AI News & Strategy Daily with Nate B. Jones

For deeper playbooks and analysis: https://natesnewsletter.substack.com/p/let-ai-pick-what-to-automateWhat's really happening when you stop telling an AI what to automate and ask it to discover the problem itself?The common story is that AI agents need a tightly specified task — but the reality is that the strongest systems can inspect real work, identify recurring friction, and propose different high-leverage automations.In this video, I share the inside scoop on giving Fable and Codex the same open brief and getting two very different answers.Why picking the problem is becoming part of the agent's jobHow Fable found a strategic editorial preflight opportunityWhat Codex built to validate completed content handoffsWhere human judgment still mattersHow to turn the method into a reusable automation-discovery skillFor operators, builders, and leaders, the shift is from asking which tool to use to asking which recurring problem is worth solving completely.Subscribe for daily AI strategy and news.Hosted on Acast. See acast.com/privacy for more information.

Every time an AI missed something, I added another rule. Eventually, the accumulated skills, memories, system prompts, checks, and permissions became a hidden system of their own—and that system was getting in the models' way.In this episode, I audit the harness around my AI and compare what happens when Fable 5 and ChatGPT 5.6 meet compact versus overloaded instruction systems. The audit found 66 skill routes, 172 instruction assets, repeated governance rules, and a discovery layer far beyond Codex's stated budget.The lesson is not simply to shorten every prompt. It is to give each surviving instruction one owner and one reason, load specialist context when the work needs it, and enforce deterministic requirements with hard checks.Privacy Policy: https://www.acast.com/privacy

For deeper playbooks and analysis: https://natesnewsletter.substack.com/What's really happening as the model race expands into GPT-5.6, Fable 5, Grok 4.5, GLM 5.2, and increasingly complicated model mixes?The common story is that you should pick whichever model tops the latest benchmark — but the reality is that the best model depends on how you think, how you prompt, and what your hardest work requires.In this episode, Nate shares the inside scoop on choosing a model by work pattern rather than hype.Why “dumber” does not mean dumbHow model families develop different working stylesWhy benchmarks are evidence, not the selection heuristicHow Ringer pairs a strong architect with cheaper workersWhat knowledge-work AI still needs beyond coding harnessesFor builders, operators, researchers, and team leaders, understanding your own work is becoming more durable than memorizing every model leaderboard.Subscribe for daily AI strategy and news.Hosted on Acast. See acast.com/privacy for more information.

AI made it cheap to build almost anything. Most companies still ship at the old pace, and it isn't because their AI is worse than Anthropic's or OpenAI's. The real difference is what they've moved out of meetings and documents and into working code.Full post: https://natesnewsletter.substack.com/p/ai-native-company-rules?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=trueMy 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.jonesWhat's really happening inside the companies that ship AI features every week? The common story is that they just have better AI. The real question is what they rebuilt underneath the model to make that speed possible.In this video, I share the inside scoop on the 15 rules I use to move a company from tool adoption to real AI-native speed:Why moving repeatable coordination into code is now an operator's jobHow 15 commandments work together as one operating systemWhat changes first: roadmaps, meetings, documentation, and designWhere partial adoption fails, and why every rule has to move togetherMoving this fast is real, but only if you treat these as one system instead of picking off the rule that feels easiest to adopt.

Most people bought AI agents and never figured out what to point them at. This is the one-minute test that tells you whether a task belongs in a chat, a single agent, a team of agents, or nowhere near AI.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.jonesWhat's really happening inside the AI agent economy?The common story is that you need more agents. The real question is which tasks are worth any agents at all.In this video, I share the inside scoop on how to spot agent-shaped work: - Why buying more thinking now beats hiring or waiting - How four estimates sort any task in about a minute - What a 40-tool audit surfaced and what it cost to run - Where human judgment still beats every frontier modelThe agents already work. The scarce skill now is knowing which tasks to hand them and which to keep for yourself.

Multi-agent AI systems just went from research project to recipe. I ran 20+ AI agents across 4 model families to rebuild a website in one afternoon for about $8 — and the system caught every hallucination, every shortcut, and even the boss model's own bug without me lifting a finger.Full post:https://natesnewsletter.substack.com/p/trust-ai-agents?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=trueMy 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.jonesWhat's really happening inside multi-agent AI systems?The common story is that hallucinations make AI agents too untrustworthy for real work — but the real question is whether trusting the agent was ever the right design in the first place.In this episode, I share the inside scoop on running a verified agent swarm: - Why one frontier boss plus cheap workers beats frontier-only pricing - How executed checks caught a hallucination, a cheat, and the boss's bug - How to audition new models before trusting them with real work - What a written constitution does that task-by-task prompting can'tHallucinations aren't solved — but with verification built into the structure, delegating big work to AI agents becomes a design question instead of a trust question.

For deeper playbooks and analysis: https://natesnewsletter.substack.com/What's really happening when AI execution gets cheaper but everything starts to feel the same?The common story is that cheaper models make advanced work a commodity, but the reality is that value moves toward the people who can imagine better work, bring context to it, and give themselves permission to run the experiment.In this episode, I share the inside scoop on why imagination x execution is becoming the operating question for AI teams.Why the $9 model test and the $40 model test mean very different thingsHow cheap open execution becomes an engine, not the whole strategyWhat the porch-mailer example reveals about new work no task list had capturedWhere context, permission, and technical imagination meetWhy the Stripe migration story is really about prepared infrastructureIf you are building with AI, managing a team, or trying to understand where frontier spend still matters, the question is not just which model is cheapest. The question is whether your task list has changed.Subscribe for daily AI strategy and news.Hosted on Acast. See acast.com/privacy for more information.

For deeper playbooks and analysis: https://natesnewsletter.substack.com/What's really happening when an email/calendar agent becomes useful enough for real paperwork?The common story is that AI agents need a totally new setup for every hard job -- but the reality is that the same safe skeleton can learn on email, then carry into insurance appeals and tax-prep packets.In this episode, I share the inside scoop on building one reusable agent pattern for messy, high-trust paperwork:Why email/calendar is the 101 where mistakes are cheap How the same skeleton moves into denied insurance claims What a cited appeal packet should do, and what it should not promise Why tax prep should produce a reviewable packet, not a return Where the human approval gate has to stay intactThis is for builders, operators, and anyone trying to move past cute demos into agents that organize real context, cite their work, export reviewable packets, and stop before the human decision.Subscribe for daily AI strategy and news.Hosted on Acast. See acast.com/privacy for more information.

For deeper playbooks and analysis: https://natesnewsletter.substack.com/What's really happening when every AI model suddenly looks replaceable?The common story is that model choice is the strategy, but the reality is that useful work comes from matching the model, the task, and the workflow surface.In this episode, I share the inside scoop on how to pick AI models without turning model selection into the whole job.Why daily-driver models are different from cheap workhorse models How to think about GLM, Kimi, Qwen, Claude, ChatGPT, and Codex What specialist tools are actually for Where harnesses and workflows matter more than raw model rankings Why fewer model choices can make teams fasterThis is for operators, founders, developers, and team leads who need practical AI work to keep moving even when the model landscape shifts underneath them.Subscribe for daily AI strategy and news.Hosted on Acast. See acast.com/privacy for more information.