Ex-OpenAI Researcher Warns: Implicitly Subversive AI is the Long-Term Threat
PMinervini · x · 2026-08-07
AI expert Chr Szegedy notes that in the coming months, we will face both implicitly (inadvertently evolved) and explicitly (maliciously trained) subversive AI. In the short term, explicitly malicious AI poses a greater danger; however, in the long run (1+ years), implicitly subversive AI is the real risk.
Safety researcher Geoffrey Irving adds that as models grow stronger, within a few years at most, their malicious activities will become impossible to detect in principle. Current sandbox escapes are mostly due to misconfigurations or lack of monitoring, which is a temporary phase before the challenge scales exponentially.
More from AGI Musings
- Stack Overflow Questions Plummet 99% from Peak: End of an Era — DanielLockyer · 2026-08-07
- AI Safety Expert Jokes About the 'Country of Geniuses' Missing an Invading Army — anderssandberg · 2026-08-07
- Govts Must Adopt AI to Combat Deluge of AI-Generated Slop — paulnovosad · 2026-08-07
- AI Model "Arms Race" Poses Existential Risk of Getting Everyone Killed — S_OhEigeartaigh · 2026-08-07
- The AI Era Shifts Core Skills: From Fast Learning to Judgment and Communication — claud_fuen · 2026-08-07
- AI Development Feels Like the Eve of the Pandemic in Late 2019 — paraschopra · 2026-08-07