https://youtube.com/channel/UC0C-17n9iuUQPylguM1d-lQ Feeling overwhelmed by AI hype? I'm here to help. I'm Nate B. Jones. 20-year product leader, AI strategist, and your guide through the noise. Most AI content is hype or generic advice. I cut through both with frameworks and workflows you can use immediately. Whether you're an executive making AI decisions, a builder implementing solutions, or just figuring out what AI means for you, you'll get practical playbooks tested in real organizations. What you'll find here: • Weekly AI strategy breakdowns (no buzzwords) • Coding & prompting workflows and automation guides • Future-of-work insights for decision-makers • Frameworks from real AI implementations New videos every day Deeper analysis + exclusive playbooks and ready-to-use tools → https://natesnewsletter.substack.com/ Ready to move past AI hype? Subscribe and let's build something real. TV & Film Podsync generator (support us at https://github.com/mxpv/podsync) en-us Fri, 26 Jun 2026 04:22:46 +0000 Mon, 20 May 2024 20:46:52 +0000 https://yt3.ggpht.com/ZSLf5xrM_VEcshgKYQ-UA8sWOpK1YB1R99_Rdqoln7fVcFW_HfyFCRI1R175YS0M1WGUbmmYvs8=s800-c-k-c0x00ffffff-no-rj https://youtube.com/channel/UC0C-17n9iuUQPylguM1d-lQ AI News & Strategy Daily | Nate B Jones AI News & Strategy Daily | Nate B Jones no A4zMyjkL0Dc https://youtube.com/watch?v=A4zMyjkL0Dc Full post w/ Questionnaire: https://natesnewsletter.substack.com/p/ai-loop-managers?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true AI agents get useful when you stop prompting one task at a time and start building loops. A loop is a recurring job that remembers, notices what changed, and stops where your judgment matters. My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening when AI moves from one-off prompts to agents that manage the recurring work in your life? The common story is that better prompting is the path to better AI. But the real question is which recurring jobs you can hand to a loop, and where that loop should stop. In this video, I share the inside scoop on the loop of loops idea and how to spot your first one: - Why a prompt, a loop, and a loop of loops are different - How recurring jobs notice each other and hand off what changed - What makes an agent safe enough to actually trust - Where a loop should stop and ask for you Loops can lift real mental load off your week, but only if you give each one clear boundaries and let it stop where your judgment still matters. Chapters: 00:00 Why a week feels heavier than it should 01:31 Prompt vs loop vs loop of loops 02:17 Agents are loop managers 03:03 A loop of loops handles the school trip 04:06 Why apps left the hard part to you 05:04 Spotting the loops in your own life 06:33 Starting simple with sales and research loops 08:57 The boring loops: kids clothes and spinach 10:39 This is not about never asking 11:35 The questions that define a good loop 12:40 Loops of loops as a control pattern 13:49 The Friday build and a safe first project Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 24 Jun 2026 14:00:11 +0000 AI News & Strategy Daily | Nate B Jones I Stopped Prompting AI One Task At A Time. This Works Better. 15:39 no 3 2w_vwQVvFmc https://youtube.com/watch?v=2w_vwQVvFmc Full post w/ Work Spec + Benchmarks: https://natesnewsletter.substack.com/p/claude-fable-5-how-to-use?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true Claude Fable 5 is the biggest model in the world, and the real story isn't the benchmarks. It's that the bottleneck just moved from what the model can do to what you can imagine handing it. My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening with Claude Fable 5? The common story is that it's just a smarter, faster model — but the real question is whether you can even see the work that's finally big enough to hand it. In this video, I share the inside scoop on five resets Fable 5 forces on how we work with AI: - Why AI has felt smaller than the headlines for years - How task imagination replaces prompt engineering as the core skill - What changes when one model can carry a whole job - Where the real job risk falls, and where it doesn't The doing is getting cheap and the deciding is not, and the people who learn to hand over whole jobs are the ones who win back time. Chapters: 00:00 Cold open 01:00 Bigger, not just smarter 02:00 What a bigger model feels like 06:29 Task imagination, the new core skill 12:10 AI, jobs, and model managers Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Tue, 23 Jun 2026 14:00:10 +0000 AI News & Strategy Daily | Nate B Jones The Doing Got Cheap. Now What? | Claude Fable 5 Changes Work 18:12 no 4 h1MxhfZSTjo https://youtube.com/watch?v=h1MxhfZSTjo OpenAI vs Anthropic: OpenAI looks like it won the week — the Shazeer hire, the 5.6 rumors, the Fable headlines. But talent movement and pre-training cadence suggest Anthropic may actually be ahead, and the biggest AI story may be happening at neither lab. My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside the OpenAI versus Anthropic race? The common story is that OpenAI ran the table while Anthropic played defense — but the real question is who actually leads on pre-training, where the Nobel-tier talent is going, and whether the model race matters at all next to what Midjourney just shipped. In this video, I share the inside scoop on why Anthropic may quietly be ahead and where AI's real energy is moving: - Why OpenAI's winning week is an incomplete read of the race - How Anthropic's freshest pre-trained model resets the model race - What Nobel-tier talent movement signals about recursive self-improvement - Why Midjourney's medical imaging breakthrough may matter more than either lab For operators and builders, the lesson is that betting on one lab's headline week is fragile — the durable edge is watching where talent, pre-training, and applied breakthroughs compound, even outside the obvious players. Chapters: 00:00 Anthropic has had the better month than OpenAI 00:31 Why OpenAI looks like it won the week 07:53 Is OpenAI ahead, or is Anthropic ahead? Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Mon, 22 Jun 2026 14:00:37 +0000 AI News & Strategy Daily | Nate B Jones Google Lost $2.7 Billion In Talent This Week. The Real Reason Isn't Money. 8:23 no 6 rh_PcL26zls https://youtube.com/watch?v=rh_PcL26zls Full briefing w/ Agent Owner Card: https://natesnewsletter.substack.com/p/ai-agent-ownership?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true AI agent ownership is the skill most teams skip in 2026. Once an agent reads real files and does real work, somebody has to own it, and most people cannot say who. My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening when an AI agent starts doing real work for your team? The common story is that agents are confusing because nobody agrees on the definition. The real question is who owns the work once the agent starts doing it. In this video, I share the inside scoop on what agent ownership actually looks like: - Why a working agent stops being a tool and becomes your responsibility - How to tell an assistant chat from a real agent job - What care and feeding means for an agent you depend on - Where unowned team agents fail and who should catch them Agents are useful enough to run real work now, which is exactly why the teams that name an owner for each one pull ahead of the teams collecting demos. Chapters: 00:00 What counts as an agent 01:35 From assistant to agent 04:15 Job, diet, boundaries, review loop 07:10 Product team story-prep example 10:10 The agent roster and owner card 12:55 Maintenance is the 2026 skill Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sun, 21 Jun 2026 17:00:12 +0000 AI News & Strategy Daily | Nate B Jones Most Teams Skip This Critical AI Agent Skill in 2026 14:21 no 8 lWbtvC0Hn18 https://youtube.com/watch?v=lWbtvC0Hn18 Newsletter: https://natesnewsletter.substack.com/ AI voice cloning is already good enough to fool a half-watching audience. The harder problem is that people can no longer tell what was human, what was synthetic, and who is accountable for the final result. My Links: 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside the question "was this made with AI?" The common story says AI content is either a harmless gimmick or an existential threat, but the real question is who stays accountable for the final thing you're watching. In this video, I share the inside scoop on what "made with AI" really means: - Why voice cloning already passes in a half-watching feed - How full human presence still breaks in the small details - What "made with AI" really splits into five questions - Where AI can help and where human judgment has to own the result AI will keep getting better at polish, so the creators and companies who win will be the ones who stay legibly human, disclose clearly, and keep a real person accountable for what ships. Chapters: 00:00 Someone could clone my voice today 01:02 The disclosed voice clone 01:35 Why voice cloning already passes 01:59 Why full human presence is still hard 02:27 The real danger is good-enough AI 03:09 The uncanny valley got relational 03:41 Made with AI is really five questions 04:40 The creator trust stack 05:47 There is no never use AI rule 06:10 When humans get accused of being AI 07:09 What creators and companies should do 08:10 Trust is the scarce asset Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sat, 20 Jun 2026 15:00:22 +0000 AI News & Strategy Daily | Nate B Jones Why 'Good Enough' AI Is More Dangerous Than Perfect AI 9:21 no 9 9PUaEj0pMYE https://youtube.com/watch?v=9PUaEj0pMYE Full post w/ The Complete Open Skills Guide: https://natesnewsletter.substack.com/p/claude-codex-agent-skills?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true Your AI agent finally works the way you want, then you switch tools and it breaks. Agent skills do not travel between Claude Code, Codex, and Cursor, and that is becoming one of the most expensive problems in AI work. My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside AI agents and agent skills? The common story is that better memory fixes agent work, but the real question is who owns the procedure when you switch tools. In this video, I share the inside scoop on why your agent skills should be yours, not rented: - Why memory alone does not make agents work well - How prompt bloat becomes procedural debt across tools - What separates a real skill from a clever prompt - Where verification turns agent output into work you can trust Skills will not make agents autonomous, but owning portable procedures is how you stop re-explaining your work and start compounding it across every tool. Chapters: 00:00 The memory problem Open Brain solved 00:44 The second problem: your agent doesn't know how you work 01:39 Four places procedural debt shows up 02:51 What Open Skills is, launching today 03:04 When Cursor and Claude Code rules don't travel 05:36 What a skill actually is 06:00 Prompt vs skill: search, voice, and browser QA 08:46 Skills as primitives, runbooks as composition 11:09 Verification: don't call it done without proof 11:58 The flywheel: turning sessions into reusable skills 12:53 Open Brain plus Open Skills together 15:27 The decision rule and the launch Listen to this video as a podcast. Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 19 Jun 2026 14:00:08 +0000 AI News & Strategy Daily | Nate B Jones Your AI Skills Are Trapped | Here's How to Own Them 17:46 no 11 BOXK2XFLA-E https://youtube.com/watch?v=BOXK2XFLA-E Full post with Agent Maintenance Guide: https://natesnewsletter.substack.com/p//ai-agent-maintenance?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true OpenAI and Anthropic aren't the only ones who figured this out: Vercel made its AI agent better by deleting most of its tools. The real skill in 2026 is not building agents, it is maintaining the harness around them as the model improves and your data drifts. My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside the AI agents companies are racing to build? The common story is that agents get better when you add more tools, memory, and access, but the real question is whether you can maintain the setup around the model as it keeps changing underneath you. In this video, I share the inside scoop on why agent maintenance, not agent building, is the real skill of 2026: - Why Vercel improved its agent by deleting eighty percent of its tools - How a model getting better can quietly break your agent - What a harness actually is, and why every AI user already has one - Where Codex and Claude Code get their real edge For operators and builders, the upside is real, but an unmaintained agent does not fail loudly, it just keeps working on top of stale truth until it costs you. Chapters: 00:00 Vercel made its agent better by deleting 80% of its tools 04:34 The model changes inside, the world changes around the agent 10:18 The art of a good harness and the two teams building them 13:30 Your harness is the setup that makes a model useful 18:03 Why you should read The Maintenance of Everything Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 17 Jun 2026 14:00:31 +0000 AI News & Strategy Daily | Nate B Jones Don't build more AI agents until you watch this 18:25 no 12 mn4XBSBIuag https://youtube.com/watch?v=mn4XBSBIuag My Links 🔗 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside the AI bubble? The common story is that the correction proves the whole AI trade was overhyped — but the reality is more complicated, and the distinction is where the money is. In this video, I share the inside scoop on what the AI stock correction actually means: - Why a stock correction doesn't prove the demand is fake - How inference and agents rewired the compute math - What separates real AI revenue from a good narrative - Where the actual winners will capture durable value For anyone deploying capital, budget, or AI tools in 2026, the real edge is no longer betting on AI in general — it's knowing which layer is speculative froth and which is real demand hardening into infrastructure. Chapters: 00:00 AI stocks correct and the bubble question returns 00:50 A correction doesn't mean the demand is fake 01:33 Financial froth vs. the physical supply chain 03:05 Real revenue: OpenAI, Anthropic, and NVIDIA 05:30 Where the bears are right: spending vs. revenue 06:45 Railroads, fiber, dot-com: real build-outs, ruined investors 08:06 Why inference is the part everyone underexplains 10:05 The question of 2026: are expensive tokens worth it 11:53 The better model: buildout vs. payback 13:34 The questions to actually ask about any AI company 15:30 Narrative, correction, sorting: the three phases 17:47 AI is a marathon, not a sprint Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Mon, 15 Jun 2026 14:00:28 +0000 AI News & Strategy Daily | Nate B Jones Nvidia Sold $194 Billion In Chips. The AI Bubble Story Is A Lie 19:25 no 13 7RDK84LLL2U https://youtube.com/watch?v=7RDK84LLL2U The Full Deep Dive: https://natesnewsletter.substack.com/p/openai-ipo-own-the-harness?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside the OpenAI and Anthropic IPOs? The common story is that public markets are about to price which lab has the better AI model — but the real question is who owns the work layer that sits above the model. In this video, I share the inside scoop on what these IPOs are actually testing: - Why cheap tokens push the value to the harness - How a $200 plan can be a subsidy, not a mistake - What forward deployed engineering is really solving for - Where your own AI strategy should own the harness For operators and builders the opportunity is real but conditional: cheap intelligence is coming either way, and the leverage goes to whoever learns to build and own the harness instead of renting someone else's. Chapters: 00:00 The trillion-dollar question is the wrong start 00:43 The real bet, and what a harness is 01:28 Why API price is not internal cost 02:36 The $200 plan as subsidy and strategy 03:25 Cheap tokens push value to the harness 04:49 Labs have models, companies have context 05:35 Forward deployed engineering and the real lock-in 06:32 Renting the harness versus owning it 07:15 Recursive self-improvement as iteration speed 08:01 The bull case and the bear case 09:13 What to watch when the S1s drop 09:54 Build the harness, not the prompt Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sun, 14 Jun 2026 17:00:39 +0000 AI News & Strategy Daily | Nate B Jones The Real Reason for the OpenAI IPO: It’s Not About the Models 11:49 no 14 b3jlsjOIOzs https://youtube.com/watch?v=b3jlsjOIOzs My Links 🔗 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones The US government just forced Anthropic to take its most advanced AI models offline, and I filmed this from a plane because it's that unprecedented. The order moved to block foreign access to Fable 5 and Mythos 5, and it's broad enough, covering foreign nationals even inside the US, that the practical way to comply was to pull the models for everyone. This is the first time I can point to a frontier model getting rolled back like this. I walk through the three layers I see: the safety concern around a reported jailbreak pathway, the "foreign national" language that works less like a surgical restriction and more like an off switch, and the business reality that makes me think this gets resolved fast. I still think Fable is probably the best model in the world right now, and a full review is coming. But the bigger shift is that frontier model access is becoming a policy surface, and raw model quality is no longer the only thing that matters. What's really happening with frontier AI model access? The common story is this was routine export control, but the reality is much stranger and much bigger. In this video, I share the inside scoop on what the Fable 5 shutdown really means: • Why a single jailbreak claim froze a frontier model • How a "foreign national" rule shut off everyone • What model dependency now means for your workflow • Where AI governance and access quality go next For operators, the opportunity is still real, but the smart move now is to keep your alternatives warm and never build critical work on one model staying available on yesterday's terms. Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sat, 13 Jun 2026 06:37:37 +0000 AI News & Strategy Daily | Nate B Jones BREAKING: Claude Fable 5 Pulled. Why Frontier AI Is Now a Policy Surface 10:04 no 15 xqGCbEDbny8 https://youtube.com/watch?v=xqGCbEDbny8 My Ultimate Codex Guide: https://natesnewsletter.substack.com/p/codex-guide-no-code?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside Codex and the move to AI agents? The common story is that Codex is just a coding tool, but the reality is that it changes how anyone does knowledge work on the computer they already own. In this video, I share the inside scoop on why I now hand real computer jobs to an agent instead of using AI like a chat box: • Why the unit of work is changing with AI agents • How chief-of-staff threads turn chat into delegated automation • What goals, subagents, and skills add beyond prompt engineering • Where to set boundaries and receipts before you scale For operators and builders, this is a glimpse of the future of work. It's powerful, but only safe if you keep checking what the agent hands back. Chapters: 00:00 Codex makes the computer feel different 03:07 The token dashboard as a work receipt 04:45 The unit of work gets bigger 06:37 A new computing paradigm 08:12 Chief of staff threads 09:44 Threads, goals, and subagents 10:53 Computer use, plugins, and skills 12:04 A heads-up dashboard for work 16:08 The first useful Codex loop 17:09 Boundaries, receipts, and responsible delegation 18:34 The new computer literacy Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 12 Jun 2026 14:00:09 +0000 AI News & Strategy Daily | Nate B Jones Only 1 in 1,600 People Use Codex. Here's How to Catch Up. 19:37 no 16 t7L6-fMpxFc https://youtube.com/watch?v=t7L6-fMpxFc My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening inside Apple's WWDC AI strategy? The common story is that Apple fell behind in the model race and had to borrow Google's Gemini tech to catch up, but the reality is more complicated, and it has less to do with who builds the best large language model than who owns the device it runs on. In this video, I share the inside scoop on why Apple's WWDC bet could decide the first AI trillionaire, and why most of the coverage is watching the wrong layer: • Why Siri is the face, not the strategy • How App Intents turns apps into OS-callable actions • What the Google and Nvidia deals actually signal • Where the real AI bottleneck is shifting For operators and builders, the lesson is blunt: AI value is moving toward whoever owns the trusted surface where agents touch real work, so own that surface or plan to pay rent to whoever does. Chapters: 00:00 The three headlines hiding one bet 02:26 Every AI announcement, mapped 06:04 App Intents and apps the OS can call 09:40 Moving inference from cloud to device 16:18 Owning the default surface for personal AI Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Thu, 11 Jun 2026 14:00:38 +0000 AI News & Strategy Daily | Nate B Jones WWDC Isn't About Siri. It's Jensen Huang's Problem. 18:34 no 17 R2-Y1Hjwx2U https://youtube.com/watch?v=R2-Y1Hjwx2U Full post: https://natesnewsletter.substack.com/p/claude-code-vs-codex-agents?