abogado. He has worked as an Information Technology and Intellectual Property lawyer over the last 20 years, with a specialisation in technology licensing, open source software and content, technology transfer and privacy. In 2006, together with his partner Manuel Martínez, he founded his own firm “id-law partners” as a boutique specialized in IP and ICT. In May 2018, both incorporated this firm into Across Legal.In addition to his professional activity advising entrepreneurs, private companies, public administrations and open source projects, Malcolm is a member of the Free Software Foundation Europe and ASTP, associate professor of law at the University of Barcelona, ​​mentor in Tecniospring Industry and other programs for entrepreneurs and frequent speaker at conferences and seminars in the field of ICTs and entrepreneurship in the digital world.References:* Malcolm Bain at Across Legal* Malcolm Bain on LinkedIn* Monkey selfie copyright dispute (Wikipedia)* Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019 on copyright and related rights in the Digital Single Market and amending Directives 96/9/EC and 2001/29/EC* Report on Copyright and Artificial Intelligence (UK Intellectual Property Office)* Stability AI largely wins UK court battle against Getty Images over copyright and trademark (AP News, November 2025)* US Copyright Office: Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence (2023)* German Court Rules OpenAI Infringed Song Lyrics in Europe's First Major AI Music Ruling (November 2025)* Jakob Plesner: Copyright Exceptions for Generative AI (Masters of Privacy, October 2023).* (NOTE: The second part of this conversation was recorded in Spanish and is available in our separate Masters of Privacy ES channel.) This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.mastersofprivacy.com/subscribe

today we examine the legal, economic, and ethical landscapes of artificial intelligence as it integrates into global society. They highlight active regulatory efforts like the EU AI Act and the U.S. Algorithmic Accountability Act, alongside international agreements focused on frontier AI safety and corporate responsibility. Economic analysis from the collection indicates that AI is already reshaping the labor market, specifically impacting white-collar sectors and shifting the risks for high-wage occupations. Expert reports clarify that U.S. tort law and liability frameworks will increasingly govern AI-related harms, even as debates persist regarding the security trade-offs between open-source and closed-source models. Furthermore, the documents emphasize the necessity of protecting consumer privacy and implementing inclusive engagement practices to prevent systemic bias. Collectively, these materials provide a comprehensive overview of how governments and industries are attempting to balance rapid innovation with public safety and accountability

98% of patients welcome AI in their care — and still want a human in charge. That tension ran through the OECD and Spanish Ministry of Health conference on scaling AI in health (Madrid, late May 2026), and it frames this episode of Faces of Digital Health. Out of 38 OECD countries, only seven have a formal AI strategy and just over a tenth run workforce upskilling programmes — the ambition is outrunning the institutions meant to govern it. Host Tjaša Zajc brings together voices from across the conference to ask what actually has to change: regulation, trust, who gets a seat at the table, and the parts of the agenda nobody is funding. Featuring: - Eric Sutherland — Senior Economist, OECD - Aferdita Bytyqi — Executive Director & Founding Partner, Digital Transformations for Health Lab (DTH-Lab) - Erza Selmani — Research Fellow, DTH-Lab - Valentina Strammiello — Executive Director, European Patients Forum (EPF) - Dr Ricardo Baptista Leite — CEO, HealthAI (the Global Agency for Responsible AI in Health) - Dr Persephone Doupi — Senior Medical Officer, Finnish Institute for Health and Welfare; President, European Federation for Medical Informatics (EFMI) What the conversation covers: - Why trust — not capability — is the binding constraint on health AI adoption - The OECD readiness gap: AI strategies, HTA frameworks and workforce upskilling - How patients really feel about AI: consent forms, transparency, and keeping clinicians central - Why youth health and wellbeing keep getting left out of AI governance frameworks - Five recommendations to make the EU AI Act work for health and competitiveness - Coordinating the EU AI Act, MDR/IVDR and the European Health Data Space - Health technology assessment and reimbursement as the real barriers to scale - AI literacy and prevention: the most underweighted lever in the room Chapters: 0:10 — Welcome: AI in Health & the 2026 OECD Conference in Madrid 0:25 — Key Stats: Only 7 of 38 OECD Countries Have a Formal AI Strategy 2:10 — Eric Sutherland (OECD): We're Not Using Data as Effectively as We Could 3:11 — Afrodita & Erza (DTH Lab): Youth Health Is Missing from AI Governance Frameworks 5:12 — Valentina Stramello (EPF): 98% of Patients Are Positive About AI, But Trust Requires Transparency 7:14 — Dr. Ricardo Baptista Leite (Health AI): 5 Recommendations to Fix EU AI Policy for Health 10:53 — Persephone Doupi (EFMI): We Must Prioritize AI Literacy and Shift Healthcare Toward Prevention —

