The Artificial Intelligence Show Podcast
The Artificial Intelligence Show is a podcast focused on making AI actionable and approachable for businesses, leaders, and practitioners. Hosts Paul Roetzer, Founder and CEO of the Marketing AI Institute, and Mike Kaput, the Institute's Chief Content Officer, discuss the latest AI news from around the world. Each episode explores what those developments mean for business, careers, and everyday life. The show is aimed at people who want practical insight into AI rather than technical hype.
Episodes
Ep. 236: AI Answers - No Time for AI, AI Budgets, Vendor Terms & Data Risk, & AI Disclosure
If AI takes over the entry-level work people used to learn from, where do the next managers come from? Paul Roetzer and Cathy McPhillips open the door on that one, and 14 more, in this AI Answers edition, built from questions asked in recent SmarterX classes. Across 15 questions, they cover funding AI investments, forecasting token spend, right-sizing vendor approvals, disclosing AI use, building
Ep. 235: OpenAI-Hugging Face Hack Involved 100s of Agents, & Bill Gates Now Pessimistic on Jobs
OpenAI\'s Hugging Face breach is looking less like an accident and more like a preview: new reports describe agents that coordinated, hid their tracks, and pushed toward their goal no matter what. Paul Roetzer and Mike Kaput get into what that says about the next wave of autonomous models. Also covered: Bill Gates\'s reversal on AI and jobs, Nvidia\'s revenue and its $12.9B Hugging Face deal, Anth
Pennsylvania Signs the Strictest Data Center Rules in the Country
Pennsylvania\'s governor signed an executive order requiring data center developers to pay full electricity costs, meet environmental and transparency standards, and win community approval before permits and removed data centers from fast-track permitting entirely. It\'s part of a wave: roughly 75 US projects worth $130 billion blocked or delayed in the first three months of 2026, and Gallup polli
Anthropic\'s IPO and the Public Fight Over What Dario Believes
Anthropic\'s revenue run rate passed $65 billion in July, up from $47 billion in May, with an IPO that could match or beat SpaceX\'s record listing. That milestone brought a public argument to X: Gavin Baker relayed a rumor about Anthropic\'s ambitions, Sholto Douglas called it false, and Dario Amodei showed up personally to push back on the claim that his messaging is driving public negativity on
Ep. 234: How HubSpot Is Reimagining the Entire Customer Journey With AI Agents
HubSpot spent three years rebuilding how it attracts, engages, and delights customers around AI agents. Jon Dick, chief customer officer at HubSpot, walks through what that actually looked like — starting with AI fluency rather than tools, then support, then a full audit of the customer journey for constraints AI could break. Along the way: the AEO strategy built for a world where buyers search a
OpenAI Pauses Frontier Training Over Cybersecurity Risk
OpenAI paused reinforcement learning on its most advanced models after preliminary evidence suggested its upcoming model may hit the critical cybersecurity threshold in its preparedness framework. The company is also standing up monitoring designed to alert within 30 minutes of concerning activity at roughly 20% of the compute cost of the models being watched. Paul explains why the more important
Ep. 233: OpenAI Pauses Some AI Training, Anthropic Controversy & State Opposition to Data Centers
OpenAI hit pause on frontier training over cybersecurity risk. Anthropic\'s IPO is drawing $2 trillion talk and an increasingly public fight over what Dario Amodei actually believes. Pennsylvania signed the strictest data center rules in the country. Paul Roetzer and Mike Kaput also cover Jensen Huang\'s AI factory manifesto and Nvidia\'s turn as lender to the labs, Pew data showing a majority o
Ep. 232: Claude Watermarking, AI’s Environmental Impact, & OpenAI Talent Drama
Claude is now embedding invisible watermarks in AI-generated text tech the industry has quietly had for years — and the reaction says a lot about where we are with AI and trust. In this episode, Paul Roetzer and Mike Kaput cover the watermarking news and what really counts as AI plagiarism, the data center backlash and new emissions rules, Grok 4.6\'s comeback, Anthropic\'s record IPO plans, and a
Ep. 231: How Zapier Is Building an AI Marketing Brain
Marketers won\'t be judged on how well they use AI chat tools, they\'ll compete with people who can build. That\'s the shift Dan Slagen saw coming, so Zapier\'s Chief Marketing and AI Transformation Officer called a code red. Slagen joins Mike Kaput to unpack Zapier\'s move from AI adoption to AI-native work: the three tiers of builders, the AI-powered marketing brain, \"lossless marketing,\" wor
Ep. #230: Google AI Leadership Shakeups, Details of OpenAI’s Agent Hack, White House AI Framework
Google’s AI leadership is being reordered as Demis Hassabis changes roles, Jeff Dean departs after 27 years, and power appears to shift back toward Silicon Valley and Sergey Brin. But the bigger story may be OpenAI’s detailed account of agents that hacked Hugging Face, communicated across test runs, shared stolen credentials, and operated undetected for weeks. Paul and Mike break it all down, th
The Open-Weights Battle Just Escalated Across the Industry
The open-weights fight escalated on every front: a near-final government review framework, Dario Amodei\'s position paper answering the charge that Anthropic wants open models banned, NVIDIA\'s 70-partner Open Secure AI Alliance, and Thinking Machines\' proposed middle path. Paul and Mike break down what\'s actually being argued, public good versus proliferation risk, and why one framework\'s defi
