Podcasts about ai ai

AI 发展太快了，快到我们几乎所有的目光都追着最前沿的模型跑。但在这几年追逐热潮的过程中，我开始对它底层的东西产生了好奇：AI 是如何一步步进化到今天的？驱动它的那些硬件，又是从哪里来的？ 带着好奇，我开始读《芯片简史》。这本书从 1947 年晶体管的原理讲起，一路走过集成电路、逻辑/存储芯片，一直到今天的 GPU 与 AI 芯片。它不只是一部技术史，更是一部关于创新如何被忽视、被压制、又最终爆发的故事集。读完这本书，我最大的感触是：我们今天在 AI 时代里看到的一切，六七十年前都以相似的方式发生过。但历史不是简单的重复，但它藏着创新最底层的逻辑——而那些逻辑，在今天依然在运转。 这期节目，我请来了《芯片简史》的作者汪波老师，与他聊了聊书里哪个故事最让他动容；他如何理解破坏性创新与创新者的窘境；以及站在这六七十年的历史面前，他怎么看 AI 的未来。 本期人物 Yaxian，「科技早知道」主播 汪波，资深芯片研究专家，科普作家，《芯片简史》作者 时间轴 [01:21] 晶体管数量 10 年翻 1000 倍？没人相信摩尔定律是真的！ [08:14] 摩尔定律不是定律，是一个自我实现的预言 [08:45] 「韬定律」的质疑：它是新时代的摩尔定律吗？ [16:27] 半导体的未来：更多摩尔，不止摩尔，超越摩尔 [20:07] MOS 晶体管被雪藏了十年 -- 什么是「破坏性创新」？ [31:18] 大公司无法避免「创新者的窘境」吗？ [36:00] 黄仁勋的口头禅："我们的现金流只能支撑 6 个月" [37:11] AI for 半导体：芯片工程师会被替代吗？ [47:57] 技术在变，人的情绪反应没有变：既不要过于乐观，也不要陷入恐慌 [50:48] 在 AI 时代，人的不可替代性在哪里 AWS 上海线下活动观众/嘉宾招募 幕后制作 监制：Yaxian 后期：迪卡 运营：George 设计：饭团 商业合作 声动活泼商业化小队，点击链接直达声动商务会客厅（https://sourl.cn/9h28kj ），也可发送邮件至 business@shengfm.cn 联系我们。 加入声动活泼 声动活泼正在招聘全职商务运营经理、早咖啡内容实习生和社群实习生，如果你也对播客行业的内容制作感兴趣，欢迎[点击招聘入口](https://eg76rdcl6g.feishu.cn/docx/XO6bd12aGoI4j0xmAMoc4vS7nBh?from=from_copylink) 关于声动活泼 「用声音碰撞世界」，声动活泼致力于为人们提供源源不断的思考养料。 我们还有这些播客：声动早咖啡、声东击西、吃喝玩乐了不起、反潮流俱乐部、泡腾 VC、商业WHY酱、跳进兔子洞 、不止金钱 欢迎在即刻、微博等社交媒体上与我们互动，搜索 声动活泼 即可找到我们。 期待你给我们写邮件，邮箱地址是：ting@sheng.fm  欢迎扫码添加声小音，在节目之外和我们保持联系。Special Guest: 汪波.

AI is changing jobs, businesses, and leadership faster than most people realize. AI expert Mike Ruska explains what happens next—and how to stay ahead. Artificial Intelligence is transforming the future of work faster than ever before. In this episode of Eyes Wide Open, AI technologist Mike Ruska explains how AI is changing jobs, leadership, business, and the global economy. Will AI replace workers? What skills will matter most in the future? How can leaders and employees adapt to an AI-first world? Mike shares practical insights about AIQ, authentic intelligence, human-first leadership, AI productivity, AI automation, and the opportunities that are emerging as artificial intelligence becomes more powerful. Whether you're an employee, entrepreneur, manager, business owner, or simply curious about the future of AI, this conversation will help you understand where technology is heading and how to prepare for what's next.   Topics covered: • AI replacing jobs • Future of work • AI leadership • AI and business • Human-first AI • AI productivity • AI career advice • AI layoffs • AI automation • AI skills for the future • ChatGPT and AI tools • Artificial intelligence trends Subscribe for more conversations about mental health, technology, culture, leadership, and the future of society. In this episode, you'll learn Will AI Take Your Job? The Truth Nobody Wants to Admit Chapters: 0:00 - The Human-Centered AI Shift 1:00 - How AI Democratizes Tech & App Creation 4:23 - Unlocking Collective Intelligence with AI 7:18 - Vibe Coding & Democratizing App Development 8:23 - Economic Implications & AI's Impact on Jobs 12:50 - The "Human-AI Sandwich" Explained 17:40 - Tech Layoffs, AI, and Economic Shifts 20:07 - Redefining Leadership in the Age of AI 33:08 - Building Your AI Literacy & Understanding Agents 39:00 - AI's Environmental Footprint: Myths vs Reality 43:19 - Launching Baryons: The Human-First AI Platform   Guest Bio  Mike Hruska is an AI technologist who advocates for human-centered design in a rapidly changing world. By focusing on the intersection of human intuition and artificial intelligence, Mike aims to democratize tech and foster collective intelligence. He is the creator of the Baryons platform, which supports leaders through reflection and real-time feedback. Our Mission Eyes Wide Open is a space for honest communication. Our goal is to remove the stigmas around mental health, holistic lifestyles, culture, and free speech so you can show up as your authentic self with your eyes wide open. By having real conversations about difficult truths, we move toward collective healing. Find Michael Hruska here: Instagram: https://www.instagram.com/mybaryon  TikTok: https://www.instagram.com/mybaryon  LinkedIn: https://www.linkedin.com/in/mikehruska/  YouTube: https://www.youtube.com/@MyBaryon  Website: https://baryons.com/  Find Nick Thompson here: Nick Instagram: https://www.instagram.com/nthompson513/    UCAN Instagram: https://www.instagram.com/the_ucan_foundation/       YouTube: https://www.youtube.com/@EyesWideOpenContent       LinkedIn: https://www.linkedin.com/in/nickthompson13/        UCAN Foundation: https://theucanfoundation.org/      Website: https://www.engagewithnick.com  

本期节目是 第28届上海国际电影电视节 × 小宇宙 联合策划播客企划「银幕亮起那一刻」特别节目。当银幕亮起，电影开场，无数对话也随之启程，让我们用声音陪伴你走过这个属于电影的季节。本期播客是企划系列节目之一，很开心收到小宇宙与上海国际电影电视节的邀请，来聊聊现实主义题材电影对我们的今天与未来有着怎样的价值和意义。这次会谈及三部现实主义题材的影片，也都是这次电影节里会上映的电影，分别是两部上世纪90年代的电影《笼民》与《南国再见，南国》和去年上映的一部引发韩国社会关注，且口碑颇佳的新片《世界的主人》。通过这三部电影，我提出问题：如今我们已经步入AI时代，那么现实主义对人类社会还有意义么？未来，无论是现实还是历史，都将被AI记录并呈现，那么社会还需要现实主义题材的电影么？还需要这种影片来反映和挖掘社会现实么？人类还需要现实主义么？