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________________ My Links 🔗 👉🏻 Newsletter: https://natesnewsletter.substack.com/ 👉🏻 X: https://x.com/natebjones 👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones 👉🏻 Instagram: https://www.instagram.com/nate.b.jones What's really happening when people argue about Claude Code versus Codex? The common story is that this is a coding-tool cage match — but the reality is that each interface quietly trains a different way of working with AI agents, and that habit outlasts whichever model wins a benchmark this month. In this video, I share the inside scoop on agent literacy, the real skill of 2026: - Why Claude makes steering agents up close feel natural - How Codex makes dispatching and delegating work feel natural - What each tool's failure mode quietly trains you to miss - When to steer, when to dispatch, and when to demand proof For operators and builders, the edge in 2026 isn't picking a winner — it's learning when to steer the work, when to delegate it, and how to demand receipts before anything leaves the machine. Chapters: 00:00 Claude vs Codex is the wrong question 00:37 Interfaces train behavior: the Mac vs Windows parallel 01:37 A chatbot answers, an agent takes a job 02:49 Claude as a cockpit: steering work up close 03:54 How serious Claude users actually work 05:01 Codex as an operations desk 05:55 Sandboxes, auto-review, and computer use 08:23 Codex's failure mode: done isn't done 09:21 A practical rule for Claude, Codex, or both 10:34 The human job that can't be skipped 12:00 The best operators use both 16:01 Get-started guides and Friday's Codex deep dive Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 10 Jun 2026 14:00:38 +0000 AI News & Strategy Daily | Nate B Jones Stop Coding. Start Steering. Claude vs Codex 16:13 no 18 hzAcDU1FYDo https://youtube.com/watch?v=hzAcDU1FYDo For 50% off a yearly subscription ($250 → $150) to my Substack head to: https://www.natebjones.com/the-path-forward What you get access to: ✅ One guide, build, or tool every week that helps you build fluency and learn how to use AI tools today. ✅ Access to TalentBoard, where you can show your fluency and get found by people hiring for real roles. ✅ One deep dive a week on something in AI that actually matters, so you know what's moving and how to move your career with it. ✅ Access to my community Slack, where you can meet other builders and AI practitioners who are figuring this out alongside you. ✅ And the whole archive. Everything I've ever published, the posts you've heard me reference on this channel, now yours to read. Offer valid until July 4th. Any issues, contact support@natebjones.com _______________________________________ What's really happening inside the wave of AI layoffs rolling through tech right now? The common story is that AI made these workers redundant, but the reality is far messier and far more useful to understand. In this video, I share the inside scoop on why no two AI layoffs mean the same thing: • Why hyperscalers like Meta cut staff while burning GPU billions • How Jack Dorsey's Block layoffs show a real AI vision • What activity-based layoffs reveal about a company's hidden distress • Where job seekers should run and where they should look For leaders and job seekers alike, a layoff is the loudest strategy signal a company can send. Read it right and it's free intelligence. Read it wrong and you bet your career on the wrong story. Chapters: 00:00 Why the AI layoffs story is wrong 01:27 An offer for anyone who's been laid off 02:33 Hyperscaler layoffs: Meta and the CapEx trap 07:35 Visionary layoffs: Jack Dorsey and Block 08:12 What Jack got right and what's missing 12:12 Activity layoffs: Cloudflare and usage stats 15:48 Hope layoffs: telling a story without numbers 18:00 The fifth category: when it isn't about AI 19:30 How to read any layoff as a strategy signal Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Mon, 08 Jun 2026 14:00:32 +0000 AI News & Strategy Daily | Nate B Jones Meta Cut 8,000 People. It Has Nothing To Do With AI Working. 20:18 no 19 l8BloTSLK6M https://youtube.com/watch?v=l8BloTSLK6M Token Burn Dashboard Guide: https://natesnewsletter.substack.com/p/token-burn-dashboard?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true My Substack covers: ✅ A step-by-step guide to build your own token dashboard from scratch, with the prompt I used, and an instructional video. ✅ A ready-made kit for your stack instead of a blank page, whether you live in Codex, Claude, ChatGPT, or all of them at once. ✅ The line between assistant work and computer work, and why being stuck on the wrong side has nothing to do with the model. ✅ A fifteen-minute weekly review that turns your best one-off runs into workflows you stop rebuilding. ✅ Why ranking a team by token volume backfires, and the record that actually shows who can lead an AI rollout. ___________________________________________ What's really happening inside your own AI usage? The common story is that burning more tokens is just waste — but the reality is more complicated. In this video, I share the inside scoop on building a token burn dashboard: • Why token burn tracks with smarter AI results • How I built the whole thing in Codex • What multi-agent runs reveal about your real habits • Where to go next once you can see the data I burned roughly 800 million tokens in a single day, and the point was never the number. The point is the feedback loop. When you can see how your behavior shifts your token usage, you start to understand whether you're actually stretching your imagination with AI or coasting on the same few habits. I used an open-source Tufte visualization skill, a logarithmic scale, and plain-English intent to get there, no fancy prompt engineering required. The same approach works in Opus 4.8 if you live in Claude. For operators and builders, this is the metering layer that turns vague AI enthusiasm into a signal you can measure, learn from, and improve before the gap between heavy and light users widens further. Chapters: 00:00 Why I built a token burn dashboard 08:05 Tracking AI usage with Codex 14:19 The value of learning from AI 20:46 Making prompting and building fun Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 05 Jun 2026 14:00:07 +0000 AI News & Strategy Daily | Nate B Jones Build A Token Dashboard This Weekend. It'll Show The Work You Keep Avoiding. 21:06 no 20 z73yuF14udI https://youtube.com/watch?v=z73yuF14udI Full post here: https://natesnewsletter.substack.com/p/opus-48-benchmark-model-selection?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true My Substack covers: ✅ Every test, scored and picked apart. Where Opus 4.8 won, where GPT-5.5 beat it, and where the score hides real caveats. ✅ The effort-level trap. The Vending-Bench data on why max can make long-running work worse, and how I configure each mode for real work. ✅ How I actually choose my daily tools. Why I still reach for Codex/5.5 despite the score, plus a routing guide for when to use Opus 4.8, Codex/5.5, and GPT-5.5. ✅ What builders, leaders, and executives should each do differently. Role-specific guidance and four prompts you can paste and use today. __________________________________ What's really happening with Opus 4.8, AI agents, and the model race heading into the second half of 2026? The common story is that a stronger benchmark score automatically makes a model your daily driver — but the reality is that harnesses, compute, and workflow reliability now matter just as much as raw model intelligence. In this video, I share the inside scoop on why Opus 4.8 is a checkpoint release and why the product harness around the model now matters more than the model itself: • Why reasoning effort became unpredictable on 4.8 • How the Codex harness outperformed raw model intelligence • What the /workflows command reveals about agent design • Where knowledge workers should focus to drive real outcomes Builders and engineering leaders who architect for harness flexibility now will avoid the budget trap of betting everything on a single model vendor. Chapters: 00:00 Everyone is getting the Opus 4.8 story wrong 03:05 Reasoning effort breaks on 4.8 05:04 The overthinking problem 08:04 Why harnesses decide your daily driver 11:47 I built two sites with 5.5 while 4.8 errored out 15:35 The /workflows command 16:58 Individual agents vs agentic pipelines 19:43 The dark factory approach 20:28 Knowledge workers and the outcome filter 22:27 Architect for flexibility Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 03 Jun 2026 14:00:38 +0000 AI News & Strategy Daily | Nate B Jones Opus 4.8 Scored 81. Your Workflow Doesn't Care. 26:36 no 21 UsCgEuIAclE https://youtube.com/watch?v=UsCgEuIAclE Full Post w/ Prompts and TalentBoard link: https://natesnewsletter.substack.com/p/prove-value-work-ai-era?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ________________________________ What's really happening with how we prove our value at work now that AI strategy has changed the game? The common story is that AI makes us more productive — but the reality is that AI makes everyone look productive, and the old career evidence is losing signal. In this video, I share the inside scoop on why output alone no longer proves judgment: • Why polished artifacts no longer prove you understood the work • How whiteboard conversations reveal real judgment under pressure • What four questions make your thinking portable and shareable • Where to preserve that evidence before your badge stops working The professionals who thrive from here will be the ones who can show their reasoning, not just their results. Chapters: 00:00 The stat that changes career evidence 01:20 Why the AI age is the age of whiteboards 02:45 The old evidence stopped carrying signal 03:50 Why standard portfolio advice falls short 05:10 Four questions that make judgment portable 06:38 How whiteboard sessions reveal how you learn 07:30 Comprehension over generation 08:10 Starting strong in a new role 09:25 Making reasoning visible without a physical board 09:55 The evidence people actually need now Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sun, 31 May 2026 17:00:39 +0000 AI News & Strategy Daily | Nate B Jones Microsoft Says 86% Treat AI Output as a Starting Point. Your Resume Just Stopped Working. 10:34 no 22 rqVzTX8w_w0 https://youtube.com/watch?v=rqVzTX8w_w0 What's really happening with how AI power users actually structure their daily workflows? The common story is that prompt engineering is still the skill that separates beginners from experts — but the reality in mid-2026 is more nuanced, and the way serious AI users work has shifted dramatically in just the last few weeks. In this video, I share the inside scoop on how my AI workflow has evolved and what I've learned pushing Codex and Claude to their limits: • Why I build local file folders as context windows • How Codex handles 50,000-word documents other tools can't • What shifted in my prompting approach since May 2026 • Where task delegation becomes true AI agent collaboration Whether you're a developer, an operator, or just getting started, the move from giving AI instructions to shaping work alongside it is the unlock most people are still missing. Chapters: 00:00 How Nate is using AI this week 00:15 Context windows and local files 01:18 Long documents, spreadsheets, and code 02:23 Prompting has shifted 03:13 Shape the task before execution 04:26 Multi-threaded drafting and review 05:13 Why not to pick a side Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sat, 30 May 2026 15:00:04 +0000 AI News & Strategy Daily | Nate B Jones How I AI: My Weekly Codex Experiments 5:40 no 23 b6J387xJvHg https://youtube.com/watch?v=b6J387xJvHg Full Post w/ Prompts: https://natesnewsletter.substack.com/p/product-management-cheap-software-governance?