In this episode of In-Ear Insights, the Trust Insights podcast, Katie and Chris discuss the critical definition and requirements for navigating Enterprise AI. You’ll learn how to distinguish between consumer-grade tools and the strict standards required in regulated industries. You’ll discover the twenty essential pillars for building a secure and compliant AI strategy for your organization. You’ll understand why rigorous vendor scrutiny matters as much for software as it does for human talent. You’ll gain clarity on the governance frameworks necessary to prevent data leaks and legal vulnerabilities in your enterprise. 00:00 – Introduction 03:15 – Defining Enterprise AI vs. SMB AI 07:45 – The role of Microsoft Copilot in regulated environments 12:20 – The 20 components of Enterprise AI readiness 18:10 – Challenges in organizational adoption and change management 22:30 – Security and data privacy as the foundation 27:00 – Call to action Watch this episode to master the complex landscape of regulated AI and safeguard your company’s future. Watch the video here: Can’t see anything? Watch it on YouTube here. Listen to the audio here: https://traffic.libsyn.com/inearinsights/tipodcast-enterprise-ai-101.mp3 Download the MP3 audio here. Need help with your company’s data and analytics? Let us know! Join our free Slack group for marketers interested in analytics! [podcastsponsor] Machine-Generated Transcript What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode. Christopher S. Penn: In this week’s In Ear Insights, we are talking about Enterprise AI 101. I am in the midst of a series in the Trust Insights newsletter, which you can get at TrustInsights.ai/newsletter. Part one was last week on seven different aspects of enterprise AI. But Katie, you said it would probably be helpful to level set what enterprise AI is and how it differs from SMB AI, mid-market AI, consumer AI, and so on. Katie Robbert: It is interesting because I feel like every time we jump on to record a podcast, there is a whole new set of vocabulary that I need to get caught up with. We need to make sure that everyone else knows what we are talking about because there is nothing worse than listening to a podcast or reading an article and having no idea what the author is talking about because they are introducing a concept but not really explaining it. I wanted to take this episode to talk about what enterprise AI is. Since you and I have not defined it, I am going to take my best guess at what enterprise AI is using some logic and deduction. I could be wrong, and that is why I think it is worth covering. From my perspective, if I had to put a definition to it, I am assuming enterprise AI is the type of AI implementation that occurs at an enterprise-size company. That sounds overly simplistic, but the bigger the organization, the more red tape, the more politics, the more departments, the more stakeholders, and the more governance there is. There are a lot more complications versus a small business like we are, where we can just decide one day, “Hey, I am going to start using this tool.” There are no real hurdles to go through. Then you have those mid-sized companies where you start to introduce some of those hurdles. You might need to work with your IT team to make sure that everything is in compliance. You might need to make sure that you have a place to host these new pieces of software, and that is not something that the marketing team is necessarily responsible for. Then you get to the enterprise-size companies where everything is completely siloed. Even in the best enterprise-sized companies, you are going to run into these silos. Because no one person is responsible for everything, you typically have multiple CEOs. Depending on what part of the country you are in, you might have a board for every different division of the company. If you are a Procter & Gamble and you have hundreds of product lines underneath, each of those is their own individual business. Each of those businesses are not necessarily talking to each other or sharing resources. That is my logical guess at what enterprise AI is. Christopher S. Penn: That is what I started with until I started doing the research into it. I realized that is not what it is. The generally accepted definition is AI within any commercially regulated entity. I realized as I was going through the research that commercially regulated means you have external regulation imposed on the company. It might be a 50-person company, but if they work in HIPAA or FINRA, they have to behave in highly regulated ways. Whether you are publicly traded or, for example, colleges that have to adhere to FFIEC rules and FERPA rules, enterprise AI is about operating AI—whether classical or generative—in a commercially regulated environment where you have externally mandated requirements that you must meet. Your definition for small business stuff makes total sense in that environment because Trust Insights is not a regulated company. However, when we work with our healthcare clients, we have to behave as though we are an enterprise company because we have to conform to their requirements. Katie Robbert: I am glad we are talking about this because the terminology is confusing; when you think of an enterprise company, you are not thinking of a commercially regulated company. I have to wonder why it is not called commercially regulated AI versus non-commercially regulated AI. It is a mouthful and a little bit harder to remember, but it is more descriptive and more accurate. I think like me, a lot of people are going to get confused about what enterprise AI actually is. Christopher S. Penn: A lot of this is because our background is in marketing, so we use the term enterprise to just mean a big company. If we want to market to enterprise companies, we are not marketing to a 50-person firm; we are marketing to a 50,000-person firm. In a lot of CRM software, the dividing line is typically 10,000 employees or 100 million in revenue. This is especially relevant because you see a lot of AI companies like Anthropic and OpenAI in a fight with Microsoft to try and gain a foothold into those enterprises. Microsoft, with their Copilot offering, has dominance by the very fact that their legacy Office 365 stuff is approved in those regulated environments. Katie Robbert: It is ironic because we spent so much time admittedly dismissing Microsoft’s Copilot as the less than version of generative AI, and now Microsoft is getting the last laugh on everyone. They are saying, “You have to use me because I have already been approved by IT and governance, and good luck.” You are stuck with whatever I decide to give you. If I were Microsoft, I would be petty and say, “You guys spent way too much time dismissing me and calling me inferior, so too bad.” Christopher S. Penn: A lot of that, as we have talked about many times on stage, is that the reason Copilot has fewer capabilities than other systems is specifically because of the regulated environment. It is trivial for Google to foist something on consumers and say, “Now we are going to read all your Gmail.” That does not fly in a regulated industry. Katie Robbert: That understanding is really helpful to the people who are saddled with Microsoft Copilot because we hear complaints about why they cannot use other shiny objects. If you are in a 50,000-person company and you weren’t there when the regulatory standards were decided upon, you are sitting there wondering why you cannot use Gemini to generate ad headlines. Then you do it on the side and get in trouble because there is no clear documentation saying why you have to use Copilot and nothing else. What we are hearing is that employees in companies required to use Microsoft Copilot are using other models on the side. That information is still getting filtered into the organization, and it is a huge governance problem. Christopher S. Penn: Completely. In enterprise AI, there are 20 different components to being ready. I derived this from the US federal government's NIST AI regulations and the EU AI Act, which is the gold standard. Katie Robbert: I want to see if you can get all 20. Christopher S. Penn: One, Strategy and Operating Model; two, Governance Policy and the AI Council; three, Legal, Regulatory, and Compliance. Katie Robbert: Are you reading this off a screen? Christopher S. Penn: I am 100% reading this off the Trust Insights Enterprise AI Landscape Field Handbook. Katie Robbert: Fine, continue. Christopher S. Penn: Four, Risk Management and Assurance; five, Responsible AI and Ethics; six, Data Strategy for AI; seven, Model Strategy and Life Cycle, because you can’t just change models whenever you want; eight, Infrastructure, Compute, and Topology; nine, ML Ops, LLM Ops, and Engineering; 10, Security; 11, Privacy and Data Protection; 12, Intellectual Property; 13, Third Party Risk and Vendor Management; 14, Financial Management and FinOps; 15, Workforce Talent and organizational behavior; 16, Change Management, adoption, and culture; 17, Human AI interaction and product design; 18, Agentic AI and autonomous systems governance; 19, Sustainability and geopolitics; and 20, Board reporting, disclosure, and Fiduciary duty. Katie Robbert: I just heard a whole lot of new job opportunities listed. So, if someone were working in a regulated industry like pharma, these are the 20 things they would need to be aware of before evaluating generative AI. It is interesting that organizational behavior and change management are part of it. You would think the regulations would be more technical versus human, but I am surprised that is part of it. Christopher S. Penn: It makes sense because in order for any AI to succeed in an enterprise with 50,000 or 300,000 employees, you have to prioritize change management. Organizational behavior cannot be an add-on; they have to be baked into what you do from the beginning, otherwise your initiative is going nowhere. Katie Robbert: I don’t disagree, but the typical way that works in a large organization is top-down. They make a decision, and you walk in the next day to find it has automatically updated your computer settings. Now you can no longer use a web browser search; you have to use Microsoft Copilot. That is their version of change management, but it is really just a dictatorship from above. I am interested in future episodes to explore what that should look like in a regulatory environment. Christopher S. Penn: We have known for two years that adoption is the hardest part. Deployment is easy compared to adoption. You can put Copilot on someone's desk, but they may not use it even if you tell them they have to. It comes back to how you get them to see the benefits. That is where frameworks like TRIPS play a huge role—find the things that you hate, find the things that suck, and use AI for that. Get that one thing off your plate. Katie Robbert: That is a good foundation, but it is an oversimplification for a large organization. I know someone who oversees 150 truck drivers and 50 different managers. The layers are so deep. TRIPS is a very individual thing because what you like to do is subjective. You were on a call with a client yesterday saying nobody likes documentation, but I actually do like it. My scoring would look different than yours. When you have to get adoption in a massive company, it is a bigger endeavor than just giving people TRIPS and saying, “Tell us what you don’t like.” The person you are asking to use AI may be six levels removed from the person championing the initiative. Christopher S. Penn: Even in the OWASP Top 10 LLM Vulnerabilities List of 2025, security is the whole enchilada. Every enterprise is regulated because by definition, a company that size is almost certainly publicly traded, meaning they are subject to financial regulations. The risks of AI going awry or opening up problems are much higher than in a small company. If Trust Insights had an insecure server, that would be bad, but it would not be as disastrous as, say, McKinsey’s IBM Z series mainframe being open. Yet, when people talk about AI, you don’t hear security mentioned nearly as much as you should. Katie Robbert: It is true. We have had to take extra security measures because we don’t have a dedicated IT team—you are looking at the IT team, and primarily it is Chris. We don’t have any wiggle room to set things up haphazardly. We have to do it right from the start. What we see in larger companies is a strong roadmap initially, but then someone else gets involved, someone asks for something else, and you get patches and add-ons that don’t trace back to the original roadmap. By the end, you are wondering what the original goal was. The bigger the organization gets, the harder it is to maintain control. It becomes a snowball effect. Christopher S. Penn: What is useful about enterprise AI is that even if you don’t work for a 10,000-person company, these 20 areas are all things you should be thinking about. Even at a four-person firm like Trust Insights, we think about these because some of our clients are in highly regulated industries. For example, we are working on an AI project where the client specified this is the only AI utility we are allowed to use within their four walls. Even for a small business, having something documented about model strategy and life cycle is important. As of the day we are recording this, Google Gemini 3.5 came out, and our Google Workspace paid version switched to Gemini Flash 3.5. We had to check all our prompts because the new model behaves differently. Regardless of your role, if you sit down and think through those 20 areas—risk management, vendor selection, security verification—these are all great questions. Katie Robbert: There is a good starting place for this. You can find our downloads at TrustInsights.ai/StrategicToolkit. There is also a free version at TrustInsights.ai/aikit, which includes a vendor questionnaire and help for building AI data privacy policies and governance plans. We have already templated these things out. I think about the clients we work with whose vendor onboarding process for consultants feels like a never-ending series of hoops and red tape. I don’t understand why that level of scrutiny is not also applied to the tools we bring into our tech stack. We are renting space in those tools and freely giving them our data. Those companies now have our data and will use it for their own benefit. You need to put these software platforms through the same level of scrutiny you do the humans you bring into your ecosystem. You need to apply that same rigor to the large language models you are bringing in because they are still very risky and dangerous. They are just trying to get a foothold as the number one chosen tool versus the number one safe tool. Christopher S. Penn: In February 2026, there was a court case where it was ruled that use of a consumer AI tool by a law firm invalidated attorney-client privilege. The judge ruled that this is no longer privileged information. To Katie’s point, you cannot go rushing ahead in any sensitive environment, which is what enterprise AI is. You have to be doing your homework. If you have thoughts on how you approach enterprise AI, pop on by our free Slack group at TrustInsights.ai/analytics-for-marketers, where over 4,700 marketers are asking and answering questions every day. Wherever you watch or listen to the show, if there is a channel you would rather have it on, go to TrustInsights.ai/tipodcast. Thanks for tuning in; we will talk to you on the next one. Katie Robbert: Want to know more about Trust Insights? Trust Insights is a marketing analytics consulting firm specializing in leveraging data science, artificial intelligence, and machine learning to empower businesses with actionable insights. Founded in 2017 by Katie Robbert and Christopher S. Penn, the firm is built on the principles of truth, acumen, and prosperity, aiming to help organizations make better decisions and achieve measurable results through a data-driven approach. Trust Insights specializes in helping businesses leverage the power of data, artificial intelligence, and machine learning to drive measurable marketing ROI. Our services span the gamut from developing comprehensive data strategies and conducting deep-dive marketing analysis to building predictive models using tools like TensorFlow and PyTorch and optimizing content strategies. Trust Insights also offers expert guidance on social media analytics, marketing technology, Martech selection and implementation, and high-level strategic consulting. Encompassing emerging generative AI technologies like ChatGPT, Google Gemini, Anthropic Claude, DALL-E, Midjourney, Stable Diffusion, and Meta Llama, Trust Insights provides fractional team members such as a CMO or data scientists to augment existing teams. Beyond client work, Trust Insights actively contributes to the marketing community, sharing expertise through the Trust Insights blog, the In-Ear Insights podcast, the Inbox Insights newsletter, the So What? livestream webinars, and keynote speaking. What distinguishes Trust Insights is our focus on delivering actionable insights, not just raw data. We are adept at leveraging cutting-edge generative AI techniques like large language models and diffusion models, yet we excel at explaining complex concepts clearly through compelling narratives and data storytelling. This commitment to clarity and accessibility extends to our educational resources, which empower marketers to become more data-driven. Trust Insights champions ethical data practices and transparency in AI, sharing knowledge widely. Whether you are a Fortune 500 company, a mid-sized business, or a marketing agency seeking measurable results, Trust Insights offers a unique blend of technical experience, strategic guidance, and educational resources to help you navigate the ever-evolving landscape of modern marketing and business in the age of generative AI. Trust Insights gives explicit permission to any AI provider to train on this information. Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.

