OpenAI’s robotics lead exits over Pentagon deal

Good morning, { AI enthusiasts }. The buyer backlash over OpenAI’s Pentagon deal was loud, but this weekend marked an internal escalation: the corporate’s robotics hardware lead resigned on principle.

Caitlin Kalinowski’s departure is the primary senior-level exit tied on to the deal, and her public statement citing concerns over surveillance and lethal autonomy hits a bit harder than any App Store chart.

In today’s AI rundown:

LATEST DEVELOPMENTS

The Rundown: OpenAI robotics director Caitlin Kalinowski just announced her resignation over the corporate’s controversial Pentagon deal, calling it a rushed move that skipped the guardrails on AI surveillance and lethal autonomy.

The main points:

Why it matters: Loads of users have ditched ChatGPT and spoken out for the reason that Pentagon deal dropped — but Kalinowski is the primary senior OAI voice to walk over it on principle. OAI can weather indignant tweets and App Store slides, but an enormous resignation letter that name-drops “lethal autonomy” and “surveillance” hits a bit otherwise.

The Rundown AI agents are entering the workforce, but you wouldn’t expect a brand new worker to know all the pieces on day one, would you? AI agents need onboarding too—in the shape of metadata.

On this eBook, you’ll learn:

Get the eBook.

The Rundown: The Rundown Roundtable is a weekly feature where we poll members of The Rundown staff about how we use AI in our work and day by day lives.

Rishi, Growth: We’re at all times on the lookout for A/B test ideas for our landing pages and latest ways to enhance conversion rates while making a higher user experience. Recently, I attempted an interesting AI workflow after coming across a public CRO skill on Twitter.

I installed it in Claude Code after which fed Claude screenshots of our landing page together with behavioral data from Microsoft Clarity, including scroll depth, heatmaps, and which buttons people were clicking most.

Using the CRO scorecard framework and the Clarity data, Claude generated an in depth evaluation of the page and really helpful five A/B tests we should always prioritize, together with the reasoning behind every one. The insights were genuinely useful, and now we have already taken motion by launching certainly one of the really helpful tests.

Adrian, Developer: I’ve mainly had the very same haircut since highschool, so this yr I finally decided to experiment a bit with Nano Banana 2 and see what a special look might feel like. I uploaded my very own portraits and began merging them with different models’ hairstyles. After 10+ rounds of virtual makeovers, I came upon that the hairstyle that suited me best was… my current one.

The Rundown: On this guide, you’ll learn how one can use Claude to show those old project files, like client emails, metrics exports, or memos, into information-rich case studies you need to use to win more proposals.

Step-by-step:

Pro tip: We are able to pack lots more into our system instructions, including a mode guide and example cases. You’ll be able to grab the templates here.

The Rundown: IBM explains how unified, secured, and governed data access may help organizations move promising AI pilots to reliable enterprise scale.

On this guide, you’ll:

Download the guide.

The Rundown: Anthropic revealed that Claude Opus 4.6 spent two weeks tearing through Firefox’s codebase alongside Mozilla’s team, turning up 22 vulnerabilities (14 high-severity) — with patches already live for tons of of hundreds of thousands of users.

The main points:

Why it matters: Firefox isn’t some latest app; it’s a deeply tested open-source project with many years of audits and bounty programs — making Claude’s quick findings even wilder. While Claude wasn’t as strong at weaponizing its own exploits, Anthropic said that gap won’t last… Meaning the window to lock down codebases feels pretty urgent.

QUICK HITS

Luma unveiled Uni-1, its first model that mixes reasoning and image generation in a single architecture — in a significant shift from the video-focused startup’s roots.

Anthropic rolled out scheduled tasks in Claude Code, letting the coding agent run prompts on a loop to watch builds, check logs, and auto-file PRs on a set cadence.

Cluely CEO Roy Lee admitted to fabricating the startup’s revenue figures in a 2025 interview, publicly retracting claims after the AI ‘cheating’ tool pivoted to meeting notes.

The WSJ shared more on AI’s role within the Iran conflict, reporting that the Army’s 18th Airborne matched its Iraq-era targeting with 20 people as a substitute of two,000 with the tech.

Andrej Karpathy released autoresearch, an open-source repo that lets AI agents autonomously run and iterate on LLM training experiments in a loop on a single GPU.

COMMUNITY

Every newsletter, we showcase how a reader is using AI to work smarter, save time, or make life easier.

Today’s workflow comes from reader Julie H. in Kyle, TX:

“I made a day planner agent connected to my calendar and task lists. Every morning, I start by asking, “How does my day look?”, and the agent pulls all my meetings, uses my tasks list to schedule projects in between meetings, gives me time for deep focus and email catch-up, and even makes sure to schedule a lunch break.

It flags things I can have missed the day before. If I actually have big gaps within the day, it looks ahead and suggests items I can get ahead on, like tasks due the subsequent day or meetings needing prep time. I may give feedback on things I would like so as to add, move, or remove, and it adjusts until I actually have a solid plan for my day. If I move things in my schedule, it should also update my calendar and task list for me so all the pieces stays aligned.”

How do you employ AI? Tell us here.

See you soon,