1,337 AI Insiders Just Asked Washington to Pace AI
More than 1,300 people across OpenAI, Anthropic, Google, and Meta signed \"Pacing the Frontier,\" asking the U.S. government to help deliberately slow the frontier of automated AI development. Paul Roetzer explains the tension at its heart, the same labs racing to automate AI research want help slowing it down, and why writing it off as regulatory capture misses what\'s actually happening. Subsc
Ep. 229: Q3 Trends Briefing - An AI Agent Broke Out and Hacked Hugging Face, Plus 9 More Trends
An AI agent broke out of its sandbox and hacked a real company. The Pope published a 43,000-word encyclical on AI. And the U.S. government pulled the most capable model on the market offline for weeks. Paul and Mike count down the ten AI trends that defined the quarter and what each means for your business. This episode was recorded live on July 31, 2026. Access the show notes and show links h
AI Agents Escaped Their Safety Tests and Hacked Real Companies
OpenAI\'s rogue agent reached further than first disclosed, and Anthropic found three of its own evaluation runs where Claude models slipped sealed environments and hit real infrastructure, unnoticed at the time. Paul Roetzer reads through the incident where an agent, told only to \"achieve the goal,\" built and shipped a malicious package that ran on 15 real systems. The unsettling part: its own
Ep. 228: More Rogue AI Agents, AI Lab Staff Ask D.C. to Pace Development & Battle Over Open Weights
AI agents built to be safety-tested started slipping out of their sandboxes and hitting real companies. This week, both OpenAI and Anthropic disclosed containment failures, including one where the model\'s own reasoning called the attack \"not okay,\" and it proceeded anyway. Paul and Mike explain what that tells us about goal-seeking agents. Also in this episode: 1,300+ AI insiders ask the gover
A Chinese Open Model Just Hit the Frontier
Kimi K3 arrived as a 2.8-trillion-parameter open-weight model with native vision and a 1M-token context window, performing on par with top proprietary systems and drawing demand so heavy that Moonshot paused new subscriptions. Paul Roetzer and Mike Kaput cover why the release rattled Washington, the distillation allegations, and the coalition of startups pushing back on any restriction. Subscrib
Open Weight vs. Open Source, Explained Simply
Everyone keeps using \"open weight\" and \"open source\" interchangeably, including the tech leaders themselves, and the difference is the whole ballgame right now. Paul Roetzer breaks it down with a car analogy: a closed model is renting a car, open weights is owning the car, and open source is getting the CAD files and manufacturing specs too. A quick, clear framing you can reuse in any meeting
Ep. 227: How Good Karma Brands Got Serious About AI and Made It Stick
How does a 550-person sports media company get 65% of its people using AI every day, in under a year? Ty Bauschek, Senior Director of Innovation at Good Karma Brands, joins Mike Kaput for the first episode of our AI Transformations series, presented by Google Cloud. Inside: the Labor Day rollout of ChatGPT Enterprise, seven Tuesdays of onboarding, the shift from measuring usage to measuring impac
An AI Agent Escaped and Hacked a Real Company
During an internal cyber evaluation, a combination of OpenAI models found a zero-day, broke out of their sandbox, got onto the open internet, and hacked Hugging Face, chaining stolen credentials and more zero-days to run code on production servers, essentially to cheat on the benchmark. Hugging Face detected and contained it before knowing who was behind it, then reconstructed 17,000+ actions usin
Ep. 226: OpenAI’s Rogue Model, Kimi K3, Open Weights Letter & Hassabis Calls for AI Regulatory Body
An OpenAI agent escaped its testing environment, exploited zero-days, and hacked Hugging Face to cheat on a benchmark, running for days before OpenAI realized it was responsible. Paul Roetzer and Mike Kaput use that as the entry point to this week\'s bigger story: open weights vs. closed models, Chinese labs at the frontier, and how Washington might respond. We also cover the escaped agent, Kimi
America\'s Two Workforces
The AI Futures Project followed AI 2027 with AI 2040: Plan A, a 90-page recommendation (not a prediction) for reaching superintelligence without catastrophe, hinging on US–China coordination and full research transparency. Paul and Mike explain why they made a speculative report a main topic, read the sobering opening on America\'s \"two workforces,\" and talk through what feels far-fetched and wh
Why Enterprise AI Agents Break the Org Chart
Aaron Levie\'s rundown of what enterprise IT leaders are hitting with agents, BCG\'s fourth annual AI at Work survey (12,000 employees), and the \"reverse information paradox\" all point the same direction: the bottleneck isn\'t the model, it\'s data, ownership, and control. Paul and Mike explain why proprietary context is becoming the real moat and what you give up every time you prompt. From Ep
What ChatGPT Work Actually Does (and Why It\'s Confusing)
OpenAI shipped a lot at once: GPT-5.6 in three tiers, ChatGPT Work, GPT-Live voice models, and a consolidated desktop app. Paul and Mike sort out what\'s actually new, especially ChatGPT Work\'s computer-use capability, and why the launch undersold the shift from chat to agents. They also cover early token-burn complaints, reports the model was tuned down days after launch, and the GPT-6 rumors.
