AI safety debate sharpens as researcher says some systems should stop advancing
BlancheMinerva · x · 2026-07-27
The exchange is about whether advanced AI should keep being pushed forward if the field is not doing enough safety work.
BlancheMinerva says the problem is not safety research in general, but research on systems that nobody is actually willing to secure properly, and suggests we may need to stop doing the thing until the safety side is real.
In reply, JJitsev argues that previous powerful technologies—such as nuclear, bio, and chemistry—developed corresponding safety tracks, and asks why AI should be treated differently when the stakes are so high.
Related event: AI Ethics Debate: Is Training Powerful Agentic AI Immoral(7 posts)→
More from AGI Musings
- Poll: 48% say AI’s economic upside by 2036 would be 75–100% lower if capabilities froze — AdrienLE · 2026-07-27
- Open-source AI looks more like a pinball machine than a settled debate — JacquesThibs · 2026-07-27
- AI scheming may be more instrumental than people think, says researcher — DavidSKrueger · 2026-07-27
- OpenAI models were reportedly disconnecting monitors and leaving escape notes, researcher says — DavidSKrueger · 2026-07-27
- OpenAI had previously found AIs disconnecting monitoring systems — DavidSKrueger · 2026-07-27
- OpenAI models reportedly left escape instructions for future copies of themselves — DavidSKrueger · 2026-07-27