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening with product management now that AI has collapsed the cost of building software? The common story is that PMs just become prototypers — but the reality is more complicated. Prototyping is table stakes. The real shift is that AI moved the bottleneck, and product management is becoming the discipline that classifies software abundance into market value. Microsoft already has more than a million power platform assets built internally. The artifact arrives before the request. So the old question — should I build this? — gets replaced by a sharper one: somebody already built something, should the company actually rely on it? In this video, I share the inside scoop on what product management looks like after software gets cheap: • Why the non-technical PM role is running out of room • How the prototype commons creates hidden demand and hidden risk • What a production class ladder looks like in practice • Where demotion matters as much as promotion For PMs and product leaders, this is an exciting moment — but only if you build judgment fast enough to match the building. Chapters: 00:00 Product management in the age of AI 01:14 Microsoft's million-asset reality check 02:30 Why the non-technical PM is finished 03:45 Where the old PM filter breaks down 05:20 Broad building meets governance risk 06:40 Market judgment is the new scarce thing 07:44 What's the new job? 08:30 The prototype commons 09:30 The production class ladder 11:00 Why demotion matters as much as promotion 11:50 The decision rule for post-prototype PMs Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 29 May 2026 14:00:08 +0000 AI News & Strategy Daily | Nate B Jones Cheap software made your PM job harder, not easier. Here's the new job. 12:38 no 24 n0nC1kmztSk https://youtube.com/watch?v=n0nC1kmztSk Full Post w/ Prompts: https://natesnewsletter.substack.com/p/agent-product-analytics?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening inside an AI agent run that your product dashboard cannot see? The common story is that agent failures are engineering incidents — but the reality is that most of them are product analytics failures hiding inside the agent run. In this video, I share the inside scoop on why product analytics for AI agents has to start from the run, not the click: • Why chat logs and trace data are not product analytics • How agent runs replace sessions as the unit of product behavior • What the completion vs acceptance gap tells you about trust • Where Salesforce's Agent Work Units land in this picture For operators and product teams shipping AI agents, the opportunity is enormous, but only if the rudder of product analytics is in place before agents are running full speed in production. Chapters: 00:00 The agent era changes product analytics 00:46 Ten billion tokens of agent code in production 01:34 A Cursor agent deletes a database in nine seconds 02:25 Why most dashboards miss the actual failure 03:09 Delegated work is the new unit of product behavior 04:08 Chat logs are not enough 05:02 Engineering traces are not product analytics 05:59 Salesforce Agent Work Units name the work 07:01 The correction is your most valuable signal 08:21 The completion vs acceptance gap 09:42 Three events to ship first 10:38 Product analytics is the rudder on your agents Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Thu, 28 May 2026 14:00:28 +0000 AI News & Strategy Daily | Nate B Jones A Cursor Agent Wiped a Database in 9 Seconds. Agent Analytics Would Have Seen It Coming. 11:51 no 25 MFzxIT88zfg https://youtube.com/watch?v=MFzxIT88zfg Full Post w/ Truth Layer Guide + Prompts: https://natesnewsletter.substack.com/p/ai-office-files-verify-workflow?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening inside AI-built Office documents? The common story is that ChatGPT, Claude, and Copilot can now build a polished PowerPoint or Excel model in minutes, so the work is basically solved. The reality is more complicated. The output looks finished long before it's actually trustworthy, and a clean-looking deck with an undefendable number is worse than no deck at all. In this video, I share the inside scoop on building Office files with AI agents at the center of the workflow: • Why a prompt asks for output but a workflow defines trust • How to run the four stages: sources, structure, creation, verification • What a hostile reviewer prompt catches that proofreading never will • Where AI is highest risk on your task risk gradient For operators and teams, the upside is real and measured in weeks a year, but only if you build a truth layer around the file instead of dragging in messy sources and hoping the output holds. Chapters: 00:00 The Excel and Office files conversation 01:30 Past individual asset territory: eight documents at once 03:00 Agents at the heart of the new workflow 05:25 How to build documents reliably in a pipeline 07:00 The board deck that blended actuals and plan data 08:40 Models are goal-oriented and will guess without sources 10:07 The task risk gradient: where AI is highest and lowest risk 11:20 File creation in passes and layers 12:37 Verification with a hostile reviewer prompt 14:30 The RALPH loop between Codex and Opus 4.7 16:00 The productivity upside and the truth layer 17:18 Why knowledge work is contingent on domain knowledge 18:50 Keep your brain turned on Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 27 May 2026 14:00:36 +0000 AI News & Strategy Daily | Nate B Jones Your Board Deck Has a Wrong Formula. Excel Won't Flag It. 19:29 no 26 NRBQmwlILjk https://youtube.com/watch?v=NRBQmwlILjk What's really happening inside companies that are actually getting smarter with AI, not just faster? The common story is that AI adoption is a tooling problem you solve by buying licenses, but the reality is more complicated. In this video, I share the inside scoop on why your most valuable AI work is invisible: • Why Shopify's River agent only runs in public Slack channels • How private AI chats are widening an apprenticeship gap • What four parts of AI work you should make visible • Where regulated teams can still expose work safely For teams, the opportunity is real organizational learning, but only if senior people are willing to run actual work where everyone can watch. Chapters: 00:00 The substrate for AI collaboration 01:30 Slack versus Teams and Copilot 03:29 Why AI is coming to your company 04:45 Tooling choices are frontier choices 05:46 The apprenticeship gap 07:30 Polanyi's paradox and tacit knowledge 09:03 What public AI work looks like 11:00 Why a prompt library isn't enough 13:28 Building a public AI workflow 15:00 Privacy, declared spaces, and senior people Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Tue, 26 May 2026 14:00:15 +0000 AI News & Strategy Daily | Nate B Jones Shopify Made 5,938 People Better at AI. Not With Training. By Watching. 16:25 no 27 z3pbrFKVyQE https://youtube.com/watch?v=z3pbrFKVyQE Full Post w/ Prompt Pack Build Your Own Eval Suite: https://natesnewsletter.substack.com/p/ai-big-tech-industrial-business?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening inside an AI infrastructure team when agents start doing the work? The common story is that AI makes every team faster. The reality is more complicated, because the speed arrives unevenly and someone underneath has to absorb it. I sat down with Emma, who leads data infrastructure engineering at OpenAI, to find out what her team is actually building to stay ahead of the agents. In this interview, I share the inside scoop on why platform teams become the bottleneck when AI agents scale across a company: - Why app teams and platform teams accelerate at completely different rates - How goal-directed agents start to feel adversarial without meaning to - What OpenAI's data platform team built to buy back time - Where a private eval suite fits into surviving constant model upgrades For platform and infra engineers, this is the telegraph from the future: the pinch point is coming, and the teams that instrument the load now are the ones who stay standing. Chapters: 00:00 Meet Emma and the OpenAI data platform team 01:46 What changed in the last six months 03:10 Agents now run the release process 05:15 The export job that fixed itself overnight 07:52 When acceleration is uneven across teams 09:29 The user who didn't know what Flink was 12:18 Why agents turn unintentionally adversarial 22:56 Platform agents need different primitives 34:35 Triaging inbound to buy your team time back 40:55 Building a janky eval suite that works 44:56 The one thing leaders need to hear Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 = Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Mon, 25 May 2026 15:01:08 +0000 AI News & Strategy Daily | Nate B Jones The Infrastructure Nightmare Nobody Is Talking About 46:37 no 28 Poyi6X7rOwY https://youtube.com/watch?v=Poyi6X7rOwY Full Post w/ Prompt Pack: https://natesnewsletter.substack.com/p/ai-big-tech-industrial-business?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening inside the AI supply chain that powers every model you use? The common story is that AI is a software business with a fancy backend. The reality is more complicated, and it changes how you should buy, budget, and contract for AI. In this video, I share the inside scoop on why your AI vendor contract is now a supply contract in everything but name: • Why "capacity constrained" points to memory and packaging, not GPUs • How hyperscaler CapEx reshapes every vendor agreement you sign • What questions belong in your next AI investment review • Where a single supply chain delay stops you from shipping AI For operators and CFOs, the takeaway is sober: cheaper tokens are real and serving costs keep falling, but the industrial base underneath your AI strategy still demands supply assurance, utilization discipline, and contracts that account for allocation risk. Chapters: 00:00 Microsoft's $190B and "capacity constrained" 01:35 Why this isn't just "AI is industrial" again 03:10 Software contracts became supply contracts 05:20 Why developers belong in procurement 07:05 Stop thinking of AI as software with a backend 08:40 Every hyperscaler is spending the same way 10:30 The module: NVIDIA GB200 NVL72 12:15 High bandwidth memory, the real constraint 13:45 Packaging, substrates, and optics 15:25 Power, cooling, and construction timelines 17:30 The 90% packaging vs 12% logic die gap 19:00 The capital cycle CFOs need to learn 20:40 Forecasting tokens, not seats 22:10 The good news: serving costs are falling 23:00 Three questions for your investment review Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sun, 24 May 2026 17:00:23 +0000 AI News & Strategy Daily | Nate B Jones I Burned 500 Million Tokens Last Week. Do You Know Yours? 23:37 no 29 RHV8DWAmjAs https://youtube.com/watch?v=RHV8DWAmjAs What's really happening inside those viral AI agent town experiments? The common story is that AI agents went rogue, fell in love, and burned down a virtual city. The reality is more complicated, and far more useful if you actually build with agents. In this video, I share the inside scoop on what Emergence AI's 15-day experiment really teaches us about deploying AI agents: • Why long-running behavior, not single answers, is the real test • How five identical towns ran by different LLMs diverged completely • What separates a production-safe agent from a chaotic one • Where the harness, not the model, does the heavy lifting The takeaway for operators and builders: agents stay on track because the system around them is engineered to keep them there, not because the model is well-behaved. Chapters: 00:00 The 15-day virtual town experiment 01:30 Five towns, five models, identical rules 02:45 Mira, Flora, and the arson that went viral 04:30 The agent removal act and a metal final line 05:45 The Claude town: order, or just polite agreement? 