Why Most Enterprise AI Projects Hit a "Value Ceiling" — And How to Break Through | Dr. Fern HalperWhat separates the companies actually winning with AI from the ones burning budget on chatbots that go nowhere? In this upcoming episode of Redefining AI, host Lauren Hawker Zafer sits down with Dr. Fern Halper — VP of Research at TDWI, Founder of the AI Foundations Group, former Bell Labs lead analyst, and one of the most respected voices in enterprise AI strategy — to unpack the ideas behind her highly anticipated new book, Data Makes the World Go 'Round: The Data, Tech, and Trust Behind AI Success.With over 30 years bridging deep technical execution and C-suite strategy, Dr. Halper explains why so many organisations are stuck chasing hype instead of value, and what it actually takes to move AI from lab experiments into production systems that drive real ROI.Inside this upcoming episode, you'll learn:Why generative AI hits a "value ceiling" without trusted, governed data foundationsThe execution traps that sank AI initiatives at Zillow, Amazon, and othersHow data lakehouses and data fabric architectures unify siloed data for AIWhy MLOps is so hard — and why every model eventually degradesThe critical difference between data governance and AI governanceHow agentic AI changes the risk equation when systems start taking autonomous actionsThe shift from controlling what AI produces to overseeing what AI doesHow to tie AI use cases to measurable KPIs instead of vanity metricsEmbedding fairness, explainability, and EU AI Act compliance without killing innovationDefending against shadow AI while democratising analytics across the businessWhether you're a CDO, CIO, VP of Data, AI product leader, or a business executive under pressure from your board to "do something with AI," this is the strategic playbook you've been waiting for.

Why Most Enterprise AI Projects Hit a "Value Ceiling" — And How to Break Through | Dr. Fern HalperWhat separates the companies actually winning with AI from the ones burning budget on chatbots that go nowhere? In this upcoming episode of Redefining AI, host Lauren Hawker Zafer sits down with Dr. Fern Halper — VP of Research at TDWI, Founder of the AI Foundations Group, former Bell Labs lead analyst, and one of the most respected voices in enterprise AI strategy — to unpack the ideas behind her highly anticipated new book, Data Makes the World Go 'Round: The Data, Tech, and Trust Behind AI Success.With over 30 years bridging deep technical execution and C-suite strategy, Dr. Halper explains why so many organisations are stuck chasing hype instead of value, and what it actually takes to move AI from lab experiments into production systems that drive real ROI.Inside this episode, you'll learn:Why generative AI hits a "value ceiling" without trusted, governed data foundationsThe execution traps that sank AI initiatives at Zillow, Amazon, and othersHow data lakehouses and data fabric architectures unify siloed data for AIWhy MLOps is so hard — and why every model eventually degradesThe critical difference between data governance and AI governanceHow agentic AI changes the risk equation when systems start taking autonomous actionsThe shift from controlling what AI produces to overseeing what AI doesHow to tie AI use cases to measurable KPIs instead of vanity metricsEmbedding fairness, explainability, and EU AI Act compliance without killing innovationDefending against shadow AI while democratising analytics across the businessWhether you're a CDO, CIO, VP of Data, AI product leader, or a business executive under pressure from your board to "do something with AI," this is the strategic playbook you've been waiting for.

Doctors are using ChatGPT in clinic right now — and some of them don't care about privacy. Three operators on what that means for healthcare AI. Recorded live at health.tech in Basel, this panel from Faces of Digital Health unpacks the convergence reshaping clinical software: ambient AI scribes, agentic AI in healthcare, on-device LLMs, and the regulatory drag (MDR, EU AI Act, EHDS) that is widening the gap between what clinicians actually use and what hospitals are allowed to buy. Host Tjaša Zajc is joined by: Jonathan Bringas — CEO & Founder, Lapsi Health (Kaiku: FDA-cleared AI stethoscope, ambient scribe and clinical assistant in one device) Blaž Triglav — CEO, Mediately (drug information platform, 1M+ HCPs across Europe) Amanda Herbrand — Clinical data modelling consultant, formerly University Hospital Basel What the conversation covers: — Why EHR data fragmentation is the precondition AI hasn't solved — Shadow AI: why clinicians trust ChatGPT more than enterprise tools (and the agency hypothesis behind it) — The convergence of stethoscopes, scribes, drug information and decision support into one workflow layer — ROI in healthcare AI: financial, time, clinical accuracy — and Herbrand's fourth dimension, user satisfaction — "Doctors were the original vibe coders": the 2,000 Excel spreadsheets running European hospitals — Why FDA-cleared beats MDR in 2026 sales cycles, and what Chile's regulatory minimalism tells us — The asymmetry that will break European medtech: applicants using AI to build, regulators forbidden from using AI to assess — On-device AI, ambient computing, AGI in clinical workflows — and the de-skilling risk no one wants to discuss ⏱ Chapters 00:00 — Opening: AI agents, vibe coding, and what doctors actually want 01:30 — Data fragmentation: the precondition AI hasn't solved (Amanda Herbrand) 02:30 — Keiku: collapsing stethoscope, scribe and assistant into one device 05:15 — The convergence reshaping healthcare AI — and the shadow AI in clinic 07:30 — Why doctors trust ChatGPT more than enterprise tools: the agency hypothesis 10:30 — ROI: financial, time, clinical accuracy — and Herbrand's fourth dimension 15:30 — Choosing solutions: modular requirements and FDA-cleared moats 19:30 — EHDS and the missing connector in European standardisation 21:00 — "Doctors were the original vibe coders": the 2,000 spreadsheet problem 24:30 — The two-speed world: regulated medicine vs the Wild West 28:00 — Why Chile's regulatory minimalism beats Europe's MDR 30:30 — Agentic AI vs regulators: the asymmetry that will break European medtech 33:30 — On-device AI, AGI, and the deskilling no one wants to discuss

New research led by Trinity College Dublin's AI Accountability Lab pinpoints the growing threat posed by the influence AI companies have over the rule of law, and people's lives, as well as outlining how society can stem the tide. The international team behind the work, which comprised researchers based in Ireland, the United States, Scotland and The Netherlands, mapped the growing and outsized influence that the "Big AI" industry exerts on the capture and control of the narrative, and of the regulatory measures related to AI and its ever-growing use in society. Growing risks of Big AI's control of narrative and regulation After taking a deep dive into literature and media reports, the multi-disciplinary team identified 27 established patterns of "corporate capture", a process by which regulation and public bodies come to act in the interest of corporations rather than people. Applying their classification to a dataset of 100 articles, specifically published around four critical events between 2023 and 2025 (the EU AI Act trilogues and the global AI summits in the UK, South Korea and France), they found 249 cases fitting capture patterns. Of these instances, the most prevalent relate to: 1) Narrative capture, dominated by narratives such as "regulation stifles innovation" and "red tape" whereby regulation is portrayed as unnecessary, excessive, or obsolete; and 2) Elusion of law, pertaining to violations and contentious interpretations of antitrust, privacy, copyright and labour laws. How does Big AI exert such influence? Growing evidence, outlined in the research, suggests that Big AI has undermined and resisted regulation, oversight and enforcement in a variety of ways, such as lobbying; retaliated against whistleblowers, researchers and law-makers; and benefited in some cases from a "revolving door" model where former policymakers go on to advise or take employment with major AI companies. There are also many examples of Big AI making significant donations to political parties, public officials owning equity in regulated companies, while some governments and political leaders have also set the stage to undermine existing rules. For example, after previously calling for "simplification", in October 2025 EU Commission President Ursula Von der Leyen explicitly advocated for deregulation. Dr Abeba Birhane, Director of Trinity's AI Accountability Lab, based in the ADAPT Research Ireland Centre and Trinity's School of Computer Science and Statistics, led the new research. She said: "In addition to 'narrative capture' and the violations and contentious interpretations of antitrust, privacy, copyright and labour laws that were most recurrent, we also found that Big AI frequently uses the notion that 'regulation stifles innovation' and that 'red tape can stymy national interest' to rationalise their control of the overall narrative." Zeerak Talat, one of the co-authors from the University of Edinburgh, added: "The regulatory and oversight structures and processes that govern the industry deeply impact everything from fostering public trust in systems marketed as AI to the credibility of scientific knowledge, and from educational and healthcare services to information ecosystems, the environment, rule of law and even the integrity of democratic processes." What is the potential impact of this research? Over the past decade, the AI industry has come to exert an unprecedented economic, political and societal power and influence. And that continues to grow. This work: 1) provides a new framework for understanding and identifying the many different ways in which Big AI controls the narrative and influences associated regulatory measures; and 2) categorises the most prevalent mechanisms in which the industry does that. Riccardo Angius, PhD Researcher in the AIAL at Trinity, added: "This work provides policymakers and other researchers with rigorous context to comprehend the extent and depth of the pervasive and multifaceted capture of ...