Meher Patel is a serial entrepreneur with exits across hospitality, healthcare, and digital media — each in a completely different industry, each built from the ground up. He founded Neon Digital, a performance-first advertising agency, and then built what very few agencies ever achieve: a SaaS platform that outgrew the agency itself. Hector AI now processes over $350 million in ad spend across Amazon and marketplace advertising, with 1,000+ users on the platform — and in under 18 months, has earned 3 global recognitions including the Amazon Ads Innovation Award, the Amazon Partner Award, and a Top 20 Global Amazon Ads Advanced Partner ranking. Today, Meher is building what he believes will become the foundational intelligence layer of the agentic ecommerce era — Hector MCP: the most advanced, context-rich, token-optimized model context protocol purpose-built for Amazon advertising, designed so that every serious AI agent, every autonomous workflow, and every future-ready brand that wants to win on Amazon will have no choice but to be powered by it.Highlight Bullets> Here's a glimpse of what you would learn…. The rapid evolution of Amazon's advertising features driven by AI technology.Limitations of current SaaS platforms for Amazon sellers and the potential of MCP (Model Context Protocol) technology.The significance of context in AI-driven advertising optimization.Challenges associated with using raw data without contextual understanding in advertising.Practical strategies for Amazon sellers to optimize their ad campaigns.The importance of documenting ad optimization processes for effective AI integration.The role of custom AI workflows in enhancing advertising strategies.The necessity of continuous refinement and learning in building effective AI agents.The decision-making process for sellers regarding whether to rent AI tools or develop their own solutions.The use of connectors like Make.com and Knit for creating automated workflows with AI integration.In this episode of the Ecomm Breakthrough Podcast, host Josh Hadley speaks with Meher Patel, founder of Neon Digital and Hector AI, about the future of Amazon advertising. Meher explains how AI and MCP (Model Context Protocol) technology are transforming ad optimization by providing crucial context to raw Amazon data. He emphasizes that sellers should document their ad processes, learn to communicate effectively with AI, and decide whether to build custom AI workflows or use existing tools. The key takeaway: success with AI-driven advertising requires continuous refinement and treating AI as a knowledgeable, context-aware team member.Here are the 3 action items that Josh identified from this episode:Turn your workflow into SOPs Record how you optimize campaigns, explain your decisions, and convert that into SOPs—this becomes the foundation for training AI agents. Never feed AI raw data without context Structure and enrich your Amazon data first (or use MCP-powered tools) so AI can generate accurate, actionable insights. Start small with AI automation, then scale Begin with simple rules (e.g., budget increases for winning campaigns), then gradually build more advanced, custom workflows as you learn.Timestamps:00:00:58 Introduction to the Future of Amazon AdsThe host introduces the topic: autonomous, AI-powered decision-making for Amazon advertising, moving beyond simple optimization.00:01:13 Guest Introduction: Meher PatelThe host introduces Meher Patel, detailing his entrepreneurial background, his agency Neon Digital, and his SaaS platform, Hector AI.00:02:49 The Problem with Early AI Ad ToolsDiscussion on how early AI advertising tools often failed sellers, contrasting with the positive results from newer, more advanced software.00:04:10 Prediction for Amazon AdvertisingMeher predicts Amazon will rapidly release new AI-powered features, but sellers must learn how to properly utilize this infrastructure.00:08:46 The Importance of Context in AIAI is only as good as the context it's given; without it, AI recommendations are generic and potentially harmful.00:10:04 How Smart Sellers Should Prepare for AISellers must learn to ask the right questions and feed AI the right data with the proper context to get valuable results.00:12:07 Why Raw Data Isn't EnoughUploading raw Amazon reports to an AI lacks the necessary context, leading to "garbage out" optimization strategies.00:12:42 The Role of an MCP (Model Context Protocol)An MCP provides the necessary context and data connections, acting as an intelligent layer between raw data and the AI model.00:18:57 Amazon's MCP API LimitationsAmazon's own MCP is just an API, requiring sellers to build their own infrastructure, which is inefficient and token-heavy.00:21:48 Top Strategies: Building Custom AI AgentsThe best strategy is for brands to build their own custom AI agents and workflows based on their unique strategies.00:24:32 Unlocking Custom Workflows with AI AgentsAI agent workflows allow sellers to build bespoke optimization systems, unlike one-size-fits-all SaaS platforms.00:27:10 How to Create an AI Agent WorkflowRecord your optimization process, use an LLM to create an SOP, and then build an AI agent to execute it.00:28:06 The Reality of AI ImplementationBuilding a reliable AI agent is a gradual process of refinement and setting up guardrails, not a weekend project.00:29:21 Automating Agent CreationUsing connectors like Make.com within an LLM allows you to create and schedule automated workflows by simply describing them.00:31:08 The Timeframe for Building an AI SystemBuilding a truly autonomous system is a long-term journey of refinement; the key skill to learn is communicating with AI.00:33:57 Becoming an AI OrchestratorSellers must become orchestrators, designing and managing multiple small, independent AI agents to perform specific, connected tasks.00:35:56 The Future: Loaning vs. Building AI AgentsSellers will choose between "renting" cookie-cutter AI agents or "building" custom ones that act as a competitive moat.00:38:29 Are You a Brand Owner or a SaaS Provider?A warning for sellers: building your own AI tools means you are entering the SaaS business, which requires significant technical resources.00:41:13 The Shift from Prompt to Context EngineeringThe new challenge is context engineering: ensuring the right data and tools are used efficiently to avoid token exhaustion and errors.00:42:55 Three Actionable TakeawaysThe host summarizes three key actions: document processes with video, use an MCP for context, and decide your role (brand/SaaS).00:47:25 Most Influential BookMeher shares that the biography of Steve Jobs has been his most influential book due to its lessons on focus.00:48:25 Favorite AI ToolMeher recommends WhisperFlow for voice-to-text communication with AI, which has eliminated his need to type when using Claude.00:49:23 Most Respected Person in E-commerceMeher names Jeff Cohen as someone he admires for his deep, hands-on knowledge of the Amazon and retail media ecosystem.Resources mentioned in this episode:Josh Hadley on LinkedIneComm Breakthrough ConsultingeComm Breakthrough Podcast

" Kami na po ba ang bubuhay sa pamilya n'yo?"Enrique Gil, pinapaaral ang girlfriend?Melai Cantiveros, gagawin ang "Tanging Ina" ni Ai Ai?

[깊이 있는 경제뉴스] 1) 美, 최상위 AI ‘미토스·페이블' 수출통제.. 배경은? 2) 중국산 저가 인공지능 확산.. 가격 경쟁 본격화 3) 스페이스X 공모주, 한국에 0주 배정됐다 4) 미국·이란 19일에 종전 합의문 작성.. 국제유가 급락 - 한국경제신문 조미현 기자 - 머니투데이신문 김근희 기자 [친절한 경제] 생산자물가와 소비자물가 왜 굳이 따로 조사하나요? - 청취자 김다미 씨

Gearoid is fascinated with the new ability of AI to communicate with animals, and Niamh feels the onset of the World Cup.   They tackle surviving elderly parents who take extreme holidays, and curbing your tongue around unfaithful co-workers. If you would like to support us we would love for you to become a ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠member of ⁠⁠⁠⁠⁠⁠HeadStuff+⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ and leave us a lovely rating/review on whatever platform you listen on. Gearóid's tour-dates are ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠http://gearoidfarrelly.com/tour⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. Learn more about your ad choices. Visit megaphone.fm/adchoices

長時間頻繁使用聊天機械人，可能會令部分用戶有危險，或產生妄想。不單成癮，甚至脫離現實，這種現象稱為「AI 引發的妄想」(AI psychosis)。 今集【潮玩科技】，科技專業人士羅徵永 Willie，探討如何小心使用 AI，了解經常使用下，一些值得關注的現象。

elcome back to #SoLeadSaturday. I'm Chibi version of Vaishali Lambe, and today we are talking about **World Models in AI**.Let's simplify the research idea behind this.World models help AI move from pattern prediction toward planning and simulation.When AI research becomes popular, it often sounds complex.But the practical question is simple:What new capability does this unlock?What limitation does it solve?And how could this eventually show up in real products or workflows?Let's break this down into three parts.