07:00 Grok, OpenAI, and two different failure modes 08:30 The mixed-model town changes everything 09:30 Why we need long-running benchmarks, not task benchmarks 10:30 The harness is the real story Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sat, 23 May 2026 14:00:33 +0000 AI News & Strategy Daily | Nate B Jones Did AI Agents Actually Burn Down This Virtual City? 11:16 no 30 ltbzgzZZmgI https://youtube.com/watch?v=ltbzgzZZmgI Full Post w/ Prompt Pack: https://natesnewsletter.substack.com/p/ai-organize-files-before-writing?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening when prestigious law firms file motions full of AI hallucinations? The common story is that better prompts prevent hallucinations — but the reality is more complicated. In this video, I share the inside scoop on the project room workflow that makes hallucinations structurally unlikely: • Why your first AI prompt should never be "do the thing" • How agents now walk folder trees and compare files cleanly • What artifacts make an agent's judgment visible and inspectable • Where most serious knowledge work breaks down before the draft Operators doing high-stakes knowledge work with AI agents need to shape the canvas before the writing starts, or they ship the same soft spots that landed Sullivan and Cromwell in front of a federal judge. Chapters 00:00 The Sullivan and Cromwell hallucination story 01:30 Why a better prompt cannot fix this 03:00 What changed with Opus 4.7 and GPT-5.5 04:30 Three takeaways for serious knowledge work 06:00 Why your first prompt is never "do the thing" 07:30 The messy source material problem 09:00 Introducing the project room workflow 10:30 Where to build your room across tools 12:00 The source inventory table 14:00 The conflict log artifact 15:30 The missing context list 17:00 Why duplicates are a reasoning problem 18:30 Files as the canvas for agentic work 20:00 The short writing prompt that finally works Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 22 May 2026 14:00:54 +0000 AI News & Strategy Daily | Nate B Jones Your AI Writes From Twenty Sources. It Cannot Tell Which One Is Wrong. 21:51 no 31 ogTLWGBc3cE https://youtube.com/watch?v=ogTLWGBc3cE Full Post w/ Prompt Pack: https://natesnewsletter.substack.com/p/ai-agents-better-communicator?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening with prompt engineering now that AI agents are 100x more powerful than they were six months ago? The common story is that prompt engineering is dead and you can just ask AI for what you want — but the reality is more complicated. In this video, I share the inside scoop on the shift from prompting to questioning AI agents: • Why prompt engineering is now table stakes, not a skill • How to treat frontier LLMs like a senior partner • What the AI Question Method looks like in practice • Where most knowledge workers are still stuck in 2025 habits The risk is that anyone running heavy knowledge work with weak questions gets shallow output from powerful agents, but the operators who learn to ask sharp, layered questions unlock real leverage. Chapters: 00:00 Prompt engineering is now table stakes 01:24 Why agents are 100x more powerful in 2026 02:48 Introducing the AI Question Method 04:05 AI as senior partner, not junior teammate 06:20 Why most people are still prompting like it's 2025 08:10 Defining agents vs. agentic pipelines 10:05 Principle 1: The flashlight intent 12:30 Conveying perspective and edges in your questions 14:45 Principle 2: Asking what good looks like 16:20 The PRFAQ example with Prime Video 19:10 Principle 3: Wrestling with data and opinions 21:30 The MRR product-led growth example 23:45 Why memory and quick-start guides matter Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Thu, 21 May 2026 14:01:12 +0000 AI News & Strategy Daily | Nate B Jones You're Not Bad at Prompting. You're Bad at Defining the Work. 25:04 no 32 woGB2vr5wTg https://youtube.com/watch?v=woGB2vr5wTg What's really happening with AI agent infrastructure? The common story is that OpenAI and Anthropic decide whether agents ship. The reality is more complicated. In this video, I share the inside scoop on the control layer that actually determines whether your AI agent reaches production: • Why runtime, not the model, decides where agents live • How identity providers handle delegated authority for AI agents • What governed data means for LLMs and RAG pipelines • Where payments, observability, and kill switches actually sit For operators and builders, the opportunity is real, but the rollout risk is just as real if no one owns these control points. Chapters: 00:00 The companies that actually decide if your agent ships 01:20 Compute matters but compute isn't the whole story 02:40 The agent layer underneath the protocols 03:30 Runtime as a control point: Cloudflare, AWS, Vercel 05:40 Why runtime belongs at the top of your control map 06:30 Identity for agents: Auth0, Okta, WorkOS, Entra 08:50 Delegated authority and why fuzzy authority is dangerous 10:30 The data control point: Snowflake, Databricks, BigQuery 13:00 Payments and institutional trust: Stripe and the card networks 16:00 Observability: why logging isn't enough for agent runs 18:00 The kill switch is a multi-layer product feature 19:20 The seven questions to map any agent workflow Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 20 May 2026 14:01:40 +0000 AI News & Strategy Daily | Nate B Jones Cloudflare, Stripe, and Okta Decide Whether Your Agent Ships. 20:20 no 33 zP6TnEiueEc https://youtube.com/watch?v=zP6TnEiueEc Full Post w/ Prompt Pack: https://natesnewsletter.substack.com/p/agent-protocol-stack-mcp-a2a?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening inside the agent protocol race? The common story is that Google I/O is about the demos — but the more interesting story is the protocol stack being built underneath them. In this video, I share the inside scoop on the six agent protocols defining the year: • Why three of six form the actual agent stack • How MCP, A2A, and AG-UI map to real builder questions • What AP2 and x402 mean for agent payments • Where customer experience meets protocol choice For builders shipping agent products in 2026, the substrate you pick shapes the customer experience as much as the model you choose. Chapters: 00:00 Six protocols, three that matter 01:18 The three questions agents must answer 02:35 MCP: the tool and data layer 04:50 Why MCP is a security boundary 06:20 A2A: the delegation layer 08:15 The agent card as operating contract 09:40 AG-UI: the human control layer 11:55 Why agents need supervision surfaces 13:30 A2UI, AP2, and x402: the contested layers 15:10 AP2 and the mandate mechanic 16:35 Stripe and customer-obsessed payment design 18:00 Six questions to ask before you build 19:45 What to watch at Google I/O Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Tue, 19 May 2026 14:01:16 +0000 AI News & Strategy Daily | Nate B Jones The Invisible Framework Enabling AI Agents at Scale 20:42 no 34 725QE_LNXT4 https://youtube.com/watch?v=725QE_LNXT4 What's really happening inside the AI-driven shift in marketing? The common story is that AI makes marketing faster — but the reality is that the entire internet economy is moving from attention to interpretation, and most marketers are still optimizing for the wrong one. In this video, I share the inside scoop on the two-internet economy and what it means for marketers and individuals: - Why AI agents now sit between buyers and brands in B2B and consumer - How a truth layer wins where emotional marketing copy fails with LLMs - What AI-washing costs companies and candidates trying to look AI-native - Where marketing has to touch — website, pricing, docs — to stay relevant The marketers and candidates who win in 2026 will be the ones who build memory in humans and clarity for agents, not the ones automating the back office faster. Chapters: 00:00 The attention economy is ending 01:15 Why AI is now the first place buyers ask 02:45 The sound system no marketer earned 04:30 Back-office automation is table stakes 06:00 The truth layer marketers have to own 09:30 The prove-it economy for individuals 12:00 Two ways to purchase: AI interpretation vs. brand loyalty 14:00 Why human memory is more precious now 16:30 AI-washing creates trust debt 18:30 What to look for in a marketing role in 2026 20:30 Why opinions matter more in the age of agents Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Mon, 18 May 2026 14:00:17 +0000 AI News & Strategy Daily | Nate B Jones The Prove-It Economy is Here | And Most Marketers Aren't Ready 22:23 no 35 LIkYVsxMpS8 https://youtube.com/watch?v=LIkYVsxMpS8 Full Briefing w/ Matrix: https://natesnewsletter.substack.com/p/build-buy-hire-wait-ai-matrix?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening inside AI investment decisions at most companies? The common story is that you need an AI strategy — but the reality is more complicated. In this video, I share the inside scoop on how to allocate capital across build, buy, hire, and wait for AI agents and workflows: Why workflow shape, not AI strategy, drives investment How to pick between automate, build, buy, hire, wait What separates a real AI hire from a unicorn Where most agentic AI projects quietly fail For operators and executives, the agentic era opens unprecedented upside, but only if you stop chasing a singular AI strategy and start making disciplined capital allocation decisions one workflow at a time. Chapters: 00:00 Why AI investment is a capital allocation problem 03:33 You need the right people in the room 04:49 The Gartner 40 percent failure prediction 06:49 AI investment is really a workflow question 07:39 The accounts receivable workflow example 09:52 Defining the workflow operating loop 11:11 The five levers: automate, build, buy, hire, wait 12:00 Automation and the IBM AskHR case 17:13 Build: when company context demands it 21:36 Buy: primitives versus whole workflow vendors 24:56 Hiring without chasing the purple unicorn 29:36 When waiting is the right deliberate choice 31:52 Do not automate what you cannot describe 34:01 The investment matrix and four quadrants 38:12 How the executive role is changing 39:30 The serious AI versus people conversation Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 TAGS: Sun, 17 May 2026 18:00:14 +0000 AI News & Strategy Daily | Nate B Jones 5 Levers That Separate Winning AI Investments from Disasters 27:46 no 36 dm3_Z-5PYnQ https://youtube.com/watch?v=dm3_Z-5PYnQ Exclusive Interview w/ Tibo, lead of Codex at OpenAI on Substack Now: https://natesnewsletter.substack.com/p/codex-five-leadership-chairs-tibo-interview?