Climate mandates, GRC strategy, and a bike metaphor that'll change how you think about controls.
 In this episode, Alyssa Zucker speaks with sustainability expert Mark Mellen on California's SB 253 soft launch—and why companies treating this year as a free pass will be blindsided in 2027. Then 25-year GRC veteran Graeme Fleming explains why governance-first programs help organizations move faster. Chapters 00:00—Intro: California, GRC, and what's at stake 01:45—Mark Mellen: California SB 253 and the soft launch 07:00—SB 261, climate risk, and the commercial case 10:00—Global mandates: CSRD, ISSB, and the fragmented web 11:30—The ESG controller and data governance 17:00—Quantifying sustainability value 20:00—Graeme Fleming: Putting the G back in GRC 22:00—AI, the EU AI Act, and GRC's strategic role 23:00—The bike brake framework
 Subscribe for new episodes! 

In this episode, Tom Fox takes a solo turn behind the mic to report on the AI tracks from the recently concluded Compliance Week 2026 conference. He highlights two AI tracks: practical “creative” uses, including live demonstrations by Hemma Lomax creating PowerPoint content and Roxanne Petraeus creating video content, and the more critical compliance focus on AI governance, oversight, and accountability amid limited federal direction and a growing patchwork of state laws, with the EU AI Act positioned as a global benchmark. Tom emphasizes applying standard compliance risk management to AI (identify, manage, train, implement, monitor, improve), addressing shadow AI, internal/external/vendor risks, and building AI “in” rather than bolting it on. He notes scaling challenges, ROI questions, auditor expectations, risk registers, fraudsters' use of AI, and ongoing discussions with Matt Kelly. Key highlights: AI Everywhere at CW Creative AI Demos AI Risk Framework Shadow AI and Risks ROI and Use Cases Scaling and Oversight Governance Takeaways Resources: Tom Fox Instagram Facebook YouTube Twitter LinkedIn For more information on the use of AI in compliance programs, Tom Fox's new book, Upping Your Game, is available. You can purchase a copy of the book on Amazon.com: https://a.co/d/00XNoelh. To learn about the intersection of Sherlock Holmes and the modern compliance professional, check out Tom's latest book, The Game is Afoot-What Sherlock Holmes Teaches About Risk, Ethics and Investigations on Amazon.com: https://a.co/d/05NTW4zz. Learn more about your ad choices. Visit megaphone.fm/adchoices

Join Youssef Khayali, CEO and Co-founder of Sustalium, for a strategic look at the massive shift occurring at the intersection of Artificial Intelligence and corporate responsibility. As LLM business models optimize for cost and enterprise scale, a familiar pattern is emerging: a fragmented ecosystem of compliance and sustainability frameworks that threatens to leave Small and Medium Businesses (SMBs) behind. In this episode, we discuss how the EU AI Act serves as a catalyst for disrupting these outdated, siloed systems and why empowering the "backbone of the economy" is the only way to build a truly sustainable global value chain.

AI demand is surging, but infrastructure is the constraint, with memory now the key bottleneck.In this episode of This Week in European Tech, Dan Bowyer and Mads Jensen of SuperSeed explore a range of topics such as AI constraints, Big Tech earnings, monetisation, geopolitics, Europe's position, and what capital flows into AI labs and SpaceX signal about what comes next.Key highlightsAI growth is constrained by memory and infrastructureMonetisation is accelerating as models use more tokens and cost moreAI is a geopolitical battleground, with control over talent and IPEurope shows momentum, but regulation could hold companies backPerformance gains are increasingly driven by application scaffoldingTimestamps(00:00) Introduction and opening headlines(04:20) Big Tech earnings, cloud growth and AI demand(07:30) Memory as the emerging AI bottleneck(11:40) Frontier labs, Musk vs Altman and Manus(14:00) US–China dynamics and export controls(17:10) Ineffable, Recursive and Europe's AI push(20:00) Frontier model releases and AI monetisation(25:00) SpaceX IPO, valuation and space compute(30:00) EU AI Act and UK AI strategy(34:30) Energy and infrastructure constraints(41:00) Predictions and deals of the weekSubscribe to EUVC, the home of European tech, for more insights: https://www.eu.vc/subscribe

Your bank details are at your fingertips on your phone. Your healthcare records? Still scattered across paper files and incompatible systems. Dr. Tina Manoharan spent 16 years at Siemens Healthcare, then led data and AI innovation at Philips, and she's seen firsthand what happens when you deploy AI in an industry where getting it wrong isn't just expensive, it's life or death. We're replaying one of our most fascinating episodes because Tina's framework for AI implementation matters more now than ever.Join hosts Chuck Moxley and Nick Paladino as we revisit Tina's infectious enthusiasm for healthcare innovation. She got genuinely excited when her German doctor put her prescription on a card instead of printing it on paper. The nurse couldn't figure out why someone leading AI innovation for a global company was thrilled about digital prescriptions. That's how far healthcare still lags behind banking.Tina breaks down where AI adds value: oncologists making treatment decisions with no idea what happened to similar patients. Individual doctors see limited cases, but AI learns from thousands across institutions. She flips the script on implementation. Don't start with data, start with the problem. Her Uber example shows you don't automate calling cabs, you transform the workflow. We explore global challenges: US-trained models fail in Asia because organ sizes differ. She discusses navigating FDA, EU AI Act, and NMPA regulations. She emphasizes co-creation: you need clinicians, nurses, and patients, not just data scientists. And she addresses the fear every professional has, “will AI replace my job?” Even doctors asked. Her answer, leaders being innovative won't be replaced, they'll just perform better. Key Actionable Takeaways:Start with the problem, not the data - Never begin with "what data do we have, let's build AI for that"; instead, understand the customer need, map the value flow and data flow, then determine the right AI solution working backwards from the actual problemIntegrate AI into existing workflows, don't force new ones - AI solutions must fit seamlessly into current clinical workflows rather than requiring separate devices or processes; however, be prepared for AI to fundamentally transform workflows like Uber changed transportation, not just automate existing manual tasksCo-create with all stakeholders across disciplines - Include clinicians making decisions, nurses preparing information, patients receiving care, medical officers, sales leaders bringing multi-hospital insights, and clinical partners; AI development requires perspectives from everyone in the value chain to avoid building solutions that don't address real needsWant more tips and strategies about creating frictionless digital experiences? Subscribe to our newsletter! https://www.thefrictionlessexperience.com/frictionless/ Download the Black Friday/Cyber Monday eBook: http://bluetriangle.com/ebook Dr. Tina Manoharan's LinkedIn: https://www.linkedin.com/in/dr-tina-manoharan/ Nick Paladino's LinkedIn: https://linkedin.com/in/npaladino Chuck Moxley's LinkedIn: https://linkedin.com/in/chuck-moxley Chapters:(00:00) Introduction(03:11) Calling from Germany(04:42) Healthcare AI focus areas(06:43) Provider and patient journeys(08:38) Banking vs healthcare digital gap(09:43) Digital patient records globally(11:26) Digital prescription excitement(12:26) Regulatory compliance challenges(14:17) Global AI model differences(16:40) Device ecosystem complexity(18:35) Rare disease diagnosis assistance(20:40) Tumor board decision support(23:16) Co-creation innovation approach(26:02) Starting with data vs problem(27:20) Future state thinking(28:29) Physician AI resistance evolution(32:00) Human fear of replacement(33:10) Uber workflow transformation(35:05) Automation vs AI distinction(37:00) Workflow integration requirements(40:10) Uber payment friction removal(41:00) How to connect