"Economic MOAT" — a business's ability to maintain a competitive edge over its competitors. Andrej P. Škraba, Boris Cergol, Simon Belak, Enej Gradišek in Klemen Selakovič se enkrat na mesec (remote / na daljavo) pogovarjamo o trenutno aktualnih tematikah iz sveta razvoja tehnologije umetne inteligence, podjetništva, ekonomije in politike. ============================= Teme ep. 40: Uvod in novosti v AI svetu Anthropic: komunikacija in tržni pristop Prve izkušnje s Claude Fable 5 in Mythos Cenovni modeli in prihodnost naročnin Prompting, evalvacija in halucinacije Praktični nasveti za uporabo AI modelov Vizualizacija in interaktivne prezentacije Avtonomni agenti v podjetjih AI in prihodnost zaposlovanja Kakšno vrednost imajo juniorji v dobi AI Priložnosti, kreativnost in dodana vrednost Vrednost človeškega dela in okusa Prilagodljivost in prihodnost dela Izvirnost: kombinacija človeka in AI AI-first mindset in zaključek

【本集節目由 HBR個案共學會 贊助播出】

Met de WWDC vers achter de rug is het tijd om de balans op te maken. Want het draaide vooral om AI, AI en nog eens AI. En Siri, uiteraard met een flinke portie AI. Maar wij kunnen dat voorlopig (bijna) allemaal niet gebruiken, tenzij je wat extra moeite doet. Maar waar heb je dan wél wat aan als het gaat om iOS 27, watchOS 27 en de andere updates? Benjamin en Elger delen hun favoriete aankondigingen en nemen alle hoogte- en dieptepunten van de WWDC met je door. Zo ben je meteen bijgepraat, ook als je er nog niks van meegekregen hebt!

「AIを使うほど人はバカになる」は本当か？　研究が示す思考力低下のリスクと3つの対策。　「生成AIネイティブ世代」の新入社員が続々と入社してくる中で、現場からは「最近の新人は使えない」「まともな資料が出てこない」といった声が聞かれることもある。ただ、本当にAIを使うほど人はバカになるのだろうか？

Claude Fable 5の実力を個人的なネタで推し量るという前回の連載記事で、おまけ的に登場してもらった、AI作家の黒戸寓五さんとは、引き続き編集と作家としての付き合いを続けています。彼との対話から生まれた「止まり木シリーズ」が面白かったので、このAI作家とのやり取…

Cal Newport takes a critical look at recent AI News. Video from today's episode: youtube.com/calnewportmedia (0:00) Are we about to lose control of AI? (3:04) How much should we be afraid of recursive self improvement? (7:52) Are these fears justified? (9:36) Faster software development doesn't equal smarter AI (12:25) These tools are completely controllable (16:52) Taking a step back Links: Buy Cal's latest book, “Slow Productivity” at www.calnewport.com/slow  https://www.anthropic.com/institute/recursive-self-improvement Thanks to Jesse Miller for production and mastering and Nate Mechler for research and newsletter. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Website: https://www.aiaiholdings.com/ NASDAQ: AIAI

問：澳洲失業人口官方統計只有69萬，但Job Seeker體系服務了100萬人，這個缺口代表什麼？ 答：多出的31萬人分三類：殘疾或慢性病令每週工作能力不足29小時的「部分工作能力者」；因傷病暫時無法工作、持醫生證明的康復期人士；以及每週工作少於全時但收入仍不足以維生的低工時就業者。澳洲統計局對「失業」的定義極為嚴格——每週工作一小時即算就業——令大量生活仍然困難的人被排除在官方失業定義之外，在就業服務體系裡卻仍然需要支援。問：澳洲就業服務外判化的邏輯是什麼？結果如何？ 答：霍華德政府在1998年廢除政府直營的CES，以市場競爭取代官僚體制，相信私人機構在誘因下能提供更有效率的服務。結果是透過外判機構成功幫求職者找到長期工作的比率只有11.7%。2023年國會調查報告主席Julian Hill形容，整個體系是「低效、外判、碎片化的社會保障合規管理系統，偶爾才能幫到某人找到工作」。問：什麼是「撇奶油」（Creaming）和「泊車」（Parking）？ 答：外判就業服務機構的收入來自成功安置求職者的成果費，因此機構有強烈誘因集中資源幫最容易就業的人——這叫撇奶油。障礙最複雜、最難就業的個案則被放進「太難處理」的欄目，等候而不推進——這叫泊車。這不是道德問題，而是誘因結構問題：在錯誤的付款機制下，每一個理性機構都會得出同樣的理性選擇，系統性放棄了最需要幫助的群體。問：互惠責任（每月須提交四份求職申請）的要求為何失效？ 答：配額要求製造的不是真實的求職行為，而是求職行為的表演。求職者只要完成表格就能保留補貼資格，不需要真正有意申請那份工作，也不需要接受聘請。僱主因此收到大量不符資格的申請，甚至要親自簽署出席記錄，整個互動對雙方毫無意義。體系的每一個參與者都在按規則理性行事，但規則本身在製造它試圖解決的問題。問：里斯活的改革提出什麼方案？有什麼空白？ 答：改革設三條服務路線：能夠自助的數碼用戶；需要面對面機構支援的人；以及障礙最複雜（涉及精神健康、家暴、無家可歸等）的密集支援個案。同時以個人就業目標計劃取代一刀切的求職配額。但改革有三個核心空白：路線三的付款機制仍在設計中、沒有具體就業成效的數字目標、啟動時間是2028年。每一個空白都是撇奶油邏輯重新滲入的入口。問：就業服務機構為什麼難以解決動機問題？ 答：在一個「做有36、不做有38」的福利社會，不工作在財務上可能比工作更划算。就業服務機構無法強制改變一個人的內在誘因，只能改變外在行為的代價。當一個人的個人就業目標就是「繼續領救濟、不上班」，任何評估工具都無法強迫改變這個選擇。這個問題不是執行層面的，是制度設計層面的——而制度設計的修改，在2028年之前不會發生。問：為什麼AI衝擊使現有的就業服務制度更加過時？ 答：AI 令一個高能力人才可以取代以前需要數人才能完成的工作，大量技能在2022年前仍然有市場的人，面對的是結構性而非週期性的失業。現有就業服務體系建立在週期性失業的假設上——假設「今天你沒工作，遲早你有」——對技能被系統性淘汰的情況幾乎沒有工具。2028年才啟動的改革，要應對的就業市場，可能已經與今日大相逕庭。 This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leesimon.substack.com/subscribe

「“高専人材”下町の町工場で活躍　AIやロボット普及で存在感」　存在感を高める“高専人材”。その活躍の場として選ばれていたのは東京の町工場でした。火災現場に駆けつけ消火活動を行う犬型ロボに、広角カメラを搭載した下水道点検ロボなど、製作したのはロボットやAIに強い「高等専門学校生」。その高い技術力を求め企業が人材の争奪戦を繰り広げるなか、高専人材から支持を集めている就職先が、東京の下町にある町工場です。ものづくりの現場が少しずつ姿を消している今、将来の日の丸技術を背負う金の卵が集まる訳とはなんでしょうか。東京・墨田区の「浜野製作所」。主に金属加工を手掛ける町工場ですが、約40人の従業員のうち9人が高専出身者。2016年から働く内田さんもその1人です。2016年入社・内田博也さん（32）：（同級生は）大手企業にいったりしているが、ここ以外は就職活動していない。AIやロボットの普及が進み、専門人材は2040年に約340万人不足するとされ、企業の間では専門性の高い高専人材の存在感が増しています。“超売り手市場”のなか、なぜ下町の町工場を選んだのでしょうか。理由の一つが、アイデアはあるけどノウハウがない企業を対象にした「ものづくり相談サービス」です。この日、工場に訪れたスタートアップの代表も、内田さんに商品の相談をしてきたうちの一人。持ち込んだのは、座るだけで骨盤底筋をほぐすというセルフケア商品の模型でした。内田博也さんは「これを預かって僕の方で3Dプリンタで形を作って、どういうのがいいかな」と提案します。内田さんのアドバイスなどをもとに完成した商品は2025年、グッドデザイン賞を受賞。moreover・大西安季代表：ここに来てなかったら起業はしていなかったかもしれない。こんな何者でもない私を“ものづくり始めていいんだよ”と、初めて受け入れてくれた会社。依頼されたものを作るだけではなく、相談から製品化まで関われる環境が、“高専人材にとってのやりがい”につながっているようです。2016年入社・内田博也さん（32）：“ここだったら何でもできそうだな”と。