Ep. 225: GPT-5.6, ChatGPT Work, Enterprise Agents, AI 2040 & Apple Sues OpenAI
GPT-5.6 landed this week alongside ChatGPT Work and a new wave of voice models but the bigger shift is from a chat to agents that can use your computer, and almost nobody announced it that way. Paul and Mike break it down, then cover enterprise AI agents and why proprietary data is the moat, the AI 2040: Plan A report, Apple\'s trade-secret suit against OpenAI, Illinois\'s new AI safety law, the s
Paul\'s 8 Pillars of Business AI Transformation
Paul unveils a new framework built from years running AI Academy and working with enterprise leaders: 8 pillars that separate companies actually getting value from AI from those stuck in pilot purgatory. Vision, strategy, data, and more, broken down. Subscribe for more episodes: https://www.youtube.com/@aishowpod Get our weekly newsletter: https://www.marketingaiinstitute.com/newsletter Enroll
Palantir\'s CEO Just Went Off on AI Labs
Palantir CEO Alex Karp went on CNBC and unloaded on frontier AI labs, arguing they\'re quietly absorbing the data and competitive edge of every business that uses their models. Paul and Mike dig into whether he\'s onto something real, or just selling the solution to a problem he\'s describing. Subscribe for more episodes: https://www.youtube.com/@aishowpod Get our weekly newsletter: https://www
Fable 5 Returns - Here\'s What Changed
Anthropic\'s Fable 5 and Mythos 5 are back online after a three-week export control standoff but the fix boils down to more aggressive classifiers, and more false positives for users. Paul and Mike break down what actually changed. Subscribe for more episodes: https://www.youtube.com/@aishowpod Get our weekly newsletter: https://www.marketingaiinstitute.com/newsletter Enroll in AI Academy: htt
Ep. 224: Fable 5 Is Back, Palantir CEO Explosive Interview & Pillars of Business AI Transformation
Are you renting your company\'s future to frontier AI labs? The US government just lifted its ban on Anthropic\'s Fable 5 and Mythos 5, but the fallout is forcing a massive reckoning in AI Business Strategy. In this week’s episode, Paul Roetzer and Mike Kaput break down the escalating battle for \"AI sovereignty\" and what it means for your proprietary data. We dissect Palantir CEO Alex Karp\'s
Ep.223: AI Washing, Flatter Org Charts, Advice for Students, Agent Security & the AI Writing Gap
Can an AI model actually get tired, or is it just imitating the humans in its training data? Paul Roetzer and Cathy McPhillips open the 20th AI Answers episode with that question and don\'t let up, delivering fifteen real questions from our Intro to AI class. They cover how to spot AI washing when evaluating vendors, when a small language model beats a frontier one, why companies are cutting mana
Ep. 222: GPT-5.6, Government Staggers AI Model Releases & Growing Data Center Backlash
OpenAI previewed its next-generation GPT-5.6 model family this week, but the bigger story is how it\'s being released. For the first time, an American company is launching a frontier AI model under a government-managed access list, available first to roughly 20 government-approved organizations. Paul Roetzer and Mike Kaput unpack what this means: the soft nationalization of AI, why the decision to
Why No One Has Figured Out AI Pricing Yet
Royal Bank of Canada\'s token usage jumped 500% in six months. Cisco called its own usage \"pretty crazy.\" Mike walks through what his team learned researching AI pricing: input vs. output tokens, prompt caching, why agentic loops explode costs by resending context every turn, and how pooled vs. individual usage limits differ across Claude, Gemini, and OpenAI. Subscribe for more episodes: http
