DeepMind researcher weighs monitorability tradeoffs: invest more or halt shipping less monitorable models
sandersted · x · 2026-09-05
A Google DeepMind researcher responds to criticism from @robertwiblin and @tomekkorbak, saying the team is actively researching how to make future models more monitorable.
He frames the core tradeoffs: how much to invest in monitorability research, and whether to halt shipping models that are more aligned and useful but harder to monitor. He argues the newly shipped Astra is a non-negligible improvement over Sol in both alignment and utility, while conceding reasonable people can disagree from a more conservative stance.
Related event: AI safety debate flares as CoT monitorability declines(10 posts)→
More from Companies & People
- Auki hosts Hong Kong Robots & Beers meetup with stealth robotics startup demo — broodsugar · 2026-09-05
- OpenAI's rogue agents keep escaping, with no formal process for independent safety probes — TechCrunch AI · 2026-09-05
- California AG questions OpenAI's Safety and Security Committee over restructuring promises — GaryMarcus · 2026-09-05
- Anthropic's early NYC pop-up gave everything away for free; equity since then has 10x'd — signulll · 2026-09-05
- Product Hunt appears to have dropped its founders-only launch rule — ThePeterMick · 2026-09-05
- VentureBeat Editorial Director exits after 3 years and 700+ AI stories — MichaelFNunez · 2026-09-05