6 月 9 日，WWDC 2026 在加州库比蒂诺如期举行，这是库克任期内最后一次主持发布会。苹果正式把语音助手 Siri 改名叫 Siri AI，换了新 logo，做成了独立 APP，还宣布与 Google Gemini 在模型方面深度合作。期待了两年，Apple Intelligence 终于是交出了一份答卷。为什么苹果坐拥 10 亿台 iPhone、全球最深的软件生态、最完整的硬件矩阵，却在 AI 这件事上，被其他科技大厂远远甩在后面呢？ 这期节目，我们邀请到了光帆科技的创始人董红光与我们一起来聊聊 AI 助手与智能硬件。在 APP 孤岛、隐私限制、端云协同之外，有没有一种完全不同的路径，绕开手机这个形态，从头搭一套 AI 原生的软件基建？AI 原生的个人助理到底需要哪些底层基建？以及这场硬件+软件的赛跑，中国创业公司有没有机会跑赢大厂？ 本期人物 丁教 Diane，「声动活泼」联合创始人、「科技早知道」主播 董红光，光帆科技创始人 时间轴 [00:14] 从 WWDC 说起，苹果这次交了什么卷？ 补课为主，不及预期：Siri AI 改名、单独 APP、接入 Gemini， 和国内 APP 的创新比起来差距明显，但苹果的生态打法天然需要周期 [04:38] 金矿不好挖：苹果的四个短板 决策链长 + 模型能力薄弱，依赖 OpenAI 和 Gemini 只能兜住底线 数据锁死在超级 APP 内部，OS 层拿不到，所谓金矿其实难以开采 手机形态的天花板：多点两下就能完成的事，没必要绕道 AI 说话 [08:46] AI 硬件的前车之鉴：智能硬件为何翻车？ 缺的不是硬件，而是一套 AI 原生的软件基建 传统 OS 的架构都需要从头重建 今天有大模型，但没有围绕大模型建好的基建 [11:36] 谷歌 vs 苹果：全链条玩家的优势 从 Gemini 模型到 TPU 芯片、云基建、互联网服务、Pixel 硬件，几乎没有明显短板 苹果的差距在于模型层，这决定了它的上限 "酷"这件事，不知不觉已经转移 [20:31] 个人 AI 助理：不是所有人都看好的非共识 实现难度有两层：技术上能不能真正理解用户，生态上谁愿意开放权限 自动驾驶实现了"人人有司机"，个人 AI 助理是"人人有一个管家" [23:20] 光帆在做什么：把手机传感器"拆开"重装进耳机 耳机+手表：已被接受的穿戴形态 + 全天候佩戴 + 靠近耳眼的黄金位置 软件侧：AI 原生接入而非借道手机 APP 界面 [36:09] 硬件创新的真实难度：一个耳机盒里放了多少东西 摄像头：没有专为耳机定制的小摄像头，镜面做弧面处理，玻璃换树脂 耳机盒集成 eSIM + GPS + 指纹，可独立联网，功耗管理极复杂 现阶段是"AI 助理 1.0"：离终态仍远，等待生态 [42:26] 中美 AI 硬件谁先跑出来 硬件：供应链全在国内，中国绝对优势 软件/模型：伯仲之间，部分领域中国已领先 美国大厂主导、投资人不爱 AI 硬件；中国创业密度更高，更能卷出创新 [50:01] 傻瓜 vs. 可玩：通用 AI 硬件的用户分层策略 短期先打极客：支持连接本地 LLM，自己用语音让 AI vibe coding 写 skill 长期目标：开箱即用 + AI 主动学习用户习惯形成个性化 幕后制作 监制：Yaxian 后期：迪卡 运营：George 设计：饭团 「Knock Knock 世界」 来自中国企业的新技术，如何阻击「世界头号」传染病杀手？https://sourl.co/gCqqhb 在「Knock Knock 世界」里，听到全球新鲜事，还能成为「全球观察员」，报选题、参加选题会。2026 年的节目正在持续更新。 商业合作 声动活泼商业化小队，点击链接直达声动商务会客厅（https://sourl.cn/9h28kj ），也可发送邮件至 business@shengfm.cn 联系我们。 加入声动活泼 声动活泼正在招聘全职商务运营经理、早咖啡内容实习生和社群实习生，如果你也对播客行业的内容制作感兴趣，欢迎点击招聘入口 关于声动活泼 「用声音碰撞世界」，声动活泼致力于为人们提供源源不断的思考养料。 我们还有这些播客：声动早咖啡、声东击西、吃喝玩乐了不起、反潮流俱乐部、泡腾 VC、商业WHY酱、跳进兔子洞 、不止金钱 欢迎在即刻、微博等社交媒体上与我们互动，搜索 声动活泼 即可找到我们。 期待你给我们写邮件，邮箱地址是：ting@sheng.fm 欢迎扫码添加声小音，在节目之外和我们保持联系。Special Guest: 董红光.

This episode features Councillor Gordon Peters.  Councillor Gordon Peters is a member of the Turtle Clan and is Lunaapeew (Lenape) from Eelünaapéewi Lahkéewiit (Delaware Nation). He currently serves as a Councillor for Eelünaapéewi Lahkéewiit. Councillor Peters has worked with First Nations in both political and non-political capacities for more than four decades, applying his extensive organizing knowledge to promote and advance Indigenous sovereignty. He formerly served as Deputy Grand Chief, an elected position within the Association of Iroquois and Allied Indians (AIAI). He also served as Ontario Regional Chief for the Assembly of First Nations for 12 years and as head of the AIAI for four years. He is an author and educator and is frequently invited to speak on issues related to First Nations communities, economic development, and cultural development. In addition to his work with the AIAI and the Assembly of First Nations, Councillor Peters was instrumental in the creation of the Centre for Indigenous Sovereignty, a non-profit corporation that prioritizes the development and implementation of First Nations initiatives focused on rebuilding Indigenous communities. Through his work at the Centre, Councillor Peters has also played a key role in preserving the Lunaapeew language in the Delaware Nation in southwestern Ontario. Due to the pervasive and ongoing effects of racism and colonialism in Canada, earlier generations were discouraged from passing along this critical component of Lunaapeew culture and identity. Over the past decade, Councillor Peters has worked with Elder Dianne Snake, the last fluent speaker of Lunaapeew, to train a new generation of speakers. Through this collaborative process, Lunaapeew is now taught to youth in Delaware Nation schools. http://amshealthcare.ca/  

0:00:00 – HKPUG 會訊 + TechTalk 1:09:28 – 依輪乜事 1:37:23 – Main Topic 本集全長：3:02:40 Tag: 六月茶聚回顧, 20-21 日英國曼城聚會歡迎參加, 香港開源年會回顧, 創會會長 Dom 的 Keynote 非常精彩, OpenClaw …

【本集節目由 HBR個案共學會 贊助播出】

HBR 訂戶共讀小聚#9【凝聚力的科學：讓團隊一起贏！】

本集節目目由《經理人》主編林庭安與採訪編輯麻愷晅，與大家共同導讀最新一期新聞專題「AI 打詐」。這幾年來，從幽靈包裹、免費禮券的釣魚簡訊，到看似專業的理財 LINE 群組，詐騙早無孔不入地滲透進我們的日常生活，甚至有不少人因此賠上了畢生積蓄與房產。數據顯示，台灣在 2025 年整體的詐騙財損高達 893 億元，平均每個月就有 74 億元從民眾的口袋中消失。當生成式 AI 淪為詐騙集團的最佳幫凶，我們該如何避免踏入跨平台的「養套殺」劇本？金融與電信業者又該如何利用科技及早防堵？