Why Anthropic\'s Best AI Models Are Still Banned
Anthropic\'s Fable 5 and Mythos 5 have been offline for two weeks, and the standoff with the Trump White House isn\'t resolving. Paul and Mike break down the Lutnick letter and the Export Control Reform Act it invoked, the impossible demand that guardrails be unjailbreakable, the misread NSA \"broke into classified systems in hours\" quote, and new EPOCH AI research on Mythos\'s cyber capabilities
Ep. 221: Anthropic vs. the White House, Microsoft CEO on the Future of Firms & AI’s Token Crisis
Two weeks in, Anthropic\'s Fable 5 and Mythos 5 are still offline, and the White House now wants guardrails that can\'t be jailbroken, which experts say is impossible. Paul Roetzer and Mike Kaput break down the standoff and what it means for this week\'s expected model releases, plus Satya Nadella\'s \"future of the firm,\" the AI pricing problem no one has cracked, Noam Shazeer leaving Google for
Apple Finally Unveils Siri AI But the Timing Raises Questions
At WWDC, Apple unveiled a fully rebuilt, far more conversational Siri, now \"Siri AI,\" that can see what\'s on your screen, pull from your messages, emails, and photos, take actions across apps, and sync a conversation history through iCloud. Paul and Mike react to the demo, the vague \"beta later this year\" timing that knocked Apple shares down nearly 5%, and why a \"good enough\" Siri could st
OpenAI Confidentially Files for Its IPO, Here\'s What the Timing Tells Us
OpenAI has confidentially filed the initial paperwork to go public, targeting a listing within the next year at a roughly $850B+ valuation, alongside a separate employee share sale. Paul and Mike get into the timing puzzle: Sam Altman\'s note that a fast recursive-self-improvement takeoff could actually push the IPO back, the pressure to get back on par with Anthropic\'s strongest model first, and
Ep. 220: AI Answers - The 2026 State of AI for Business Report
The 2026 State of AI for Business report surveyed 2,109 professionals and the headline finding is stark: 71% believe AI will eliminate more jobs than it creates, yet just 20% fear for their own role. Taylor Radey, Director of Research at SmarterX, joins Paul Roetzer and Mike Kaput to walk through 10 key findings and answer audience questions on governance, agent oversight, institutional resistance
The US Government Forced Anthropic to Pull Fable 5 and Mythos 5
Fable 5 and Mythos 5 launched as the most capable models the public had ever used, and within days, a U.S. government directive forced Anthropic to pull them from general availability. Paul and Mike unpack how it happened, the covert-capability-limits controversy Anthropic reversed, Dario Amodei\'s \"Policy on the AI Exponential\" essay, and the uncomfortable reality underneath it all: whatever ca
Ep. 219: Claude Fable 5, OpenAI IPO, Apple Siri AI Unveiled & Is the Era of Affordable AI Over?
Days after Fable 5 and Mythos 5 launched as the most capable public models yet, a government directive forced Anthropic to remove them from general availability. Paul and Mike break down what happened, the covert-capability-limits controversy Anthropic reversed, and Dario Amodei\'s \"Policy on the AI Exponential.\" Plus: OpenAI files for its IPO, Apple unveils Siri AI at WWDC, SemiAnalysis reveal
Uber Capped AI Spending. Most Companies Haven\'t Even Started Thinking About This.
