Did Altman already secretly claim OpenAI hit AGI? Netizens dig through old interviews

RileyRalmuto · x · 2026-09-23

RileyRalmuto recalls an interview where Sam Altman seemed to implicitly claim OpenAI has already reached its stated goal: while reviewing evals of their current SOTA (or internal evals for Astra), Altman said watching the agents "really hit me like 'alright, well we really did it.'" The poster argues the statement was deliberately subtle and the host missed its weight. It stems from ChrisGPT's question about whether Anthropic avoids claiming AGI due to high standards or fear of looking like it hasn't paced the frontier. The poster wonders if anyone at Anthropic has hinted at something similar.

Original post →

More from AGI Musings

AGI Musings channel →