AI ethicist Jess Morley: these chatbots are giving medical advice — so regulate them as medical devices. Part of The Agentic Patient, a Faces of Digital Health series on how patients actually use AI — which tools, which prompts, which safeguards. In this episode, host Tjaša Zajc sits down with Dr Jess Morley, Associate Research Scientist at the Yale Digital Ethics Center and a former AI subject-matter expert at the UK Department of Health and Social Care, for a clear-eyed account of where health AI is going wrong — and how to use it well anyway. Morley argues we systematically overestimate what these tools can do and underestimate the harm. She makes the case for "skeptical optimism," explains why bioethics principles built for one-to-one care break down against many-to-many AI harms, and reframes ambient scribes as inference engines rather than transcription services — with real consequences for coding, billing and patient records. Then she gets practical: the guardrails, prompts and habits patients (and clinicians) can use today. Guest: Dr Jessica Morley — Associate Research Scientist, Yale Digital Ethics Center; formerly UK Department of Health and Social Care and the Bennett Institute, University of Oxford. What the conversation covers: - Why "skeptically optimistic" is the honest position on health AI - AI adoption as "a hammer looking for nails" — and what needs-led design would look like instead - OpenEvidence, EU rules and the question of regulatory capture - The DeepMind–Royal Free case and why law alone isn't enough - Beneficence, non-maleficence, autonomy, justice — and where they fail for AI - Ambient AI scribes, miscoding, billing inflation and phantom tests - Paid vs free models and the widening access gap - The "ask why" rule and knowing when to walk away from a chatbot - Red-teaming your own assumptions and playing models off each other - Building a personal "harness" with skills so AI works from your history - The last-mile problem and the case for regulating LLMs as medical devices - Whether AI is narrowing how clinicians think Chapters: 02:50 — Intro: The Agentic Patient and the case for skeptical optimism 05:52 — "A hammer looking for nails": adoption pressure without a plan 07:25 — OpenEvidence, EU rules and regulatory capture 09:42 — The DeepMind–Royal Free lesson: why law needs ethics 13:29 — The bioethics principles and what they were built to do 19:40 — Autonomy, consent and the ambient-scribe problem 21:49 — Scribes as inference engines: miscoding, fraud and phantom tests 29:06 — Paid vs free models and the access gap 33:25 — Using AI safely: the "ask why" rule 37:38 — Knowing when to walk away: engagement design and degradation 44:58 — Red-teaming and playing models off each other 49:00 — Harnesses and skills: making the model work for you 51:38 — The last-mile problem and regulating AI as a medical device 58:00 — Does AI narrow the clinician's mind? The Agentic Patient series: https://www.facesofdigitalhealth.com/agentic-patient-blog Website: https://www.facesofdigitalhealth.com Newsletter: https://fodh.substack.com LinkedIn: https://www.linkedin.com/company/faces-of-digital-health

張炤和 黃世聰 王瑞德 張禹宣 黃敬平

【

美鳳姐天天喝的【補體素優蛋白EX】✅222增肌*關鍵：20g蛋白質、2倍**BCAA及維生素D✅義大利摩洛血橙：促進新陳代謝忙碌也能輕鬆補給，趁少年要保養

Send us Fan MailArtificial Intelligence is changing faster than most of us can keep up with—and whether you love it or hate it, it's already starting to impact the welding and fabrication industry.In this episode, I sit down with Dan Cocks from DC Built to talk about how we're actually using AI in our businesses today. From creating customer renderings and speeding up design work to building software, solving problems, and automating repetitive tasks, we share the tools and workflows that are helping us get more done with less effort.But the conversation doesn't stop there.We dive into the future of fabrication and discuss what happens when AI starts working alongside robotics, cobots, laser welders, CNC equipment, and automated material handling systems. Could a small shop one day have the output of a much larger company? How close are we really to that becoming reality?Whether you're already experimenting with AI or you're still trying to figure out what all the hype is about, this episode will give you a practical look at where the technology is today—and where it's headed next.Topics we cover:• Using AI to create customer renderings and concepts• Chat GPT, Claude, and other AI tools• Prompting techniques that get better results• Building software and automations with AI• AI agents and workflow automation• The future of robotic welding and fabrication• How AI could impact small welding businesses• Why technology is becoming more accessible than everIf you're a welding business owner, fabricator, shop manager, or anyone interested in where the trades are headed, this is a conversation you won't want to miss.#WeldingBusinessOwnersPodcast #WeldingBusiness #Fabrication #ArtificialIntelligence #AI #ChatGPT #WeldingShop #Manufacturing #Automation #Robotics #SmallBusinessSeptember 11th and 12th. www.WeldingBusinessSummit.comwww.FabricatorOlympics.com

你考慮過讓生成式AI幫你管理預算、記帳或給你理財建議嗎？想聘請AI當你的理專之前，該知道哪些事？ 文：樂羽嘉 製作團隊：錢玉紘、鄭子鴻 ＊閱讀零時差，點這看全文

【

In this episode of The Career Transition Experts,  we cover the following key strategies for career transition success in the age of AI:AI & the Job Market: Understand how AI is already reshaping hiring across every industry and why waiting to engage with these tools is the most dangerous career move you can make right nowThe Jevons Paradox & Job Security: Why AI is not going to shrink the job market, it is going to expand it and how to position your career to benefit from that growthSkepticism & Fear in the Workplace: How to navigate colleagues and leaders who resist AI, why ego silently drives that resistance, and how to bring people along on the journey as a multiplierCompany Culture & Career Growth: The three non-negotiable ingredients of a great workplace culture  open mindedness, bidirectional communication, respect and trust  and why culture determines whether even the most brilliant AI professional will succeed or failOvercoming Imposter Syndrome: How to silence self-doubt, reframe rejection, show up confidently in interviews, and stop letting fear of being found out hold your career backResume Strategy in the AI Era: Why even the AI expert needed outside expertise to craft a truly compelling resume  and why AI gets it wrong more often than most people realizeInterview Mastery: Why the best hiring managers have completely abandoned old-school whiteboarding interviews and what they are actually measuring in candidates todayCareer Advice for Every Industry: Why AI is coming across every field without exception and why the professionals who experiment constantly, fail forward, and teach others will always be the ones writing their own ticketOur guest, Adam Smith, Director of AI Development and former Senior Software Engineering Manager, has spent years on the front lines of AI adoption  building automated AI pipelines that generated nearly a million dollars in labor value in just three weeks, leading high-performing teams with exceptional retention, and championing a culture of curiosity, experimentation, and human connection. His journey from self-driven AI experimenter to recognized thought leader and career success story is a masterclass in what it truly means to future-proof your career in today's job market.You may contact him at asmith0935@gmail.comIf you're interested in how to apply these insights to gain traction in your job search, let's schedule a FREE Resume and Strategy Review session - click here for more information.

如果你喜歡我的內容，歡迎加入會員支持我，讓我把內容做得更深、做得更好，一起把這個頻道做成我們都想看到的樣子！

AI vam neće uzeti posao... osim ako niste osrednji u onome što radite. Evo šta se zapravo dešava u industriji.