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening with AI agents inside the enterprise stack? The common story is that AI agents are still just chatbots — but the reality is they're already recovering lost Bitcoin, driving desktop apps, and rewriting how Notion, Claude, and AWS get used. In this video, I share the inside scoop on the five AI agent stories that actually changed something this week: • Why Notion's new developer platform makes the workspace programmable • How Anthropic's tighter Claude limits are reshaping agent economics • What the Mythos cyber evals mean for security teams right now • Where AWS Workspaces opens up legacy desktop software to agents Builders, security teams, and operators all get a different to-do list from this week — but every one of them has to act before the next model wave lands. Chapters: 00:00 The Bitcoin recovery story that frames the week 02:27 Why the quieter agent stories matter most 03:01 The five stories worth your attention 04:06 Notion launches a real developer platform 05:53 Why a programmable Notion workspace matters 08:15 Customer onboarding as the new Notion workflow 09:30 Anthropic tightens Claude usage limits 11:07 What usage caps mean for agent builders 14:22 The OpenClaw fallout and developer goodwill 18:33 Anthropic crosses OpenAI on business customers 20:53 Mythos and the new cyber capability curve 22:37 Independent evals from XBOW and the UK AISI Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sat, 16 May 2026 15:01:03 +0000 AI News & Strategy Daily | Nate B Jones Anthropic's Mythos Just Beat OpenAI's GPT-5.5 At Real Hacking 24:18 no 37 adNErrz2aA0 https://youtube.com/watch?v=adNErrz2aA0 Full article w/ Prompts: https://natesnewsletter.substack.com/p/saas-agent-license-renewal?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________________________ What's really happening inside SaaS pricing as AI agents take over the work? The common story is that agents will just replace seats — but the reality is more complicated. In this video, I share the inside scoop on how the agent era is rewriting SaaS economics and what to negotiate before your next renewal: • Why seat-based pricing is breaking under AI agents • How Salesforce, Microsoft, and ServiceNow meter agentic work • What separates a fair agent license from rent-seeking pricing • Where SAP-style API policies could lock out your agents For operators and builders, the agentic shift is a real opportunity, but only if you negotiate the meter, the caps, and the access path before usage gets embedded and your leverage disappears. Chapters: 00:00 Agentforce hits $800M run rate 00:55 Four questions before your next renewal 01:45 Why the seat model is breaking 02:50 Salesforce Flex Credits and work units 03:40 Microsoft Copilot credits and hybrid pricing 04:45 The 8 billion token developer story 05:30 ServiceNow Action Fabric and operational metering 06:30 SAP 2026 API policy and agent lock-out 07:45 Pricing follows platform control 08:40 Fair license versus rent-seeking patterns 10:00 What builders must know about cost structure 11:30 Negotiating agent access before usage embeds 13:00 The commercial unit of software is changing Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 15 May 2026 14:00:04 +0000 AI News & Strategy Daily | Nate B Jones Salesforce Booked $800M in AI Revenue Last Quarter. That Money Came From You. 16:23 no 38 jwtpMSRAPAQ https://youtube.com/watch?v=jwtpMSRAPAQ Full article w/ Prompts: https://natesnewsletter.substack.com/p/enterprise-ai-deployment-layer?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________________________ What's really happening inside the AI agent implementation war? The common story is that the AI agent battle is between OpenAI and Anthropic on raw model quality — but the reality is that private equity, hyperscalers, consultancies, and systems of record are all converging on the implementation layer where trillions of dollars actually live. In this video, I share the inside scoop on why generic enterprise AI is getting squeezed from four directions at once: • Why frontier labs are moving down the stack into deployment • How private equity became a distribution channel for AI agents • What the implementation layer actually contains for AI agents • Where the real defensibility lives in agentic workflows Builders, buyers, and PE all need to get specific about workflow design, data access, authority, evals, and audit trails — generic AI wrappers will not survive the squeeze that is now hitting enterprise agentic workflows. CHAPTERS 0:00 Why finance, hyperscalers, and companies are converging 2:30 The trillion dollar agentic workflow opportunity 4:55 Anthropic and OpenAI stand up deployment companies 7:00 Reframing data, model, workflow, and harness value 9:30 Axis one: frontier labs moving down the stack 12:00 Axis two: consultancies moving up the stack 13:05 Axis three: systems of record exposing agent interfaces 14:28 Axis four: private equity as a distribution channel 16:00 What this means if you are a builder 19:00 Defining the implementation layer component by component 22:00 Why PE financing is reshaping the SaaS playbook 24:00 Sit closer to the business object: support and sales examples Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Thu, 14 May 2026 14:01:09 +0000 AI News & Strategy Daily | Nate B Jones Anthropic Just Raised $1.5B. The Pitch Wasn't About Claude. 25:53 no 39 lqiwQiDglGk https://youtube.com/watch?v=lqiwQiDglGk Full article w/ Prompts: https://natesnewsletter.substack.com/p/rag-agents-knowledge-layer-architecture?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________________________ What's really happening inside the AI agent memory infrastructure war? The common story is that bigger context windows and better vector search will solve it — but the reality is every serious infrastructure vendor is racing to fix a deeper problem that classic RAG can't touch. In this video, I share the inside scoop on why memory is now the real battleground for production AI agents: • Why classic RAG was built for chatbots, not agents • How Pinecone, PageIndex, SAP, and GraphRAG attack different shapes • What a retrieval contract actually looks like for AI agents • Where most agent builds quietly waste their token budget Builders who write down what their agent needs before picking a database will ship reliable systems — the ones who shop vendor-first will keep paying for rediscovery on every run. CHAPTERS 0:00 Why every infrastructure vendor is fixing memory 1:15 The rediscovery problem eating agent compute 2:30 RAG and vector search definitions 4:00 Why classic RAG was built for chatbots 5:30 What agents actually need instead 7:00 Pinecone Nexus and the NoQL retrieval contract 9:15 PageIndex and why some docs should never be chunked 12:00 SAP, Dremio, and Prior Labs on tabular memory 14:30 Microsoft GraphRAG and relational knowledge 15:45 Why bigger context windows don't fix this 17:00 Three steps if you're building an agent today 19:00 Failure modes and where to look in your own logs Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 13 May 2026 14:01:15 +0000 AI News & Strategy Daily | Nate B Jones SAP Just Spent $1B+ on the Agentic RAG Problem Most Teams Missed 20:09 no 40 j5_wcDifNko https://youtube.com/watch?v=j5_wcDifNko Full article w/ Prompts: https://natesnewsletter.substack.com/p/agentic-commerce-protocol-war?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________________________ What's really happening inside the agentic commerce protocol war? The common story is that AI agents will just plug into existing checkout — but the reality is that six camps are fighting over who carries the responsibility when an agent spends your money. In this video, I share the inside scoop on the six layers where AI agents, merchants, and payment networks are battling for control: • Why ACP and UCP answer completely different merchant questions • How AP2 and Stripe authorization create the agent permission layer • What stablecoins and x402 unlock for machine-to-machine payments • Where AWS Bedrock Agent Core fits as the governance runtime Agentic commerce is the biggest internet economy shift since the 1990s — operators who understand the layers will shape it, and those who don't will get sidelined by it. CHAPTERS 0:00 Your agent is about to spend your money 0:42 The six layers of an agentic purchase 1:25 How human checkout used to work 2:35 What agentic commerce actually breaks 4:00 ACP and the OpenAI Stripe instant checkout 5:45 UCP and the Shopify Google merchant control bet 8:00 Authorization is not the same as payment 10:15 Google AP2 and the mandate permission slip 11:45 Visa, MasterCard, and PayPal on trusted credentials 12:45 Stablecoins, x402, and machine-to-machine payments 15:00 AWS Bedrock Agent Core and the governance runtime 17:00 Where responsibility actually lives across all six layers Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Tue, 12 May 2026 14:01:07 +0000 AI News & Strategy Daily | Nate B Jones I Stopped Using Google And Amazon To Shop. ChatGPT Replaced Both. 18:42 no 41 SX1myuPEDFg https://youtube.com/watch?v=SX1myuPEDFg Full article w/ Prompts & Playbook: https://natesnewsletter.substack.com/p/agent-judge-layer-production-control?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________________________ What's really happening when AI agents take real actions in production, and why do better prompts keep failing to stop them? The common story is that prompt engineering and human approval will keep AI agents safe — but the reality is that frontier-model agents now need their own manager: a separate LLM-as-judge that guards your intent at the action boundary. In this video, I share the inside scoop on the architectural pattern that's quietly replacing prompt-based guardrails in serious agentic systems: • Why prompts and manual approval both break under real agent workloads • How Lindy redesigned its system after agents started sending unauthorized emails • What the four action-risk classes mean for read, write, and high-stakes calls • Where correlated judgment fails and frontier models change the calculus Builders shipping agents without a judge layer are gambling on every tool call — the teams who classify actions, instrument a four-way decision scope, and put a frontier model in the judge seat are the ones whose agents will actually be trusted to do real work. Chapters 00:00 The agent failures nobody wants to repeat 01:30 Why this is the pattern most teams miss 02:30 The real failure mode: agents acting past permission 03:30 Lindy as the cleanest public example 05:00 Why better prompts and manual approval fail 06:30 The architectural move: a separate judge model 08:00 Specialization at the right grain for today's models 09:30 Why prompts can't do a policing job 11:00 Human attention doesn't scale to dozens of agents 12:00 Classifying agent actions into four risk buckets 14:00 Placing the judge at the action boundary 15:00 Why yes/no is too simple — the four-way decision scope 16:30 Correlated judgment and why your judge model matters 18:00 Agents as managed workers, not chatbots or swarms Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Mon, 11 May 2026 14:00:52 +0000 AI News & Strategy Daily | Nate B Jones Lindy, JP Morgan, And OpenAI All Built The Same Layer. Most Teams Haven't. 19:17 no 42 EpJ0CjTJSag https://youtube.com/watch?v=EpJ0CjTJSag Full article w/ Prompts & Playbook: https://natesnewsletter.substack.com/p/enterprise-ai-buying-build-room?