Eric Schmidt

Within about two years (by mid-2025), AI systems will be able to find zero-day exploits in software or discover new kinds of biology.

“There are scenarios not today, but reasonably soon, where these systems will be able to find zero-day exploits in cyber issues, or discover new kinds of biology. Now, this is fiction today, but its reasoning is likely to be true.”

Said on

24 May 2023 · Wall Street Journal CEO Council Summit, London

Source

AI poses existential risk of people being 'harmed or killed,' ex-Google CEO Eric Schmidt says · CNBC

Should come true

by 2025

How sure they sounded

reasonably soon; its reasoning is likely to be true

Categories

AITechnology

Importance

4 of 5 · AI that finds zero-day exploits or new biology has major global security consequences.

Outcome

Accurate Happened 1 November 2024.

Inferred window "reasonably soon" = about 2 years. In November 2024 Google's Big Sleep AI agent found a previously unknown exploitable memory-safety bug in SQLite, described as the first public example of an AI agent finding such a flaw in widely used real-world software, inside the window.

Evidence:

• https://googleprojectzero.blogspot.com/2024/10/from-naptime-to-big-sleep.html