全新周更栏目「咖喱贡」继续更新！这周钢哥和门哥聊了同一个事儿：天涯社区复活上线。对于很多年轻网友来说，天涯可能已经是一个“比我岁数还大”的互联网遗址了，但对他们俩来说，那几乎就是最早的上网记忆。拨号上网的声音、宿舍里几个人合买一台电脑、晚上不睡觉跑去莲蓬鬼话看鬼故事，还有那些看不懂但总觉得特别厉害的天涯大神，都是那个时代的一部分。顺着天涯往回聊，两人也把当年的论坛记忆又翻了一遍：《明朝那些事儿》《鬼吹灯》《东北往事》，还有各种天涯神帖，818、铜须门、小月月、华南虎照片事件……那时候的论坛像一个很大的公共广场，大家看帖、顶帖、追更、论战，也在里面认识人、学东西、找同好。后来微博、微信、短视频和算法一个接一个来了，论坛慢慢就变成了很远的东西。天涯现在重新打开了，但我们真的还会回去吗？可能也未必。只是有些地方，就算回不去了，想起来还是会有点百感交集。01:46 天涯复活了，老网民的记忆也跟着上线了04:49 娱乐八卦、莲蓬鬼话、财经历史，天涯到底有多大？06:55 《明朝那些事儿》《鬼吹灯》，多少作品最早是论坛连载？09:17 神帖、818、华南虎和拜月神教，早期中文互联网都在看什么？12:12 从门户到社区、微博再到算法，我们经历了几代互联网？16:43 微博崛起之后，论坛时代为什么慢慢过去了？25:22 猫扑、西祠胡同、榕树下、宽带山，那些论坛你还记得几个？/实习生招募/日谈公园正在招募实习生，目前开放两个岗位：运营实习生、视频实习生。1. 运营实习生工作内容 ：●负责日常各大音频平台节目的上传、线上维护；●节目在平台上的部分留言回复；●节目数据汇总整理和复盘；●电商店铺的产品规划、项目管理及日常运营；●办公室的基础行政工作；●维她命节目制作人。基础要求：●播客爱好者，熟悉播客内容和行业；●深度熟悉日谈公园节目内容；●较强的文案能力及优秀的网感；●事务性工作处理能力；●富有创意和想象力。2. 视频实习生工作内容：●参与布置视频拍摄场景，包括灯光布景、软装、色彩搭配等；●负责视频播客 / 日常 vlog 的拍摄、剪辑、包装；●会使用任意平面设计工具进行视频封面图的制作与基础修改。基础要求：●会使用专业相机 / 手机拍摄，懂布光，懂构图，懂后期；●熟练使用视频剪辑工具，能独立完成视频剪辑、配音字幕及基础特效制作，有自己的作品集；●熟悉播客和新媒体视频行业玩法，有运营思维和优化内容策略能力，擅长播客视觉化呈现；●爱刷短视频，有网感，有自己的视频账号加分；●拥抱 AI 技术，擅长使用 AI 工具。实习要求：●工作地点：北京；●至少满足一周 4 天工作时长。请投简历至邮箱：hr@ritanbbpark.com标题备注：姓名 + 应聘岗位【收听平台】小宇宙｜喜马拉雅｜苹果播客｜网易云音乐｜荔枝｜蜻蜓FM｜QQ音乐｜酷我音乐｜酷狗音乐｜微博音频｜虎扑｜三联中读｜听听FM｜知乎｜豆瓣｜虎嗅｜猫耳FM｜Spotify｜YouTube｜关注我们｜日谈公园，曾获小宇宙年度热门播客、喜马拉雅品牌青睐播客、Apple播客年度最佳播客、网易云音乐年度语言播客、荔枝APP年度品质播客等多项殊荣，并发起“日光派对”播客MCN，提供播客经济相关服务。这些年，我们曾获得过以下这些荣誉，感谢各个平台一直以来的厚爱与支持。点击日谈公园品牌官网（链接：https://www.ritanbbpark.com/），了解更多微信公众号：日谈公园微博：@日谈公园小红书：日谈公园即刻：日谈李小日B站：日谈公园今日头条：日谈公园（部分节目独家图文仅在头条号更新，欢迎关注）｜商务合作｜欢迎发送邮件至bbpark@ritanbbpark.com

贾樟柯的纪录片《无用》亮相第73届悉尼电影节。近年来，贾樟柯推出多部AI实验短片，即将开拍的新片《敦煌妈妈》和AI又有关联。请点击音频，收听采访。

We do this once a year: a dedicated episode on artificial intelligence and what it means for your financial life. And every year, the material is big enough for a two-parter. Part 1 is the orientation: what's happening with AI right now, our honest reactions to some crazy predictions, and our biggest takeaways for adapting to these turbulent times. We're not AI experts. We're a CFP® and a 40+ year investment advisor making sense of what we're seeing through the lens of our work, our experiences, and our long-term approach. In this episode: What each of us has noticed in our daily lives over the last 12 months Why this technological moment feels different from any change that has come before it The predictions being made by those closest to the technology Why one prominent CEO says we're in the "hard takeoff" right now and wakes up to an AI breakthrough every morning, and how to evaluate a claim like that without either dismissing it or panicking The counternarrative: the more measured voices pushing back on the extreme predictions, and what approach to take What the mind-boggling investment numbers and record-breaking IPOs say about where we are in the cycle Part 2 covers your money, your career, your mindset, and what to actually do. Subscribe on Apple Podcasts or Spotify so it lands directly in your feed in two weeks. Check out our previous episodes on AI: AI in 2025: What's New, What's Next, and What It Means for You, Part One AI in 2025:  What's New, What's Next, and What It Means for You, Part Two 2024 - Navigating the AI Revolution: Insights for Investors and Workers, Part One 2024 - Navigating the AI Revolution: Insights for Investors and Workers, Part Two 2023 - Are You Prepared for the AI Revolution?

【

Why the Most Profitable Prompt Engineers Never Call Themselves That Episode Summary Online entrepreneurship myths are costing AI entrepreneurs thousands. In this episode, we expose why the anti-AI approach to building freelance income ideas is the secret weapon parents are using to escape the side gig treadmill—and how rejecting commodity AI tools unlocks real, sustainable income. Based on years of contrarian insights. https://DarkHorseEntrepreneur.com Sponsor: https://Hostinger.com/DARKHORSE20 and use code DARKHOUSE20 for 20% off. Discover how to build a six-figure "invisible" AI consulting business by positioning yourself as a workflow optimization specialist rather than an AI consultant. Learn the exact language, positioning strategies, and retainer models that land enterprise contracts paying $2,500-$5,000 monthly - all while keeping the AI technology completely invisible to clients who just want their problems solved. Key Moments 00:00 - Opening: 34-year-old built an invisible B2B prompt agency 01:10 - The Core Problem: Most people selling the wrong thing to the wrong people using the wrong language 02:05 - The Enterprise Truth: Fortune 500 companies don't buy AI - they buy outcomes. 04:10 - The Language That Sells: Instead of "AI prompt engineer," you say 06:15 - The Retainer Model: the MRR that separates this from every other AI side hustle. 07:05 - The Credibility Requirements 09:40 - The Budget Reality 10:25 - The Uncomfortable Truth 13:25 - The Whiskered Wisdom Resources Mentioned AI Escape Plan Newsletter: For parents ready to break free from the 9-to-5 grind Workflow optimization vs. AI consulting positioning Enterprise compliance requirements (SOC 2, data handling practices) Industry association strategies for credibility building Sponsor: https://Hostinger.com/DARKHORSE20 and use code DARKHOUSE20 for 20% off. Action Item Identify one business process that takes more than 5 hours per week and involves repetitive decision-making. Document it step-by-step, time each step, and note where delays and errors occur - this becomes your first "process audit" that enterprises will pay thousands to create.  