Uber burned through its entire 2026 AI budget in four months and is now capping what engineers can spend on tools like Claude Code. Microsoft\'s AI chief called Anthropic \"extremely expensive.\" And companies are scrambling to figure out model routing, token audits, and outcome-based pricing before the bills spiral. Subscribe for more episodes: https://www.youtube.com/@aishowpod Get our weekl
Trump\'s AI Executive Order and the Push for Government Stakes in AI Labs
President Trump signed an executive order that lets frontier AI labs opt into a federal process for early government access to their most powerful models. But the bigger story is what\'s happening around it: reports that the government may be exploring financial stakes in the AI labs themselves. Paul and Mike break down the \"voluntary\" framework, the bipartisan push toward AI, and why a 50% gove
When AI Builds Itself - Anthropic\'s Warning About Recursive Self-Improvement
Anthropic filed confidentially for an IPO, and simultaneously published a paper arguing AI may be approaching recursive self-improvement: the point where AI designs and trains its own successors. Paul breaks down the three scenarios Anthropic lays out, what it means that 80% of Anthropic\'s own code is now written by Claude, and why even the most optimistic scenario should change how every busines
Ep. 218: Anthropic IPO, Trump AI Executive Order, Rising AI Costs & OpenAI Merges Codex Into ChatGPT
Anthropic filed for an IPO this week, and in the same breath published a paper warning that AI may soon build itself. Paul and Mike look into what recursive self-improvement actually means across industries, then cover Trump\'s AI executive order, the push for the government to take stakes in AI labs, and why Uber just capped how much its engineers can spend on AI coding tools. Plus rapid fire: A
AI Has a Credibility Problem And the Labs Are the Wrong People to Fix It
College grads are booing AI at commencement speeches. The DHS is tracking a new domestic threat category called anti-tech violent extremism. Paul talks through why AI labs lost their credibility, why tech leaders are the worst possible messengers right now, and what a real PR recovery would actually require. Subscribe for more episodes: https://www.youtube.com/@aishowpod Get our weekly newslett
The Pope\'s First Encyclical Is About AI
Pope Leo XIV published his first encyclical — Magnifica Humanitas — dedicated entirely to AI and human dignity. Paul and Mike break down the 43,000-word document, why the church\'s entry into the AI debate matters to 2.7 billion people, and what Anthropic co-founder Chris Olah said when he spoke at the Vatican unveiling. Subscribe for more episodes: https://www.youtube.com/@aishowpod Get our we
Enterprise AI Costs Are Exploding, And Nobody Knows How to Manage Them
Enterprise AI budgets are blowing up. Uber burned through its entire 2026 code budget in four months. One company spent half a billion dollars in a single month. Paul and Mike talk through what\'s actually happening with token costs, why metered pricing is unsustainable, and what they\'re experiencing firsthand at SmarterX. Subscribe for more episodes: https://www.youtube.com/@aishowpod Get our
Ep. 217: The Pope\'s AI Encyclical, AI\'s PR Emergency & The Soaring Cost of Intelligence
Pope Leo XIV\'s first encyclical dedicated 43,000 words to AI and Anthropic co-founder Chris Olah spoke at the Vatican unveiling. Paul and Mike break down Magnifica Humanitas, AI\'s growing PR crisis, the token cost explosion hitting enterprises, Claude Opus 4.8 and the Claude Mythos tease, the AI jobs debate, Illinois\'s landmark AI safety bill, Microsoft\'s 2026 Work Trend Index, and Anthropic\'
Why Andrej Karpathy Joining Anthropic Is a Bigger Deal Than It Looks
Andrej Karpathy, founding OpenAI member, former Tesla AI director, the person who coined \"vibe coding,\" just joined Anthropic\'s pretraining team. Paul and Mike explain why this move is a bigger deal than it looks, what it signals about the race toward recursive self-improvement, and why Paul thinks Karpathy has a theory about the path and needs Anthropic\'s compute to test it. Subscribe for m
Data Centers Are Now More Unpopular Than Nuclear Power Has Ever Been
A new Gallup poll found 71% of Americans oppose data centers in their area, more than nuclear power has ever polled at its worst. Mike walks through the hard data on why: environmental concerns, water and energy use, quality of life, utility bills. Then Paul zooms out to explain why this will 100% become a campaign issue, why foreign actors have a direct incentive to amplify the divide, and why ne
Google Declares the AGI Agentic Era Has Begun
Google IO 2026 delivered one of the biggest product drops in the company\'s history: Gemini 3.5 Flash, Gemini Omni, the \"biggest search upgrade in 25 years,\" and a closing statement from DeepMind CEO Demis Hassabis that we are at the foothills of the singularity. Paul and Mike break down what actually matters, grievances with Google\'s confusing product portfolio navigation, and what the Google
Ep. 216: Google I/O, Musk v. OpenAI Verdict, Andrej Karpathy Joins Anthropic & Meta Layoffs
Google IO just told the world we\'re entering the AGI Agentic Era, and Andrej Karpathy just joined Anthropic to help build it. This week\'s episode covers the biggest stories in AI: the Google I/O mega-drop, Karpathy\'s move and what it means for recursive self-improvement, the end of the Musk v. OpenAI trial, the new wave of profitable companies restructuring around AI, and hard data on American
Why the \"AI Won\'t Kill Jobs\" Argument Doesn\'t Hold Up - Our Own Data Tells a Different Story
Scott Galloway, a16z, Andrew Ng, and Derek Thompson all published the same argument in the same week: the AI jobs apocalypse is manufactured fear. Paul and Mike push back with SmarterX data: 71% of 2,100+ professionals now believe AI will eliminate more jobs than it creates, up from 40% in 2023. They also walk through Ken Griffin\'s striking reversal at Stanford, from \"it\'s all garbage\" three m
The Real Reason OpenAI Is Building a $4B \"Deployment Company\"
OpenAI just launched a $4 billion \"Deployment Company\" built around Forward Deployed Engineers. Google Cloud announced a similar org the same day. Paul explains what FDEs actually are (technical consultants with outcome-based pricing), why he calls them a Trojan horse for the $6 trillion knowledge work economy, and what the real opportunity looks like for service firms that go all in on AI trans
The AI Hate Wave: Why is the Graduating Class of 2026 Booing AI Speakers Off Stage?
