joy larkin

v.3 of “Who Tests The Frontier AI Labs? What We Know” is live. Three weeks of new system cards, safety reports, and evaluator posts have changed the landscape. ◦ Google DeepMind now names six outside testers in its Gemini 4 Argon safety report. Its earlier reports named none. 1/4

◦ UK AISI’s role shrank at Anthropic and OpenAI in September. The White House lifted its hold on Oct 11. The labs are becoming more open about who tests them, and what those testers can say publicly is the real story now. 3/4

v.3 of “Who Tests The Frontier AI Labs? What We Know” is live. Three weeks of new system cards, safety reports, and evaluator posts have changed the landscape. ◦ Google DeepMind now names six outside testers in its Gemini 4 Argon safety report. Its earlier reports named none. 1/4

📌 Upcoming AI Research & AI Safety Program application deadlines as of October 20, 2026 via cleverhack.com/ai-research-... Included programs: Successif Organization Builders Program, BlueDot’s AGI Strategy, Technical AI Safety and Frontier AI Governance courses, the MATS Residency... 1/2

Folks from around the world kept asking about the AI career links on my frontier lab tracker... so I gave them their own page. cleverhack.com/ai-research-... So far I've discovered 81 programs, fellowships, research orgs, and funding routes into frontier AI and AI safety.

A few days ago, I created a repo to collect news about OpenClaw. There's been a striking dissonance between the hype and what's being said about OpenClaw outside the AI bubble. 🔗 github.com/joylarkin/op...

Vendors continue to update OpenClaw security warnings. And as people get hands on with the tooling, we're seeing daily headlines about unfortunate user outcomes. Hype serves its purpose, but it shouldn't obfuscate reality — even if it’s about the latest in agentic AI.

A few days ago, I created a repo to collect news about OpenClaw. There's been a striking dissonance between the hype and what's being said about OpenClaw outside the AI bubble. 🔗 github.com/joylarkin/op...

Notwithstanding recent conversations, I've been building a neolab and emerging AI lab tracker — with how fast the industry is moving, a new lab seems to appear every few days. Link ➟ cleverhack.com/neolab-and-e...

🪧 New from me: Frontier AI Lab Tracker — a curated, up-to-date index (Sept 2025) of AI teams pushing model scale, multimodal research, and governance work. Useful for investors, product leaders, and researchers mapping capability and safety activity. Feedback or labs I missed? DM me.

🆕 Announcing the latest dataset update (07092025) for the AI Market Maps repo. While the README.md file is updated almost daily, this batch update should be useful for VCs, analysts, and others with a need to access raw URL data. 📄 github.com/joylarkin/Aw... 🤗 huggingface.co/datasets/joy...

The ongoing Cambrian explosion of new-to-market AI app builders has been the one trend I've been most surprised by while developing my AI Coding Landscape. It seems as if a new app builder appears on the timeline each day.

Heads up, y’all! My AI Market Maps repository has been updated for the summer and now features 275+ entries from top VCs and industry analysts. 🔗 Link: github.com/joylarkin/Aw... 📡 RSS: raw.githubusercontent.com/joylarkin/Aw...

🏛️📄 2025 is shaping up as a year of ever-increasing AI regulation. I recently mapped out the global AI policy landscape, and the patterns are striking. Full listing of 2025 AI policy developments, including a state-level US legislative breakdown → cleverhack.com/global-ai-re...

The good news? Most governments are trying to balance technological advancement with responsibility. The challenge? Navigating fragmented regulatory requirements. For AI builders: Understanding these frameworks early equals a competitive advantage and works as an incredible market differentiator.

🏛️📄 2025 is shaping up as a year of ever-increasing AI regulation. I recently mapped out the global AI policy landscape, and the patterns are striking. Full listing of 2025 AI policy developments, including a state-level US legislative breakdown → cleverhack.com/global-ai-re...