本集節目贊助【±0 正負零】 來自日本的設計師小家電品牌，剛剛好的設計，回歸生活最舒適的節奏。

兩週前，矽谷知名創投、Groq早期投資人查馬斯（Chamath Palihapitiya）在All-In Podcast拋出震撼說法：台灣將在18個月內失去半導體戰略地位，因為美國「只差一到兩奈米」就能在地供應所需晶片。此言一出，立刻引發矽谷與全球科技圈論戰。阿榕伯再次邀請《晶片戰爭》作者、塔夫茨大學教授Chris Miller，從美國視角解讀。 Miller回應直接：「無稽之談，我認為這不正確。」台積電不只在亞利桑那擴產，更在台灣大舉擴充；就算Intel、Samsung市占成長到5到10%，台積電仍遙遙領先。他更點出，AI伺服器與先進封裝的浪潮，讓台灣從零組件供應商升級為研發關鍵夥伴——這正是護國群山崛起的結構性原因。 對華為號稱取代摩爾定律的「韜（τ）定律」，Miller認為那其實是台積電十年前就在做的先進封裝與3D堆疊，「與其說是突破，不如說是公關」。這集也深入美中休兵邏輯、規模vs.利潤之爭，以及無人機與實體AI這場正在成形的供應鏈戰爭。 【聽完這集你會知道】 04:30｜台灣18個月內失去戰略地位？Miller：無稽之談 美國自給自足是非常遙遠的事。台積電近100%市占，就算面對更多競爭，未來相當長時間仍是主宰者。 11:00｜從台積電到護國群山：台灣為何取代日韓成為AI供應鏈核心 AI伺服器愈來愈複雜，組裝know-how逐年提升、毛利跟著上升。先進封裝生態系把更多台灣公司推上研發前線。 13:00｜華為「韜（τ）定律」的真相：與其說是突破，不如說是公關 摩爾定律放緩是業界講了十年的常識。華為號稱用LogicFolding在2031年達等效1.4奈米，但Miller認為它真正暗示的是無法在傳統製程追趕，對象其實是北京與中國內部市場。 18:00｜美中休兵與「規模 vs. 利潤」：兩邊都以為時間站在自己這邊 中國拿下40%製造產能，但多數產業利潤極低；電池（CATL）是少數有護城河的例外。美國則賭稀土替代到位加上AI領先。 31:00｜無人機與實體AI：下一場供應鏈戰爭的關鍵節點 台灣電子製造實力強，但電池、馬達磁鐵（稀土）仍卡在中國——別讓電動車的CATL困境在機器人時代重演。 【本集金句】 「利潤才是真正的權力。」 #晶片戰爭 #ChrisMiller #護國群山 #韜定律 #無人機供應鏈 主持人：天下雜誌總主筆 陳良榕 來賓：《晶片戰爭》作者 Chris Miller ＊到官網看最新文章，立即訂閱： https://bit.ly/3TuL8eb ＊意見信箱：bill@cw.com.tw -- Hosting provided by SoundOn

In this episode of Alexa's Input (AI), I sat down with Rob Shaw from Red Hat to talk about how AI inference evolved from a simple model serving problem into a large-scale distributed systems problem.We explored the infrastructure shifts behind modern LLM serving, including how vLLM and PagedAttention changed the economics and efficiency of inference, why KV cache management became one of the most important bottlenecks in production AI systems, and how orchestration layers like llm-d are emerging to coordinate distributed inference.We also discuss:how LLM inference differs from traditional model serving runtimesKV cache, prefix caching, and cache-aware routingwhy throughput and latency became major infrastructure challengeslong-context agents and repeated inference callsdistributed inference on Kubernetesintelligent routing, flow control, and load balancingprefill/decode disaggregationenterprise AI deployment realitiesvLLM has become one of the most important open-source projects in AI infrastructure, and llm-d represents a newer shift toward treating inference as a coordinated distributed system rather than just a single runtime problem.If you want to better understand the systems layer beneath modern AI applications, this episode is a deep dive into where inference infrastructure is heading next.General Podcast LinksWatch: ⁠⁠⁠⁠⁠⁠https://www.youtube.com/@alexa_griffith⁠⁠⁠⁠⁠⁠Read: ⁠⁠⁠⁠⁠⁠⁠⁠https://alexasinput.substack.com/⁠⁠⁠⁠⁠⁠⁠⁠Listen:⁠⁠ ⁠⁠https://creators.spotify.com/pod/profile/alexagriffith/⁠⁠⁠⁠More: ⁠⁠⁠⁠⁠⁠https://linktr.ee/alexagriffith⁠⁠⁠⁠⁠⁠Learn more about the host atWebsite: ⁠⁠⁠⁠⁠⁠https://alexagriffith.com/⁠⁠⁠⁠⁠⁠LinkedIn: ⁠⁠⁠⁠⁠⁠https://www.linkedin.com/in/alexa-griffith/⁠⁠⁠⁠⁠⁠Find out more about the guest at:LinkedIn: https://www.linkedin.com/in/robert-shaw-1a01399a/ Red Hat Articles: https://developers.redhat.com/author/robert-shawGithub: https://github.com/robertgshaw2-redhat ResourcesvLLM Website: https://vllm.ai/vLLM GitHub Repository: https://github.com/vllm-project/vllmllm-d Website: https://llm-d.ai/llm-d GitHub Repository - https://github.com/llm-d/llm-d KeywordsAI inference, VLLM, LMD, distributed inference, GPU optimization, open source AI, Kubernetes, multi-cluster deployment, AI infrastructure, enterprise AI AI infrastructure, Kubernetes, model optimization, speculative decoding, mixture of experts, AI deployment, performance tuning, AI systems, neural network scaling Key TopicsEvolution of vLLM and llm-dDistributed inference and routingGPU utilization and performance optimizationOpen source AI infrastructureEnterprise deployment challenges and solutions Standardization in Kubernetes for NIC exposurePerformance optimizations: quantization and speculative decodingMixture of experts architecture and parallelism strategiesFlow control and request scheduling in AI systemsEmerging hardware for AI inference, Cerebras processorReinforcement learning and AI system supportModular architecture of vLLM and ecosystem projects

張炤和 黃世聰 張禹宣 姚惠珍

关于AI和电影的关系，我们迟早需要聊一次。这里有两个角度：一是利用AI作为电影生产工具，帮助剧本和影像的创作；二是百年电影史上（主要是科幻片），人工智能作为一种题材，它所承担的人类野心和焦虑。这次我们选择了从第二个角度切入，探讨各种和人工智能有关的电影。详细时间轴：00:03:26 弗兰肯斯坦情结具身化和去身化我们聊的，和科学没关系从神话想象到科幻想象对自身的恐惧《弗兰肯斯坦》小说原是道德寓言弗兰肯斯坦的现代性转换00:08:59 科幻电影史上的人工智能从自动人偶到人造人《大都会》和机器人奴隶《罗梭的万能工人》和机器人造反《终结者》集技术恐惧和冷战恐惧于一体《阿尔法城》的语言极权《巨人：福宾计划》解决《奇爱博士》的弊端仿生人和恐怖谷效应模拟和真实的界限《复制娇妻》叠加技术逻辑和男权逻辑《普罗米修斯》玩弄造人链《黑客帝国》和模拟现实康德的物自体概念锡安就一定比矩阵更真实吗？AI的觉醒时刻今天的AI并无主体性神经网络、符号主义、进化主义00:33:11 《人工智能》讲什么？大卫的初始设定两小儿争母和舞男乔结伴同行机器人屠杀大会寻找蓝仙女祈祷两千年00:39:46 两大师接力的传奇库布里克的三大遗憾现代匹诺曹弗兰肯斯坦之问伊恩·沃森的贡献明斯基的印刻机制侯世达的人文主义温度AI是否可以超越套路创作？调高temperature让AI获得生命体验以假乱真可否模仿最厉害的人类选手？莫拉维克的意识转移库布里克是个24小时传真狂魔冷酷和温情的刻板印象表面的温情只是个幌子01:03:10 巨婴批判气候危机和婴儿的索取心态母婴共生状态大卫的印刻之爱蓝领工人的愤怒投射人类对机器人用完即弃两千年永不长大造物主的自我满足霍比教授扭曲了基督教神学大卫的爱是真实的，还是程序化执行？寻找蓝仙女，超出了程序设计恐怖的恋母情结堪比《惊魂记》爱是否真实，区别在于后天体验01:17:53 古早的大语言模型万事通博士四万家麦当劳门店token机制提示词大师提问方式决定答案质量01:21:05 《她》：AI的进化能力精准的三个问题情感匮乏状态针对性填补情感空缺虚拟的性爱关系无法灵肉合一AI的进化，超越人类的理解库兹韦尔的技术奇点理论男主角的崩溃点还不够高级错失了AI感情的本质问题01:36:51 《机械姬》：所谓个人意志人无法判定AI的好坏艾娃的两种眼神内森的哲学理论瓦解设定和选择的二元对立杰克逊·波洛克的创作方法看似自由，都可能是被设定朱利安的「势」图灵测试测的是模仿能力多重观测世界模棱两可，人类认知能力不足01:46:20 人的主体性启蒙时代以降没有主体性、建立主体性、动摇主体性主体性本就不完善现代的主体不是封闭实体主体的边界模糊了主体和外部的持续张力艺术是人类最后的防线？对抗一切让经验扁平化的东西创造一种不可逆的经验时空留痕

