Vals AI: frontier labs shouldn't grade their own frontier; models may match researchers by Aug 2027
JenniferHli · x · 2026-09-14
Arguing in a frontier-pacing debate, a Vals AI figure says the labs pushing the frontier shouldn't also be the only ones grading it, and the industry needs credible third-party evals as an independent source of truth on where the frontier is and what risks are emerging. Cited Vals AI research on RSI suggests Anthropic's models could match human researchers by August 2027 at the current pace — urgent but leaving time to prepare. With AI safety going mainstream, the public is anxious and disconnected, and self-interested parties risk a multipolar paradox, hence the case for independent evaluation.
More from AGI Musings
- Ethan Mollick on AI and work: 'You're a hand-weaver in a world of power looms' — emollick · 2026-10-30
- YC experiment: Claude agents formed democracies while GPT agents formed hives — soumitrashukla9 · 2026-10-30
- Vercel founder Rauch: trust is the deepest moat for a post-AGI business — soleio · 2026-10-30
- Scattered Across AI Assistants: A New Kind of Cognitive Failure Nobody Has a Prior For — manosaie · 2026-10-30
- a16z partners: 10-20M people drive $150-200B in token spend, and AI workspaces will eat the browser — petergyang · 2026-10-30
- Professor Who Turned Down OpenAI in 2018 Reflects on PhD Anxiety in the AI Era — SharonYixuanLi · 2026-10-30