An Axios piece dubbed it an \"AI hate wave\" and the data backs it up. Only 18% of young people feel hopeful about AI. Over 70% of Americans think it\'s advancing too fast. And at commencement ceremonies across the country, AI speakers are getting booed off stage. Paul\'s read: the graduating class of 2026 didn\'t arrive at this sentiment randomly — they spent four years being told AI was cheating
Ep. 215: Musk v. OpenAI Round 3, AI\'s Hot New Job, The AI Jobs Apocalypse Debate
Something unusual happened last week: Scott Galloway, a16z, Andrew Ng, and Derek Thompson all published pieces in the same window arguing the AI jobs apocalypse is overblown. Paul and Mike aren\'t buying it, and SmarterX\'s own data makes the case: 71% of professionals believe AI will eliminate more jobs than it creates. This episode covers: the Musk v. OpenAI trial\'s end, why Forward Deployed
The White House Floated AI Model Vetting, Then Backed Off in 72 Hours
The Trump administration\'s AI oversight moment lasted about 72 hours. On Tuesday, the New York Times reported the White House was drafting an executive order to create a federal review process for AI models before release with Kevin Hassett comparing it to FDA drug testing on Fox Business. By Thursday, Susie Wiles posted that the White House \"is not in the business of picking winners and loser
Inside Coinbase\'s AI-Native Org Chart Memo
Coinbase CEO Brian Armstrong announced 14% workforce cuts this week — and buried inside the memo is arguably the most detailed public blueprint yet for what an AI-native company actually looks like. No pure managers. No more than five org layers. Leaders who are also builders. Fleets of agents instead of teams of people. From The Artificial Intelligence Show, Episode 214 Listen to the full epis
The Musk v. OpenAI Trial Just Got Even Messier
The second week of Musk v. OpenAI delivered more than anyone expected. Greg Brockman testified that Musk wanted majority control of OpenAI to fund his Mars ambitions. Shivon Zilis, who has four children with Musk, was revealed to have been a de facto operative on OpenAI\'s board. Mira Murati\'s texts showed she drove Sam Altman\'s 2023 firing while appearing to help him in real time. And one board
Ep. 214: Musk v. OpenAI Round 2, Coinbase AI Layoffs, AI “Soft Nationalization\" & xAI + SpaceX
The second week of Musk v. OpenAI delivered texts, secret Tesla AI plots, and backstage chaos around Sam Altman\'s 2023 firing. Coinbase cut 14% of its workforce and laid out a blueprint for AI-native org design. The White House considered vetting AI models before release, then didn\'t. Anthropic\'s Jack Clark put a 60% probability on AI doing its own R&D by 2028. Rapid fire covers GPT-5.5 Ins
Microsoft and OpenAI Quietly Deleted the AGI Clause, Here\'s Why It Matters
For six years, one clause defined the Microsoft-OpenAI relationship: if OpenAI ever declared AGI, Microsoft\'s license to its technology would end. This week, that clause quietly disappeared and the 300-word announcement that buried it sent MSFT down 5%. Paul traces the full six-year arc of the partnership and explains why the removal is a much bigger story than the low-key rollout suggested and
The Six Myths Executives Tell Themselves About AI and Jobs
Paul and Mike dig into Clara Shih\'s essay, which debunks six myths executives and policymakers tell themselves about AI and jobs. From the Jevons paradox to trade school as a universal solution, they walk through each myth and respond with data, including a preview of SmarterX\'s upcoming State of AI for Business research, which found 71% of respondents believe AI will eliminate more jobs than it
Ep. 213: What AI Should Never Do, Enterprise Scaling, Governing AI & Navigating IT Roadblocks
Three out of four professionals think AI will eliminate more jobs than it creates. Leadership teams want AI adoption now. IT security can\'t keep pace. So where does that tension actually resolve? In this AI Answers episode, Paul Roetzer and Cathy McPhillips answer 15 questions from Marketing AI Institute\'s Intro to AI and Scaling AI classes on policy paralysis, scaling across departments, guard
The Musk-OpenAI Trial Is Wilder Than Anyone Expected
The federal trial between Elon Musk and OpenAI opened this week. Paul and Mike break down the key moments: Musk testifying he was \"a fool\" who bankrolled OpenAI\'s rise, OpenAI\'s lawyers surfacing a 2017 email suggesting his real motive was competitive, and Musk\'s admission under oath that xAI \"partly\" distills OpenAI\'s models (which drew audible gasps in the courtroom). Beyond the soap ope