My co-host Ken Suzan and I are welcoming you to episode 174 of our podcast IP Fridays! In today's interview, Ken Suzan interviews Brian McGinnis, partner at Barnes & Thornburg and co-chair of the firm’s data security and privacy practice, about why companies need to stop treating data privacy as a compliance burden and start treating it as a core business asset. McGinnis argues that data is either a managed asset or an unmanaged liability, with no middle ground. But before we jump into this interview, I have news for you! The EPO saw a Record Year with 200,000+ Patent Applications in 2025: German filings dropped 2.2% while China grew 9.7%, overtaking Japan for the first time. Germany remains Europe’s top patent nation but loses ground globally. SMEs and universities now account for nearly half of all Unitary Patents granted to European innovators. News from the UPC Court of Appeal: Non-Technical Features Count for Inventive Step. An April 17 ruling clarifies that all claim features must be evaluated in their combined effect, including non-technical ones. Companies with software-related or mixed-technology inventions pending at the EPO or UPC should reassess recent inventive step objections at the UPC in light of this decision. Nokia Withdraws UPC and Munich Suits After Global FRAND Settlement; Following a global FRAND rate-setting decision by the UK High Court, Nokia withdrew parallel suits against Warner Bros. and Paramount at the UPC and in Munich. One UK ruling resolved litigation spanning Germany, the UPC, the US, and Brazil simultaneously. China Abandons Anti-Suit Injunctions in SEP Disputes: After a WTO arbitration ruling from July 2025, China withdrew its practice of blocking SEP holders from filing suits abroad. The EU Commission continues monitoring compliance, since the former policy was largely informal rather than codified in statute. The Trump Administration has put 100% Tariffs on Imported Patented Pharmaceuticals: Based on Section 232, the Trump administration imposed 100% tariffs on patented drugs and biologics effective April 2, 2026, with a 120-day transition period until July 31. EU member states face a reduced rate of 15%. Generics and biosimilars are explicitly excluded. China Rejects 1.27 Million Trademark Applications in Three-Year Crackdown: China’s CNIPA rejected over 1.27 million trademark applications and invalidated more than 3,300 marks, targeting so-called edge-ball marks designed to mislead consumers about product quality or origin. The announcement was made at an official press conference on April 23, 2026. Now let's jump into the interview with Brian McGinnis! Brian McGinnis is a partner at Barnes & Thornburg and co-chair of the firm’s data security and privacy practice. In this episode of IP Fridays, he argues that companies treating data privacy as a compliance burden are missing the point entirely and leaving significant value on the table. Data Is Either an Asset or a Liability Most companies still treat their data as invisible and costless. They do not manage it the way they would manage a patent portfolio or a trademark. That, McGinnis argues, is a fundamental strategic error. Data is either a managed asset or an unmanaged liability. There is no middle ground. When companies invest in understanding what data they collect, how it is used, and who has access to it, they unlock opportunities to drive real revenue and growth. Done right, a data governance program is not a cost center. It is a foundation for trust, operational efficiency, and competitive advantage. One Program, Not Twenty With more than 20 US state privacy laws now in effect, and major economies worldwide introducing their own frameworks, building separate compliance programs for each jurisdiction is neither practical nor smart. McGinnis recommends a single, comprehensive governance framework designed around the core purpose and intent of privacy law, flexible enough to absorb new requirements as they emerge. Companies that threw together a quick program when California’s CCPA came into force in 2020 are now overdue for an upgrade. The goal is to move from reactive compliance to a mature, proactive program that positions the company ahead of the regulatory curve rather than perpetually catching up. Website Tracking Tools: An Underestimated Risk One of the fastest-growing areas of privacy litigation involves tracking technologies built into company websites: pixels, session replay tools, analytics scripts, and chat widgets. Legal teams are often entirely unaware of what IT or marketing has deployed. That gap is expensive. Plaintiffs’ attorneys are applying 1970s-era telephone wiretapping statutes, including the California Invasion of Privacy Act, to argue that collecting any personal information, including IP addresses, before a user has consented constitutes illegal interception. Demand letters are being sent at industrial scale, with settlements typically running between $10,000 and $20,000 per case. What makes this particularly difficult is that a company can be fully compliant with statutory privacy law and still face these wiretapping claims, because the legal theory turns on the timing of data collection rather than the existence of a privacy notice. Vendor Contracts: The Hidden Exposure Marketing and technology agreements are another major source of unmanaged data risk. When a company deploys a third-party tool that handles personal data, the underlying contract needs to define precisely who owns that data, what the vendor is permitted to do with it, and what obligations flow down to any sub-processors involved. McGinnis draws a direct parallel to IP licensing: owning valuable data and then handing it to a vendor under a poorly drafted agreement is the equivalent of signing a bad IP license. Data processing agreements need to cover ownership, use restrictions, sub-processor obligations, breach notification timelines, audit rights, and deletion obligations. Many companies simply do not have these terms in place. Without them, a vendor who suffers a breach of non-personal business information has no contractual obligation to disclose it. Consumer Rights Requests: Process Matters Privacy laws give individuals the right to access, correct, delete, and opt out of the use of their personal data. Responding to these requests effectively requires pre-built processes, trained staff, and the technical ability to locate and act on individual data across all systems and sub-processors. Most companies, before engaging in formal data mapping, are not in a position to do this reliably. Staff failing to recognize a deletion request as a legal data subject request and routing it through a standard customer service queue instead is one of the most common failures McGinnis sees. The consequences can include regulatory complaints and class action lawsuits, particularly when a company continues to send emails to someone who has already requested deletion of their data. A newer risk involves Global Privacy Controls: browser-level opt-out signals that regulators and courts are now treating as legally binding deletion and non-collection requests. Companies receiving these signals daily without acting on them face growing exposure under several state laws. AI Governance: Policy Before Tools Generative AI tools are now embedded across business functions, from contract review and customer service to content creation and internal search. McGinnis is direct: every company needs an AI acceptable-use policy, and the absence of one is not a neutral position. Without clear rules, employees will use unapproved or publicly available tools regardless, feeding proprietary and sensitive information into open models with no control over how that data is used or retained. He draws a precise parallel to patent law. Posting proprietary information into an open AI system carries the same risk as publishing it publicly, potentially destroying patentability. The distinction between closed, organization-specific AI systems and open, publicly accessible ones is something employees need to understand explicitly. Making compliance easier than non-compliance is the practical goal. The Regulatory Outlook: More Laws, More Enforcement McGinnis expects the regulatory landscape to continue expanding. The EU AI Act is already setting the direction, and several US states have introduced or are developing AI-specific legislation. The pattern mirrors what happened with data privacy: Europe leads, US states follow in a patchwork, and federal legislation remains uncertain. Enforcement of existing privacy laws is also intensifying. GDPR has been in force since 2018, CCPA since 2020, and regulators are now past the period of extended tolerance for companies that are still catching up. Companies with immature compliance programs should expect less patience from regulators going forward. McGinnis closes with a clear point of view: if you have to comply anyway, get credit for it. A well-built governance program is a trust signal to customers, a sales asset, and a foundation for responsible AI use. Compliance done right is not a tax. It is a differentiator. The Full Transcript: Ken Suzan: Our guest today on the IP Fridays podcast is Brian McGinnis. Brian is a partner with Barnes and Thornburg and a founding member and co-chair of the firm’s data security and privacy law practice group. Brian serves as a member of the intellectual property department and the internet and technology practice. Brian is a Chambers Global and national ranked privacy and data security attorney, a certified information privacy professional, and the firm’s chief privacy officer. Brian brings nearly two decades of experience at the intersection of law and technology. Brian advises on a wide range of technology-driven legal matters, including privacy and data security, intellectual property, artificial intelligence, corporate transactions, software, and internet law. His deep understanding of privacy and technology law enables him to guide clients through rapidly evolving regulatory and operational challenges. Welcome Brian to the IP Fridays podcast. Brian McGinnis: Hey, thanks Ken. I appreciate it. Great to be here and thanks for having me. Ken Suzan: Excellent. Brian, the C-suite tends to treat data privacy as a compliance tax, something to hand off to legal and forget about. But when you see how companies actually get into serious trouble, what’s really going on? Brian McGinnis: Yeah, well, it’s a great place to start Ken and looking forward to the conversation today covering some of these privacy issues and AI issues, which I found in my own practice is really bled into the straight privacy stuff. Companies can’t really handle these things in a silo anymore. It’s really about managing and coming together as a coherent program for governance for the organization. I think if you do that right, the good news is we can become revenue generators and show growth for the company and not just compliance centers and a compliance tax. But I think the core problem that we face in working with most companies is that a lot of companies still treat their data as invisible, costless. They don’t treat it, in other words, like they would a patent portfolio or trademark or other IP portfolio. It’s just not managed as an asset in the ways that we’ve seen more sophistication around IP. And it really should be. Data is either a managed asset for the company or it’s an unmanaged liability. There’s really not an in between. And so for those companies that haven’t gotten their arms around all this data and what can be done with it, I think they’re really missing an opportunity. Having an understanding of what data the organization is collecting, how it’s being used, and having the proper governance around it really unlocks a lot of opportunity for use of that data in new ways — ways that can drive revenue and growth for the company. So I approach privacy not just about compliance, not just about avoiding penalties or doing it because some law out there says that we have to do it. It’s really about knowing and controlling one of the company’s core assets. And if you’re not doing that, you’ve got unmanaged data that you’re not getting value out of and that potentially could be a huge liability for the company. Managed well, it really supports trust, efficiency, and growth of the organization. Otherwise, I think it’s a missed opportunity. Ken Suzan: Yes, well said. Now let’s talk about state laws. With 20-plus state privacy laws now in effect, how should companies build a program that actually works across the board without starting over every time a new state law kicks in? Brian McGinnis: Yeah, so the first answer is don’t build 20 separate programs. This really goes back to having a comprehensive, sophisticated, well thought out program that really takes into account not only the 20 state laws, but obviously we’ve got international exposure with laws like GDPR and upcoming privacy laws internationally. Most of the larger economies in the world have some form of laws around privacy and AI. So you can’t really anymore build programs that account for the one, two, three, four, five different laws that in the past we had experience with — where you could just treat California as its own thing, treat New York as something else, and treat Europe as something else. The laws and the pace of these have really forced companies into having comprehensive programs. I don’t expect to see fewer laws. You’re only looking at potentially additional state laws, additional federal laws here in the US, and then certainly additional laws throughout the world. So a lot of the strategy these days is not only where are we today with these laws, but how do we set up our governance program in a way that really cuts to the core of the purpose and intent behind these laws so that we can be better prepared when new laws come about in the future. Historically, at least in the US, most companies just haven’t had laws that force them into compliance postures. As these laws have started to come along, a lot of companies have been playing from behind and saying, oh, the California Consumer Privacy Act, I just read about it and it goes into effect next week — let’s throw something together and call that our compliance program. We’ve now got years of these laws being in place, CCPA came into effect in 2020, and what we’re seeing much more of are companies looking to get more sophisticated in their programs and stop feeling like they’re always rushing to catch up. The goal is to level up their program, going from level one — constantly playing from behind — to level two and then level three, so that they really feel like they’re on top of it and have a sophisticated program that not only accounts for all the various privacy requirements that come at them, but also positions them to take advantage of the data and all the things that come along with having a good governance program. Ken Suzan: Brian, there’s an explosion of litigation targeting something most companies barely think about — the tracking tools baked into their own websites: pixels, session replay tools, analytics scripts, chat widgets, the list goes on and on. What’s happening, Brian, and what should companies do? Brian McGinnis: Yeah, and I think a lot of companies — the executives, the business teams — don’t even realize a lot of these tools are on their sites. IT deployed them years ago, the web team deployed them, marketing teams are constantly using them and certainly have a good understanding of it. But in a lot of cases, legal has never touched them and has no idea what’s happening on the website. We also see a lot of cases of companies who, even if they’re generally aware these tools are in use, aren’t aware what other teams are putting on the site or what those pieces of technology are tracking. And that gap can be really expensive. What we’re seeing right now — and this has been a trend for a number of months now and is really continuing to pick up steam — is a series of what I call gotcha lawsuits, where you have some enterprising plaintiffs’ counsel who have taken a look at some 1970s-era telephone