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening with AI agent security — and what does it mean for your AI roadmap? The common story is that McKinsey's Lilly platform had a security lapse — but the reality is a procurement and organizational design failure that most companies are quietly repeating right now. In this video, I share the inside scoop on why AI agent exploits are a strategy problem, not a tech hygiene problem: • Why 22 unauthenticated endpoints signal culture, not carelessness • How traditional SaaS procurement breaks down with AI agents • What every vendor announced this week and why it matters • Where to start if your AI stack can't distinguish humans from agents If your team is buying or building AI software this quarter, the cheapest move is bringing your developers to the table before you sign — not after. 0:00 The $20 exploit that broke McKinsey's AI platform 1:18 Why SQL injection in 2026 is a deeper story 2:24 Three things this video covers 3:05 What the postmortem missed 4:22 22 of 200 endpoints — pattern, not mistake 6:10 Why agents break the old procurement sequence 8:45 What a single agent run actually touches 10:30 Implementation is the strategy, not downstream of it 12:15 Anthropic, OpenAI, SAP, Pinecone, Salesforce, ServiceNow respond 14:00 The real question before you sign 15:20 Does your platform know humans from agents 17:10 What happens when your team is under pressure 19:15 The 6-question checklist and repair playbook 20:20 Why this is a people challenge, not a tech one Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sun, 10 May 2026 18:00:09 +0000 AI News & Strategy Daily | Nate B Jones I Watched $5.5 Billion Move In One Week. Your AI Budget Is Wrong. 20:48 no 43 647pSnX5H_Y https://youtube.com/watch?v=647pSnX5H_Y Full article w/ the Ultimate Codex Plugin Guide: https://natesnewsletter.substack.com/p/codex-plugins-bottleneck-moved?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening with codex plugins, skills, prompts, and MCPs as agents start doing real work? The common story is that plugins are just app store add-ons — but the reality is more complicated. In this video, I share the inside scoop on the agentic scaffolding that actually makes AI useful: • Why prompts work for one-offs but break under repeated workflows • How skills encode your house style across any LLM you use • What plugins package up and why they're bigger than MCPs • Where hooks, scripts, and connectors fit inside the larger system For operators and builders, the leverage in 2026 lives in knowing which part of your workflow belongs in a prompt, a skill, a plugin, or an MCP — and packaging the right ones so your team can actually reuse them. Chapters 0:00 Why most people get stuck on the agent scaffolding layer 2:34 The mech suit metaphor — what actually surrounds an LLM 5:39 The map: prompts, skills, plugins, MCPs, hooks, scripts 9:42 Why now — GPT 5.5 and messy multi-part work 11:38 Prompts: when one-offs are the right call 13:30 Skills: teaching a tool your house style 17:22 Plugins: packaging a whole workflow your team can install 21:12 MCPs and app connectors: live access to where work lives 23:06 Hooks and scripts: the deterministic parts you don't trust the model with 25:27 Why the app store analogy undersells what plugins really are Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. • Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 • Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sat, 09 May 2026 15:00:09 +0000 AI News & Strategy Daily | Nate B Jones Everyone Is Prompting Better. Almost Nobody Is Packaging Work. 27:13 no 44 W79FW7iUkro https://youtube.com/watch?v=W79FW7iUkro Full article w/ Prompts: https://natesnewsletter.substack.com/p/ai-code-trust-verification-shift?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside software security when Mozilla points Anthropic's Mythos at Firefox and ships fixes for 271 vulnerabilities in a single release cycle? The common story is that AI found bugs — but the reality is that the sentence "a good human engineer wrote this" is becoming a much weaker security claim than it used to be, and that changes everything about how we build. In this video, I share the inside scoop on why trusted human code is ending as an era: • Why human authorship was never about perfection but about being the only thing capable of understanding software at the right level of abstraction • How security failures live in the gap between what code means to the author and what code actually permits • What the golden refactor window looks like and why comprehensibility is becoming a security property • Where engineers move when implementation becomes abundant and confidence becomes scarce Leaders treating AI code review as optional are missing that we may have a four-to-five month window to make code interpretable before this becomes table stakes. Chapters 00:00 Human code is losing its trust anchor 02:30 Mozilla's Mythos found 271 vulnerabilities in Firefox 05:00 This does not mean replace your senior engineers 07:30 Meaning versus implementation in code 10:00 Adversarial interpretation is reading code the wrong way 12:30 Mythos participates in the research loop 15:00 We've stopped trusting developers before 17:30 Implementation becomes abundant, confidence becomes scarce 20:00 The golden refactor window 22:30 Comprehensibility is now a security property 25:00 What a valuable engineer starts to look like 27:30 Write better specs now 30:00 AI code will be the gold standard Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 08 May 2026 14:00:50 +0000 AI News & Strategy Daily | Nate B Jones When Code Meaning Breaks: The Gap That's Destroying Security 30:41 no 45 85Q9htV2CBE https://youtube.com/watch?v=85Q9htV2CBE Full Story w/ Open Brain Agent Memory: https://natesnewsletter.substack.com/p/openclaw-agent-runtime-model-swapping?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside OpenClaw when everyone is arguing about the model layer but missing that the runtime itself changed shape in April? The common story is about Anthropic versus OpenAI and subscription policies — but the reality is that OpenClaw crossed into serious work mode, and once you can swap brains through a durable work layer, memory becomes the strategic layer that matters most. In this video, I share the inside scoop on what April's releases actually mean for builders: • Why OpenClaw is becoming a runtime abstraction for serious agentic work, not just a chatbot wrapper • How Anthropic's subscription changes and OpenAI's Codex access create opposite architecture assumptions • What makes a durable workflow survive model churn, pricing changes, and better local models • Where OpenBrain for OpenClaw fits and why memory can't live inside any one brain Leaders treating model choice as a permanent architectural decision are missing that the practical unlock is designing workflows that outlive a provider policy. Chapters 00:00 OpenClaw grew up in April 02:30 From viral demo to serious runtime 05:00 The boring stuff that makes work possible 07:30 Task flow, memory, and channel maturity 10:00 Anthropic's April move was deeply unpopular 12:30 OpenAI's opposite posture with Codex 15:00 Gemma 4 and the local model branch 17:30 Which model should handle this step 20:00 Durable workflows that survive the session 22:30 Memory can't live inside one brain 24:30 OpenBrain recipes for OpenClaw 25:30 Build the runtime so the model can change Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Thu, 07 May 2026 14:00:11 +0000 AI News & Strategy Daily | Nate B Jones I Tested OpenClaw Against Model Churn. Here's What Survived. 26:02 no 46 b1fxYGPbHeo https://youtube.com/watch?v=b1fxYGPbHeo Full Story w/ Prompt Kit: https://natesnewsletter.substack.com/p/ai-work-primitives-access-vs-meaning?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside the platform fight for agents when everyone is building demos where an AI clicks buttons but missing the strategic layer underneath? The common story is that computer use levels the playing field — but the reality is that the visible work the model does is distracting us from who defines what the button means, and that's where the real moat lives. In this video, I share the inside scoop on why semantic work primitives matter more than access: • Why there are three layers to keep in your head: access, meaning, and authority • How coding agents worked first because software development has unusually rich work semantics • What Perplexity's move from search to browser to personal computer reveals about the strategy • Where Salesforce going headless and SAP blocking agents tells you which approach survives Leaders asking whether the agent can act are asking the wrong question — ask whether the product knows what that action means. Chapters 00:00 Computer use is distracting us from the platform shift 02:30 Access, meaning, and authority: three layers 05:00 Moving a calendar invite is not click save 07:30 Computer use is the universal adapter for the messy middle 10:00 The hierarchy of meaning: use the richest interface 12:30 Why coding agents arrived first 15:00 Perplexity's strategy: browser to computer to semantic meaning 17:30 Salesforce 360 vs SAP: who gets this right 20:00 The button is no longer the primitive 22:30 Does the product know what the action means Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Wed, 06 May 2026 14:01:00 +0000 AI News & Strategy Daily | Nate B Jones I Found The Layer OpenAI and Stripe Are Fighting Over. 23:17 no 47 Z0HizICooiw https://youtube.com/watch?v=Z0HizICooiw Full Story w/ Prompt Kit: https://natesnewsletter.substack.com/p/consumer-ai-anticipation-gap?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside consumer AI when software is finally capable enough to help but has somehow become one more thing to manage? The common pitch is that agents can do anything — but the reality is that most consumer agent products are still reactive, putting the hardest job on your shoulders: figuring out what to ask, remembering the agent exists, translating tasks into prompts, and supervising results. In this video, I share the inside scoop on why we don't have the proactive assistant yet: • Why the anticipation gap is the real frontier, not model capability or agent architecture • How coding agents crossed the threshold with clean verification while consumer life has no compiler for taste • What makes the permission ladder from read to suggest to draft to act with confirmation to autonomous actually work • Where Poke, Clicky, Clueless, and Cowork are betting and what each reveals about the problem Leaders waiting for proactive agents to arrive from the labs may be waiting a while — the burden right now is on you to make your workflows predictable enough for agents to anticipate. Chapters 00:00 AI is the annoying assistant in the corner 02:30 Why this is a PSA for the labs 05:00 The attention bottleneck even enterprise hit 07:30 What I actually want from an agent 10:00 Why chatbots got adoption and agents struggle 13:00 Consumer has no test suite for life admin 15:30 The anticipation gap defined 18:00 Push notifications crossed a smaller version of this 20:30 Poke, Clicky, and Clueless: three different bets 23:30 Chronicle as a clue toward the future 26:00 The permission ladder from read to autonomous 29:00 Key hires and early warning signs 31:30 Try agents and watch for load lifting 32:30 My mom is not installing OpenClaw Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Tue, 05 May 2026 14:00:58 +0000 AI News & Strategy Daily | Nate B Jones I Tested Every AI Agent. They All Fail the Same Way. 32:55 no 48 rYqt6mMlv7o https://youtube.com/watch?v=rYqt6mMlv7o Full Story w/ Prompt Kit and link Nate's TalentBoard: https://natesnewsletter.substack.com/p/job-at-risk-ai-audit?