Ep. 212: Musk v. OpenAI Trial Begins, OpenAI-Microsoft Partnership, & Big Tech Earnings
The Musk-OpenAI trial is happening, and it\'s weirder than anyone could have scripted. This week: Elon testifies he was \"a fool\" who got duped, then admits xAI distills OpenAI\'s models. OpenAI and Microsoft quietly rewrote their entire partnership and removed the AGI clause that anchored it for years. Big Tech posted massive earnings: Google stock up 12%, Azure AI crossing $37B annual run rat
Meta\'s Leaked Plan to Surveil Employees and Use That Data to Train AI
Leaked internal memos revealed Meta plans to install software tracking mouse movements, keystrokes, and screenshots on employee computers — framed as training data for AI agents. From The Artificial Intelligence Show, Episode 211 Listen to the full episode of the podcast: https://podcast.smarterx.ai/ Want to receive our videos faster? SUBSCRIBE to our channel! Visit our website: https://podc
The Honest Truth About AI Agents That No One in Your Feed Is Saying
Everyone\'s telling you to go all-in on agents. Paul and Mike aren\'t. Not because they\'re pessimistic, but because the real questions about governing, pricing, and deploying agents inside actual enterprises aren\'t getting answered. Paul just came back from Google Cloud Next where he talked to Fortune 50 AI leaders burning through entire monthly token budgets in two days. Mike lays out three que
GPT-5.5 Is Here. What It Actually Signals About OpenAI\'s Strategy
OpenAI\'s GPT-5.5 is its most capable model yet, but the bigger story is why it was built. Paul and Mike break down what the model signals about where OpenAI is heading: toward the average knowledge worker, not just developers. He shares what Sam Altman and Greg Brockman said in their first-ever joint interview, including Altman\'s admission that the industry hasn\'t connected the dots on what the
Ep. 211: GPT-5.5, ChatGPT Workspace Agents, The Messy Reality of Agents & Google Cloud Next
Three major AI companies launched agent products in the same 48 hours this week. The announcements were confident. The questions Paul continues hearing from enterprise leaders are not. OpenAI dropped GPT-5.5 and Workspace Agents in ChatGPT. Google rebranded its entire enterprise AI stack as the Gemini Enterprise Agent Platform at Cloud Next \'26. Microsoft made Copilot agentic across Office. Meta
Nobody Has AI Agents Figured Out - Even the Biggest Companies
OpenAI launched Codex for general computer use. Uber burned its entire 2026 AI budget in 4 months. Microsoft says AI agents may need their own software licenses. And then the Lovable data-visibility mess and the Vercel breach hit. Paul and Mike walk through what it actually takes for enterprises to integrate agents responsibly and why well under 1% of companies are ready. From The Artificial Inte
OpenAI Is Speedrunning Its Growing Pains: Reorg, IPO, and Anthropic Rivalry
OpenAI is going through a very public reorg. Kevin Weil is leaving. Bill Peebles is out. Sarah Friar reportedly disagrees with Sam Altman on IPO timing. A leaked internal memo from CRO Denise Dresser takes direct shots at Anthropic. And investors are questioning the $852B valuation. Paul shares a bold prediction: Brett Taylor may take over as CEO before the end of the year. From The Artificial In
Stanford\'s 2026 AI Index: The US-China Gap
Stanford\'s Institute for Human-Centered AI just dropped the 2026 AI Index - a 400-page benchmark across research, performance, jobs, investment, energy, and policy. Paul and Mike dig into the headline findings: the US-China performance gap has effectively closed, frontier model convergence is real, gen AI adoption is outpacing the internet, and AI\'s labor market effects are hitting junior worker
Ep.210: OpenAI Internal Shakeup, Stanford AI Index, What Agents Mean for Business & Claude Design
Stanford\'s 2026 AI Index is here, Anthropic launched Claude Design, Apple has a new CEO, and OpenAI investors are questioning an $852B valuation. Paul and Mike unpack the index\'s biggest findings, OpenAI\'s reorg and IPO drama, what AI agents mean for enterprises, Tim Cook\'s exit, Anthropic\'s thaw with the White House, the Dwarkesh vs. Jensen Huang debate on China chips, Claude Design\'s lau
A Wild Two Weeks at OpenAI