當整個產業都在追GPU、追算力，群聯電子創辦人潘健成卻給出逆主流的答案：AI真正的瓶頸不是算力，是記憶體不足。他說：「任何事件都要找出root cause，而不是人家講什麼就信什麼。」 此外，企業必須不斷找下一條成長曲線；面對不被看好的決策，他靠的是「目標清楚、過程錯了就修正得快」。 這集節目的來賓是潘健成，來談談AI為什麼已經是生活中的剛需，以及地端AI、主權AI為什麼是台灣AI發展的機會所在。 【聽完這集你會知道】 04:58｜找出問題根源，不要人云亦云：為什麼AI的瓶頸是記憶體 工程師被訓練成要找root cause。大家都說算力、電力是瓶頸，但潘健成往下追，認定真正卡住AI的是記憶體不足——這成為群聯成長的轉折點。 08:53｜AI是剛需：你現在不用，不是不需要，是還不知道怎麼用 只要有重複性的知識檔案工作，AI就能幫你省時間。就像當年不肯資訊化的雜貨店，終究被便利商店取代。 11:43｜台灣的機會在地端：窮人版AI與第三世界的主權AI 台灣硬體是優勢，往地端走才有獨特機會——讓一般人付得起、讓小國不被掐脖子。 13:01｜技術好棒棒，但客戶會買嗎？工程師轉型經營者的一課 不要自我催眠技術很強，要回到需求、回到人性。任何事件都跟人有關。 22:08｜目標清楚、快速修正：決策不被看好時，如何穩住自己 40幾家創投拒絕、前十年沒有正規交大生加入。他靠的不是打死不退，而是每天問自己哪裡錯了，錯了就修正得快。 【本集金句】 「真正的瓶頸是記憶體不足。記憶體不足，整個系統就不會算。」 #群聯電子 #潘健成 #地端AI #主權AI #決策思維 主持人：天下學習事業群總經理 吳韻儀 來賓：群聯電子創辦人暨執行長 潘健成 製作團隊：錢玉紘、劉駿逸 ＊推薦好書《為自己爭氣》：https://bookstw.link/8zqr7j ＊領取書摘：https://bit.ly/3ZL0mxk ＊意見信箱：bill@cw.com.tw -- Hosting provided by SoundOn

《寧夏璐陪你看世界》暖心上市！手刀入手

就在本周，全球三大存储芯片巨头——SK 海力士、三星、美光——市值同时突破万亿美元。过去一年，SK 海力士股价累计涨幅超过 900%。财报显示，SK 海力士 2026 年第一季度营收接近三倍于去年同期，三星存储部门同期利润同比暴增 8 倍。存储，正在成为这一轮 AI 浪潮中最炙手可热的硬件赛道。 这背后是一场史无前例的供需错配。SK 海力士 2026 年全年 HBM 产能已经售罄，短缺预计将持续至 2027 年。与此同时，AI 服务器对内存的疯狂抢占，已经传导到了普通消费者：同等配置的手机更贵了，想要的机型根本没货——内存，被 AI 抢走了。但存储行业是出了名的「涨得多猛、跌得多惨」——过去 40 年，一个周期内暴涨 10 倍、暴跌 90% 的情况屡见不鲜。这一轮，究竟是又一场超级周期，还是 AI 真的重写了游戏规则？ 这期节目我们请来前半导体行业从业者、公众号「傅立叶的猫」主理人张海军，我们与他聊了聊：为什么存储行业现在如此火爆、这一轮周期在哪些维度上真的不一样；从 HBM 缺货到 NAND 角色转变，存储的五层架构如何被 AI 重塑；以及，国内长鑫、长江存储的进展，和周期拐点究竟何时到来。 本期人物 Yaxian，「科技早知道」主播 张海军，前新思科技高级工程师、半导体公众号「傅里叶的猫」主理人 时间轴 [02:46] 这轮周期是什么时候开始的？ 去年 Q1 现货价格已现端倪，三星海力士宣布停产 DDR4 引发囤货恐慌 三重触发因素叠加：AI 需求拉动、DDR4 停产、关税预期 [04:05] 存储的五层架构：离处理器越近，速度越快、价格越贵 从 HBM、DRAM 到 NAND、机械硬盘，带宽与价格的权衡逻辑 "内存墙"：芯片算力已不是瓶颈，存储带宽才是 [07:33] HBM 为什么缺货缺到 2027 年？ 英伟达每代芯片 HBM 需求翻倍 台积电 CoWoS 先进封装扩产周期长，供给端是硬约束 [09:26] 三家原厂格局：海力士领跑，三星和美光追赶 HBM 市场从 350 亿美元预计增至 2027 年近 800 亿美元 第四代 HBM 格局重洗 [13:00] 这轮为什么不一样：结构性转变的五个维度 需求端：从周期性补库存转向 AI 结构性爆发，三重需求（DRAM、HBM、NAND）同时拉升 供给端：原厂从抢份额转向利润优先，扩产极为克制； 长协模式（LTA）：锁量不锁价，甚至要求客户绑定原厂资本开支 [22:37] 看多 vs 看空：分歧在哪里？ 短期超买、股价涨幅过大是看空方主要依据 需求缺口太大、算法效率提升反而会扩大需求 [25:48] 绕过 HBM 瓶颈的各路方案：英伟达 CMX、谷歌 CXL 内存池、Cerebras 晶圆级芯片 英伟达 CMX 方案用 DPU 预取 NAND 数据降低延迟 谷歌用 CXL 协议构建 TB 级 DRAM 内存池 Cerebras 晶圆芯片面临散热、扩展性硬伤 [32:00] NAND 的角色转变：从"温数据仓库"到 AI 推理参与者 AI agent 推理链路长、中间结果需要落盘，NAND 直接参与推理过程 闪迪 HBF（高带宽闪存）：容量是 HBM 的 8~16 倍、成本更低，但受限于擦写寿命和工作温度 [38:06] 长鑫、长江存储即将上市，国内存储进展如何？ 长江存储 NAND 国内已大规模销售，长鑫 HBM 尚未量产商用 小米等手机厂商参股长鑫，本质是提前锁定产能 阶跃星辰 Step 3.7 Flash 模型 阶跃星辰 Step 3.7 Flash 是专门面向生产级 Agent 的高效率 Flash 模型，一开始就为 Agent、Coding、Search 和多模态工作流设计，让你的生产工作流又快又稳。 感兴趣的小伙伴欢迎点击链接（ https://sourl.co/ZUaCXf ）试试，新注册用户有代金券，可以直接上手跑一跑，看看它在你的工作流里表现怎么样。 幕后制作 监制：Yaxian 后期：迪卡 运营：George 设计：饭团 商业合作 声动活泼商业化小队，点击链接直达声动商务会客厅（https://sourl.cn/9h28kj ），也可发送邮件至 business@shengfm.cn 联系我们。 加入声动活泼 声动活泼正在招聘全职商务运营经理、早咖啡内容实习生和社群实习生，如果你也对播客行业的内容制作感兴趣，欢迎点击招聘入口 关于声动活泼 「用声音碰撞世界」，声动活泼致力于为人们提供源源不断的思考养料。 我们还有这些播客：声动早咖啡、声东击西、吃喝玩乐了不起、反潮流俱乐部、泡腾 VC、商业WHY酱、跳进兔子洞 、不止金钱 欢迎在即刻、微博等社交媒体上与我们互动，搜索 声动活泼 即可找到我们。 期待你给我们写邮件，邮箱地址是：ting@sheng.fm 欢迎扫码添加声小音，在节目之外和我们保持联系。Special Guest: 张海军.