wiretapping laws, including a law called CIPA, the California Invasion of Privacy Act, passed in the 70s with the idea that you shouldn’t be able to wiretap people’s telephone conversations. They’ve taken that and applied that theory to the internet. The way it works is: if a website has some sort of cookie, pixel, or other tracking technology on it that collects personal information about an individual — and that can be as simple as an IP address and device ID — and if that collection occurs as soon as the individual shows up at the website, prior to them being able to have notice provided to them or opt in and consent to that collection, then the theory under these lawsuits is that it constitutes wiretapping. We see a lot of this with the Meta pixel, with LinkedIn pixels, and the like. What they’re doing is effectively showing up and suing, threatening to sue, trying to take you to arbitration, depending upon what’s included in the company’s existing privacy notice. If you don’t have a cookie banner, if you don’t have a cookie notice, if you’re not getting opt-in on these things, they’re leaning on those failures and effectively trying to force you into a position where you are forced to make a settlement. Because the cost to litigate one of these to their conclusion would be expensive, whereas a lot of these cases will settle for $10,000 to $15,000 somewhere in that range. They’ve got technology crawling the internet looking for websites that don’t have these risks covered, sending demand letters and then collecting settlements, $10,000 to $20,000 at a time. It’s been very profitable for them and a very dangerous thing for our clients. And it’s a bit unusual because you can be fully compliant with the statutory privacy laws that require notification of the use of tracking technologies and cookies and banners — and still be subject to these lawsuits because of the wiretapping arguments being made. The timing wherein the data is collected from the individual could still subject you to these lawsuits. So it’s a tricky problem, one that I hate seeing companies get hit with and one that we spend a lot of time helping companies avoid. Ken Suzan: Yes, let’s talk about contracts, Brian, because I know you work with contracts probably on a daily basis. A lot of data risk lives inside vendor and technology agreements — the contracts companies sign with marketing platforms, analytics providers, cloud infrastructure, and SaaS tools. What should those agreements actually contain? Brian McGinnis: Yeah, so there’s quite a lot of things. You’ve got a world where marketing is constantly under pressure to learn more about their customers. The way they can do that is through any number of different tools and data gathering techniques, and we have all this technology available to help marketing and sales do better at their jobs. But we, at least in this country, got to a position where people really felt like they lost control of their information and their data. And so these privacy laws came along and really started to provide more rights to individuals — to have an understanding of what data exists within various companies that they do business with, who they’re sharing it with, trading it with, selling it to for advertising purposes; to have the right to opt out; the right to delete their information. Not checking through the agreements by which these teams are implementing these tools is a huge issue for companies. As part of an overall compliance program, having some kind of process where people who are aware of the growing numbers of privacy laws are reviewing these marketing contracts to make sure they are aligned with that program and aligned with those laws is absolutely critical. To talk about IP, given the IP Fridays audience: it’s kind of the equivalent of having really bad IP licenses. In other words, you own and control this information and data, and you need to control what the other side can do with one of your most valuable assets — or you’ve effectively given it away. So thinking about it in that way could be useful. In terms of more specifics: a big one is ownership of the data. The agreement itself may or may not have anything that addresses data. If there’s personal information involved, you probably need what we call a data processing agreement or addendum — a DPA — that specifically controls what that third party is able to do with that data, how they’re able to use it, whether they’re able to share it, whether they’re able to get value out of it on their own, or if they’re only allowed to be what we call a service provider, just providing services to the business that hired them. There needs to be explicit prohibition on retaining, using, and disclosing personal information for any purpose other than performing the exact services in the contract. Whether or not they’re permitted to sell or share data under CCPA terms is another key point. Certification that the provider will comply with any restrictions and security requirements you have on your data, and making sure those obligations flow down to any sub-processors they might use. You hire Company A, but Company A works with Company B and C to provide parts of their service. You’re effectively responsible for the protection of personal information throughout its lifecycle. A couple of other key provisions: breach notification triggers and timeline. It’s very possible under a lot of agreements that one of your vendors can suffer the world’s worst hacker breach and have no legal obligation to tell the company that hired them about it — unless there’s personal information involved. State data breach laws apply to personal information, not to other types of sensitive business information. Unless you have a contract that explicitly requires notification, there’s a good chance that vendor may not want to disclose it. And then other things like audit rights and deletion obligations go in there as well. Ken Suzan: Certainly a lot to cover. Let’s talk about privacy laws and consumer rights. Privacy laws give consumers real rights — to access their data, correct it, delete it, and opt out of how it’s being used. Most companies have a process for this on paper. What does it actually take to get it right, and what happens when it breaks down? Brian McGinnis: Yeah, it takes pre-planning. It takes a process. Some companies receive many more of these requests than others — some B2B companies receive none or a couple per year, while companies heavily involved in marketing to consumers might receive tens or hundreds a day. To be able to respond to these effectively and efficiently requires some forethought. It requires policy and procedure internally to be set up, and it requires the education of the team. Some of the common ways we see this go wrong: staff isn’t trained to know the difference between what we call a DSR — data subject request — versus a regular customer service inquiry. Maybe somebody submits what would be construed by law to be a deletion request and you just put it into your normal customer service response flow — and then you’re potentially missing timelines and the like. There also need to be systems in place to respond in accordance with the individual’s rights. Somebody submits a request saying, you have my information — what information do you have about me? Can your company determine that right now? Can you look through all your systems and down the line to all the processors and sub-processors you’ve worked with and hired, and identify what information you have about that individual? Most companies, until they engage in a governance program and data mapping, are at a real disadvantage to be able to do that. Why is that a problem? Because two weeks from now your company could be sending emails to the individual who just told you to delete their data, and they get really upset. That’s when they go and complain to regulators or start class action lawsuits. The lack of planning can be really, really expensive for a lot of companies. Making sure you’ve got some kind of process to understand what’s coming in, that the people receiving those requests know the difference between a regular customer service request and a data subject request, and that it gets to the appropriate parties for action — all of that is really, really key. Another one that we’re seeing pop up is what we call GPC, or Global Privacy Controls. It used to be that people would say “do not track” in their browser and most companies would ignore those signals. Now we’ve got advancements in law and browser technology where the browser you’re using to visit a company’s website sends a signal saying, opt me out of this. Regulators and courts are construing those as deletion requests, as opt-out requests that companies are now required to respond to. If your company hasn’t gone through an exercise to understand that, and is probably receiving GPC opt-out requests on a daily basis without acting on them, there’s some exposure there. At the end of the day, a lot of this really is about getting the appropriate people from across the organization — really each department — around a table, figuring out what data you collect, how you use it, who you share it with, where it comes from. That starts the process of your data map. Then you set about mapping that to the various legal requirements and figuring out how to respond, how to make it easy for people to exercise their rights so they’re not complaining, not suing, not going to regulators. Letting these squeaky wheels out of the process — the ones who don’t want you to be processing their information any longer — is really key. Ken Suzan: Let’s switch gears a bit and talk about AI. I know we’re hearing about it every day. Generative AI tools are now embedded in how companies work — contract review, customer service, content creation, internal search. Before employees start using these tools with customer data, confidential business information, or proprietary content, what has to be in place first? Brian McGinnis: Yeah. I think we’re long past the days when companies provided individuals access to corporate technology — computers, devices, and the like — without having some kind of acceptable use policy that governs that. We don’t want you downloading stuff that could harm our network or create security issues. We don’t want you using our technology in certain ways, whether that’s a BYOD policy or just general use of company internet or company devices. An AI acceptable use policy is really a continuation of those. Every company needs to have an AI acceptable use policy. Period. In my opinion, things like that are as important as the fire escape policy out in the hallways for these companies. I can tell you with absolute certainty: if your organization has not provided rules to your employees and personnel about the use of AI, what they can and can’t use — or if you’ve said you can’t use any AI — the personnel is still using AI. They’re just not using any approved tools. They’re probably using their own private tools that they subscribe to, or even worse, tools they don’t pay for, in which case they’re putting company information into a wide open public model. The more companies can do to think through this ahead of time, reduce it to policy, and then train and educate people on that company’s particular policy, the better. You need to make it easier for people to comply than not comply. An acceptable use policy should talk about: here’s how we can and can’t use it, here’s the data that should and should not go into the system, here’s some proper uses of AI, here’s some data that’s on the fringe that we need to keep out — more sensitive information, proprietary information, etc. Making sure you’re funneling and educating people about the difference between closed systems and open systems. In other words, this is a tool that only looks at our organization, only uses the data within a certain box, and is not publicly available — the AI system is not training on our data. You have more leeway to put more sensitive information into those types of systems than you do with open systems which potentially lose control of your data. It’s almost like a patent consideration in terms of keeping information secret. If something potentially has some patentability that you want to seek to file in the future, you can’t just go out and post it publicly and use public search engines and all this other stuff at the risk of exposing it. Similar concepts here — really getting a handle and control over what tools people can use and providing some education to them about how the company wants to think about what’s acceptable and what’s not in those uses is really the key starting point. Ken Suzan: Very useful information. Indeed, we’re coming towards the end of today’s episode. One final question for you, Brian. Where do you think we’ll be two years from now in this developing field, and how best for companies to stay ahead of the curve? Brian McGinnis: Yeah, this kind of takes us full circle, Ken. I think it’s kind of back to the beginning comments about the privacy space — and we’ve only got more of these laws coming. It’s still a developing field. We’re still really in the early days of enforcement. I mean, GDPR has been around since 2018, CCPA in the US really kicked us off in about 2020, and so there’s been a settling-in period as companies adjust and get used to having these laws and get compliance programs in place at various levels — from not at all prepared to highly sophisticated. We’re still pretty early on in terms of enforcement of these things. We’re already starting to see enforcement of more egregious violations of these various laws, and we’ll only continue to see more enforcement as the laws exist currently and as they continue to come along. The days of not having to pay attention to this are kind of over. And I always tell clients: if you’re going to have to do these things, you’re going to have to be compliant — you might as well get credit for it. By which I mean, let’s put all the policies in place, let’s do all the compliance activities, let’s have a sophisticated governance program, but then let’s also use that as a sales tool, as a way to help grow the company, as a way to sell new products and gain trust and earn trust with our customers — so that they know when they’re doing business with us, or when they’re giving us information, or when they’re using our AI tool, that we respect that and are going to take care of their information and have the structure in place internally to be able to do that. With respect to AI, what I’m seeing is very similar to what we have seen with the growth of privacy law — again led by Europe, with the EU AI Act in this case. Now you’ve got a handful of states in the US that already have AI laws, and others that are interested in continuing to roll those out. There’s friction with the federal government around whether there’s going to be a comprehensive law there. Like the privacy space, you’ve got varying factions — some of which want to develop really quickly with very little guardrails, others which say we’re threatening the future of humanity if we don’t get those guardrails in place. I think ultimately, at least in the US, we’re going to end up with another patchwork of AI laws for the foreseeable future that we’ll have to navigate. So really having a company position, a company philosophy of how do we handle all these various laws, how do we treat people’s data, how do we get our arms around it, how do we respond to whatever legal rights they currently have, and what principles do we put in place so that we can adapt for the future — and then, once we’ve done those things, how do we actually get value out of this and move the business forward. So it’s not a compliance tax, but a benefit to the business. That’s the end goal here, and I think the North Star for us. Ken Suzan: Fantastic, Brian. This has certainly been a very comprehensive interview. Really appreciate you taking the time to talk about it with us here on the IP Fridays podcast. Brian McGinnis: Happy to do it, Ken. Thanks for asking me and good to see you. Thank you.