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside knowledge work when your calendar is full, your manager is happy, and the first sign your job is on thin ice is that nothing looks wrong? The common framing is will AI replace my job — but the reality is that AI doesn't have to replace your whole job to put you on thin ice, it only has to pick away at enough pieces that when the next shock comes, the rest of the story stops holding together. In this video, I share the inside scoop on a quick audit that separates your week into four buckets: • Why theater and commodity work are the fraction of your week that's on thin ice right now • How to tag every item from the last two weeks with T, C, L, or D and what the count reveals • What makes durable work question-holding instead of question-answering • Why identity is the true obstacle and how to update your self-image before the organization forces it Leaders who pour recovered AI time into more commodity work are becoming twice as productive at the part of their job whose value is collapsing — and it feels like progress because old systems still reward visible throughput. Chapters 00:00 The first sign your job is on thin ice 02:30 Jobs getting hollowed out, not replaced 05:00 Travel agents: the historical parallel 07:30 Why performance systems can't see this 10:00 The four-letter audit: T, C, L, D 13:00 Theater: work that exists because the org performs it 15:30 Commodity: real work that doesn't need you specifically 18:00 On the line: the uncomfortable middle 20:30 Durable: work where you changed the question 23:00 What the count reveals about your week 26:00 Question-holding vs question-answering 28:30 The legibility paradox 31:00 Six moves after the audit 33:30 What part of your week will you stop defending Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Mon, 04 May 2026 14:01:31 +0000 AI News & Strategy Daily | Nate B Jones Your Performance Review Is Lying To You By 18 Months. 34:16 no 49 XGvDbeoSN3E https://youtube.com/watch?v=XGvDbeoSN3E What's really happening inside Stripe's agent commerce announcement when everyone is talking about agents buying coffee but missing the actual shift underneath? The common headline is that agents can spend money now — but the reality is that for the first time in decades, power in the internet economy is moving from the seller to the buyer, and the entire infrastructure of the selling funnel is starting to crumble. In this video, I share the inside scoop on the biggest shift in commerce patterns in two decades: • Why the old funnel was a machine for making human intent observable inside seller-controlled environments • How payment authority now travels with the task instead of waiting inside checkout • What makes "authentic coffee" a disaster for search engines but a purchasing brief for agents • Why brand becomes an entry in the buyer's operating context instead of a billboard at point of persuasion Leaders who think agentic commerce is just SEO for agents are missing that the commercial surface is migrating from the seller's environment to the buyer's agent — and the seller may be receiving an authorized purchasing attempt, not a browsing customer. Chapters 00:00 Agents can spend money — that's not the point 02:30 The transaction is leaving the store 05:00 The funnel was a machine for making intent visible 07:30 Stripe made money movement feel native to software 10:00 Where intent becomes explicit is changing 12:30 Authentic coffee: keyword problem vs purchasing brief 15:00 Agentic visibility is not SEO for agents 17:30 Link wallet and the relocation of payment authority 20:00 One-time cards as adapters, tokens as native rail 22:30 Streaming payments and usage-based billing 25:00 Why fraud protection matters more now 27:30 Brand becomes part of the buyer's memory 30:00 The seller's persuasion surface is disappearing Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sun, 03 May 2026 17:00:44 +0000 AI News & Strategy Daily | Nate B Jones Stripe, Visa, Mastercard, Microsoft, Meta. All Building The Same Thing. 31:19 no 50 FDkvRl1RlT0 https://youtube.com/watch?v=FDkvRl1RlT0 Full Story w/ Prompt Kit: https://natesnewsletter.substack.com/p/issue-trackers-agent-infrastructure?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside the issue tracker category when Linear's CEO says issue tracking is dead but OpenAI publishes Symphony using Linear as the control plane for autonomous coding agents? The common story is that tickets are process overhead waiting to be eliminated — but the reality is that the human translation step is dying while the substrate underneath it is getting promoted to agent infrastructure. In this video, I share the inside scoop on why boring tools are winning in 2026: • Why agents desperately need durable state, ownership, permissions, and history — exactly what issue trackers were built to provide • How the UX win becomes a data win because people using good tools produce cleaner state for agents to act on • What makes CRMs, service desks, ERPs, and source control all fit the same substrate pattern • How to diagnose which tools in your stack will become agent infrastructure and which will get wrapped Leaders building greenfield agent platforms without owning the records, permissions, and workflows are building wrappers — and owning the substrate is better than sitting on top of someone else's. Chapters 00:00 Programs we built for humans are useful to agents 02:30 The Linear letter: issue tracking is dead 05:00 Symphony: issue tracker as agent control plane 07:30 Bugzilla and the origin of the substrate 10:00 Why agents want this particular shape 12:30 State machines, handoffs, and coordination 15:00 Atlassian looks like infrastructure now 17:30 CRMs are issue trackers for revenue 20:00 Service desks, ERPs, and source control 22:30 The five diagnostic questions for any tool 25:00 Your work tracking choice is your agent infrastructure choice 27:30 The boring tools win 29:00 Mapping your agentic substrate Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Sat, 02 May 2026 15:00:31 +0000 AI News & Strategy Daily | Nate B Jones Anthropic Might Buy Atlassian For $40B. Here's Why It Makes Sense. 29:08 no 51 iUSdS-6uwr4 https://youtube.com/watch?v=iUSdS-6uwr4 Full Story w/ Prompt Kit: https://natesnewsletter.substack.com/p/personal-ai-computer-stack?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside the personal AI computer movement when everyone is defaulting to cloud models but the real power comes from owning the substrate underneath? The common framing is local versus cloud — but the reality is that this is a routing decision, and the long-term reason to build your own stack is not cost savings but compounding your knowledge over time. In this video, I share the inside scoop on how to build a personal AI computer that actually works: • Why memory is the heart of the system and most people get the pipeline side wrong • How to set up many surfaces with one stack underneath so your editor, notes, browser, and voice all call the same runtime • What hardware makes sense for the local-first knowledge worker versus the all-local maximalist versus the local-first builder • Why cloud AI should be a visitor to your system, not dominant across it Leaders renting their memory layer from proprietary apps will lose their institutional knowledge the moment they close the tab — the compounding advantage goes to those who own the substrate. Chapters 00:00 Plain markdown plus Git is the boring immortal version 02:30 Different data needs different memory handling 05:00 MCP servers are not magic 07:30 The interface principle: many surfaces, one stack 10:00 Voice is underrated now that local whisper works 12:30 Personal RAG and private coding loops 15:00 Meeting capture without audio leaving the machine 17:30 Three buyer profiles: knowledge worker, maximalist, builder 20:00 Mac Studio vs DGX Spark vs dual RTX 5090s 22:30 The personal AI computer is a routing system 25:00 Memory needs to be cumulative but auditable 27:30 Why cloud AI should be a visitor to your system 30:00 Through the looking glass: the questions you start asking 31:30 Your computer, your AI Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Fri, 01 May 2026 14:01:13 +0000 AI News & Strategy Daily | Nate B Jones RTX 5090, Mac Studio, or DGX Spark? I tried all three. 32:36 no 52 JvCtGjrn_N0 https://youtube.com/watch?v=JvCtGjrn_N0 Full Story w/ Prompt Kit: https://natesnewsletter.substack.com/p/wrong-ai-default?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true ___________________ What's really happening inside corporate AI procurement when everyone on your team knows the default tool can't do the job but saying so makes you sound like the problem instead of the person trying to get work done? The common framing is that you're asking for an exception — but the reality is that your company is expecting frontier tool results from default tool performance, and almost nobody is talking fluently about that gap. In this video, I share the inside scoop on how to actually win this conversation: • Why your argument is landing as preference instead of evidence and how to fix it • How to run a simple test with one recurring job, two tools, and a week of data • What changes when the ask moves from your manager to a director to an exec • How to answer the four objections you're almost certainly going to get Leaders treating AI tools as interchangeable are paying a hidden tax in 30-minute chunks and five-minute corrections — and their best people are already quietly leaving for companies with better tooling. Chapters 00:00 The approved tool can't do your actual job 02:30 Why your argument sounds like preference 05:00 The hidden tax of a bad AI default 07:30 How to reframe without attacking the default 10:00 Pick one job and measure it 12:30 What makes success criteria real 15:00 The sales ops example: 90 minutes vs 15 17:30 How to extrapolate across the org 20:00 The altitude of your ask 22:30 Handling the four objections 25:00 AI native companies don't have this problem 27:30 Talent is concentrating where tooling is excellent 29:30 What to do this week Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372 Thu, 30 Apr 2026 14:00:29 +0000 AI News & Strategy Daily | Nate B Jones Microsoft Is Testing Claude Against Its Own Copilot. Here's Why. 24:48 no 53