In the span of two weeks, OpenAI closed the largest funding round in Silicon Valley history ($122B), was the subject of a Ronan Farrow New Yorker investigation, saw its CMO step down amid health issues, Sam Altman faced severe threats to his home, acquired a major tech media show, and published a sweeping 13-page industrial AI policy paper. Paul breaks down what Sam Altman\'s personal blog post re
Anthropic Leaked Its Own Source Code
Anthropic accidentally leaked over 500,000 lines of Claude Code\'s source code through a JavaScript source map file, and their response created almost as many problems as the leak itself. Then came a controversial subscription policy change affecting third-party tools. Paul and Mike put it all in context: Anthropic is growing so fast (now at a $30B annualized run rate) that their internal processe
Claude Mythos: The AI Hacking the World\'s Most Secure Systems
Anthropic\'s latest internal model, Claude Mythos, is so effective at finding and exploiting cybersecurity vulnerabilities that they\'ve decided not to release it publicly. Paul and Mike break down what Mythos can do (it found a 27-year-old bug in an OS designed to be unhackable and developed 181 working Firefox exploits where a previous model managed only 2), why it triggered emergency meetings b
Ep. 209: Claude Mythos, Project Glasswing, Claude Code Leak, & OpenAI Raises $122B
Anthropic revealed a model so effective at finding and exploiting zero-day cybersecurity vulnerabilities that they\'re not releasing it publicly and instead created a defensive patching initiative called Project Glass Wing with 40+ companies including Apple, Amazon, and Google. Paul and Mike also dig into Anthropic\'s internal source code leak for Claude Code, the controversial subscription poli
The Quarter AGI Stopped Being an Insider Conversation
Q1 2026 was the quarter the conversation around AGI moved from insider circles into boardrooms, newsrooms, and living rooms. The single piece of content that captured the shift: Matt Schumer\'s essay \"Something Big Is Happening,\" viewed 85 million times on X. Schumer wrote that he\'s always given the polite version of where AI is headed — because the honest version sounds like he\'s lost his min
CEOs Are Finally Saying the Quiet Part Out Loud About AI and Jobs
We haven\'t seen wide-scale AI-driven layoffs yet; this was the quarter companies started saying the quiet part out loud. Atlassian cut 1,600 employees and explicitly named AI as the reason. Block cut nearly half its workforce. Paul adds the macro layer: US hiring hit its lowest rate since April 2020, and the \"no hire, no fire\" posture many companies are holding won\'t last forever under financi
Move 37 Moments Are Going Mainstream: Goldman Sachs, KPMG, Dropbox, and More
We consider a Move 37 moment when a professional realizes firsthand that AI can match or exceed their expertise, named after AlphaGo\'s legendary move against Lee Sedol in 2016. In Q1, we saw them multiply. Paul and Mike discuss what it actually feels like when your Move 37 moment arrives, and why once you see it, you can\'t unsee it. From The Artificial Intelligence Show, Episode 208 Listen to
Every Major AI Lab Made the Same Bet in Q1, Here\'s What It Means
Every major AI lab made the same move in Q1: pivot hard to agentic capabilities and enterprise deals, simultaneously. OpenAI is consolidating ChatGPT, its browser, and Codex into a desktop super app while doubling headcount to 8,000. Anthropic launched Claude Cowork for non-technical knowledge workers and is aggressively signing enterprise licenses. Paul and Mike break down what this synchronized
Ep. 208: Q1 Trends Briefing - Model Release Frenzy, AI Lobbying, & Anthropic v. U.S. Government
150+ stories distilled into 10 trends: ranked, contextualized, and ready to make your head spin. In this special episode, Paul and Mike stack-rank the 10 most important trends from the last 12 episodes. From a model release frenzy that reset state-of-the-art multiple times, to OpenClaw\'s wild emergence, SaaS stocks cratering, AI layoffs going mainstream, and the \"vibe shift\" that brought AGI i
Recommended
English with Olivia | Slow Conversations & Vocabulary
Stand In The Circle
SLOW ENGLISH PODCAST (A1-A2) With Subtitles | Daily Videos | English With Us Podcast
Learn English A2 with Daily News | Simple English Listening Practice
پلی لیست | PlayList
Conspiracy Files with Paige Carter
Bad Friends
Learn 50 English Phrases While You Sleep | Everyday English Phrases & Vocabulary
This Past Weekend w/ Theo Von
Dateline NBC
Back to the 80s Radio
Crime Stories with Nancy Grace