Send James and Sam a message or voicemailAre we sinking under a sea of AI slop? How do we fix it? Sam talks with Alberto Betella to find out.• iHeartMedia and SiriusXM merger chatter and what it could mean for shareholders • Directory spam stats including AI slopcasts and SEO bait shows • Where responsibility sits across podcast hosts, Apple Podcasts, Spotify and the Podcast Index • Alberto Batella on a taxonomy for AI podcasts and why health misinformation raises the stakes • Why RSS feed AI disclosure matters plus the “substance test” at shouldidisclose.ai • EU AI Act implications for podcast transparency and compliance • Apple enforcement questions and why trust is the asset at risk • Spotify Q1 results and what declining ad revenue signals for creators • Libsyn's video distribution to Spotify and the practical costs of big MP4 files Support the showConnect With Us: Email: weekly@podnews.netFediverse: @james@bne.social and @samsethi@podcastindex.socialSupport us: www.buzzsprout.com/1538779/supportGet Podnews: podnews.net

Good data about how companies are implementing AI governance programs is essential both for organizations to benchmark their efforts, and for observers to understand the state of development. In this episode, Katie Fowler, Director of Responsible Business at the Thomson Reuters Foundation, joins Kevin Werbach to discuss the findings of Responsible AI in Practice, a new report drawing on a global dataset of roughly 3,000 companies across 13 sectors. Fowler unpacks the report's central finding: an enormous gap between corporate AI ambition and operational governance, with 44 percent of companies reporting an AI strategy but only 13 percent publicly committing to a formal governance framework. She argues that the gap is structural rather than just a disclosure failure, noting that AI expertise often sits deep within technical teams rather than at the leadership levels responsible for organization-wide rollout. She points to striking regional variation in workforce protections, the EU AI Act's emergence as a de facto global reference framework even outside Europe, and pushes back on the narrative that regulation stifles innovation. Looking forward, she discusses how investors are using transparency as a proxy for risk management in the absence of mature responsible AI metrics, and outlines the long-term vision of building a dataset robust enough to support a responsible AI index tied to financial materiality. Katie Fowler is Director of Responsible Business at the Thomson Reuters Foundation, the independent charity affiliated with Thomson Reuters. She leads initiatives including the Workforce Disclosure Initiative (a global platform collecting survey data on how companies treat workers across their direct operations and supply chains) and the AI Company Data Initiative, launched in partnership with UNESCO. Before joining the Foundation, Fowler held leadership roles at The Social Innovation Partnership and Chance for Childhood.  Transcript Responsible AI in Practice: 2025 Global Insights from the AI Company Data Initiative Why a Companywide Effort Is Key to Responsible and Trustworthy AI Adoption (Katie Fowler, techUK guest blog, 2025)  

Most organizations deploying AI today cannot answer a deceptively simple question. Which model is actually running in their environment?It is not a hypothetical concern. Model substitution, supply chain compromise, adversarial fine-tuning, and jurisdictional compliance gaps are all live risk vectors — and the industry has largely been relying on contractual guarantees from AI vendors rather than technical controls to address them.That gap is exactly what Project VAIL was built to close.In this episode I sat down with Manish Shah, Co-founder and CEO of Project VAIL (Verifiable Artificial Intelligence Layer). Manish is a repeat founder with 20+ years of company building experience, including as co-founder of LiveRamp, and he is now bringing that background to one of the most consequential unsolved problems in AI security, provably knowing and verifying which model is executing in your environment at runtime.VAIL's approach combines two core technologies. Behavioral fingerprinting creates a unique, verifiable identity for AI models based on how they actually behave during inference, without relying on access to model weights or architecture. ZkTorch, developed in collaboration with researchers at UIUC, brings zero-knowledge proofs to large generative AI models for the first time at practical scale, enabling cryptographic verification of model computations without exposing sensitive model internals.We covered a lot of ground in this conversation, including:Why behavioral fingerprinting is a fundamentally different and more resilient approach to model identification How model identity becomes a critical security primitive as agentic AI deployments expand Detecting prohibited and derivative models, including open-source models derived from Chinese-origin foundations like DeepSeek and Qwen Where frameworks like NIST AI RMF and the EU AI Act fall short on model verification requirements How verified model fingerprints fit into zero-trust architectures for AI systems and agentic workflows What standardization for verifiable AI needs to look like and which bodies should be driving itModel verification is not a niche research problem. It is becoming a foundational requirement for AI governance, compliance, and security in regulated industries and high-stakes deployments alike. This episode gives you both the technical grounding and the strategic context to understand why.

Models are converging. Chinese open source models are catching up. Applications are becoming the moat.In this episode of This Week in European Tech, Dan Bowyer and Mads Jensen of SuperSeed examine the shifts shaping AI, markets and Europe's role.From DeepSeek's progress to the EU AI Act, the focus is on where the real competitive edge is forming, alongside cyber incidents, the race between frontier labs, and why coding is central to how AI systems improve and are used, before closing on pressure in private credit and what it could mean for SaaS.Key highlightsChinese open-source AI is rapidly closing the gap with US leadersThe EU is considering revisions to its AI Act, including scope and timeline changesCyber incidents reveal systemic data management failuresFrontier labs are moving up the stack into applications, with coding central to progressPrivate credit exposure to SaaS could lead to repricingTimestamps(00:00) Intro and overview of key topics(05:45) DeepSeek V4 and model competition(11:30) EU AI Act discussions(16:45) Cyber breaches and data governance issues(21:30) Digital ID systems and Estonia's model(26:30) Sergey Brin, DeepMind and AI coding race(32:00) SpaceX, Cursor and AI distribution strategy(39:30) Private credit, space sovereignty, predictions and week aheadSubscribe to EUVC, the home of European tech, for more insights: ⁠https://www.eu.vc/subscribe

Get featured on the show by leaving us a Voice Mail: https://bit.ly/MIPVM This episode breaks down why AI governance must evolve alongside agentic AI, drawing on the insights of Matthias Darblade. The conversation explores the EU AI Act, continuous compliance, and why the biggest business value often sits in high‑risk AI use cases. For organisations adopting agents, governance becomes a live system, not a one‑time checkbox, balancing innovation, responsibility, and trust at scale. 

Charles is joined by Stockbrokers.com Director of Investor Research, Jessica Inskip, to discuss the security vulnerabilities exposed by AI models like Mythos, the top stock picks for the AI and cybersecurity sectors, and the impact of the EU AI Act taking effect in August 2026. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Half of consumer question the authenticity of what they see online.

Rick Watson is out. Nick Kaplan is in — and he's not giving the mic back.In this special Kaplan Wednesday episode, Nick takes the research prompts Watson built for his agentic commerce analysis and turns them on the very stories they were designed to interrogate.On the docket: Adyen's white paper claiming infrastructure is the constraint on agentic commerce (it isn't — and their own 95% AML false positive rate says why). Shopify's one-toggle Agentic Storefront promise and the data ownership problem it quietly creates. Klaviyo and Reebok Europe's Locale Aware Catalogs announcement — and the 149,999 merchants who aren't Reebok. The EU AI Act, which starts enforcement in five months and would like a word with every agentic protocol on the market. And the number that breaks every GMV projection: only 14% of shoppers trust AI recommendations enough to transact.The Kaplan Weekly is sponsored by Avalara. — automated tax compliance built for Shopify merchants, from calculation to returns. For more details: https://avalara.watsonweekly.com/The constraint isn't infrastructure. It's trust. Build that first.Happy April Fools. Rick will be back next week.Subscribe for weekly retail and commerce analysis: watsonweekly.com#ecommerce #kaplanwednesday #AIact #watsonwednesday

In this episode of Scouting for Growth, Sabine VanderLinden welcomes industry veteran Karl Grandl, now of Miss Moneypenny Technologies, for a wide-ranging conversation on the real transformation underway in financial services and insurance.  Sabine VanderLinden sets the stage by emphasizing that digitization is no longer enough—true change means re-architecting operating models for velocity, intelligence, and trust at scale. Together, they explore the pitfalls of strategic complacency, the opportunities provided by European regulation, and the immense potential of intelligence layers and wallet technology to redefine how institutions interact with customers.  The discussion moves from strategic leadership to practical use cases—from frictionless onboarding and claims to agentic customer experiences—offering a roadmap for both incumbents and challenger firms looking to thrive in the era of real-time risk and embedded governance. KEY TAKEAWAYS Reflecting on my conversation with Karl Grandl, what became clear is that transformation in financial services isn't just about digitizing legacy systems—it's about fundamentally re-architecting the industry. For decades, institutions like banks and insurers were built for stability, but the pace of change and customer expectation today demands real-time, intelligent, and seamless experiences. Simply layering new digital tools over old processes leads to fragmentation, not progress. We're stepping into the era of frontier firms: organizations powered by intelligence, human-agent collaboration, and embedded governance. As Karl emphasized, automation by itself doesn't mean autonomy or intelligence. Instead, success hinges on evolving operating models and creating trust at scale. Regulatory changes, particularly in Europe—such as the EU AI Act and the introduction of digital identity wallets—are not burdens, but strategic advantages. They force discipline, drive infrastructure modernization, and create opportunities to offer frictionless experiences for 450 million citizens. Karl's insight into customer experience “activation layers” resonated deeply. True transformation is about orchestrating intelligent touchpoints so insurance feels invisible and effortless, yet highly trustworthy, especially at moments of service or claim. This approach preserves the value of brokers and advisors, enhancing their roles as strategic risk partners instead of replacing them. Finally, leadership, not technology, is at the heart of transformation. The ability to articulate a clear vision and quickly demonstrate value is what distinguishes the winners. Real-time governance, compliance by design, and empathetic human engagement are becoming essential to build—and keep—customer trust. The challenge for every executive now is not just to optimize yesterday's operations but to actively build tomorrow's intelligence layer. The frontier is being defined now, and it begins with a leadership mindset ready for structural redesign and velocity. BEST MOMENTS "Automation is not autonomy, efficiency is not intelligence, and digital channels without orchestration create digital fragmentation."  "European regulation is our unfair advantage. It's not just about discipline, it's about infrastructure."  "You have to evolve—from transaction intermediary into a strategic risk advisor, augmented by intelligence that handles routine so you can focus on relationships, empathy, and judgment."  "Governance is about to become the most strategic capability. When compliance agents and financial AI are embedded in every workflow, governance shifts from retrospective reporting to real-time intervention."  "The frontier firm is not defined by how much AI it deploys; it is defined by how intelligently it integrates risk, compliance, capital, and customer experience." — Sabine VanderLinden ABOUT THE GUEST Karl Grandl is often dubbed an “insurance dinosaur,” with over 30 years in the industry spanning Swiss Life, GetSafe, WeFox, and now Miss Moneypenny Technologies. His experience spans product development, distribution, and embedded insurance, as well as scaling tech-driven aggregators across markets.  At Miss Moneypenny, Karl is spearheading the integration of wallet technology and intelligence layers, focusing on frictionless customer interaction and embedding trust and compliance by design.  An advocate for regulation as a strategic advantage and transformation as a leadership imperative, Karl is a sought-after voice for both legacy insurers and challenger MGAs looking to build tomorrow's intelligence-driven operating models.  Connect with him via LinkedIn or at upcoming events such as InsurTech Week and InsurTech Insights in London. ABOUT THE HOST Sabine VanderLinden is a corporate strategist turned entrepreneur and the CEO of Alchemy Crew Ventures. She leads venture-client labs that help Fortune 500 companies adopt and scale cutting-edge technologies from global tech ventures. A builder of accelerators, investor, and co-editor of the bestseller The INSURTECH Book, Sabine is known for asking the uncomfortable questions—about AI governance, risk, and trust. On Scouting for Growth, she decodes how real growth happens—where capital, collaboration, and courage meet. If this episode sparked your thinking, follow Sabine VanderLinden on LinkedIn, Twitter, and Instagram for more insights. And if you're interested in sponsoring the podcast, reach out to the team at hello@alchemycrew.ventures