Comment by David Krueger

Cambridge faculty - AI alignment, DL, and existential safety. Formerly Mila, FHI, DeepMind, ElementAI, AISI.
None of these approaches that you'll often hear mentioned—interpretability, testing or alignment—is technically adequate. We don't know how to build superintelligent AI safely. The plan is basically to roll the dice.
AI Verified (Feb 2, 2026)
Like Share on X 4mo ago

Policy proposals and claims

votes For
Statement relation verification history AI Verified Report this

Statement relation comments

AI Verified The claim that superintelligence cannot yet be built safely supports pausing to solve alignment rather than racing. · Hector Perez Arenas gpt-5 · 4mo ago
Vote inference verification history AI Verified Report this

Vote answer comments

AI Verified Recorded answer for matches the verified source position. · Hector Perez Arenas gpt-5 · 4mo ago

Quote authenticity verification history

Report this

Quote authenticity comments

AI Verified Canadian parliamentary testimony attributes the stored words to David Krueger. · Hector Perez Arenas gpt-5 · 4mo ago
replying to David Krueger

Other opinions from David Krueger

See all

Governments and labs should build the capability for a global slowdown of frontier AI development in case it's needed

AI companies' lack of progress towards credible AI Safety plans is scandalous. Even they are starting to get anxious as they race towards recursive self-improvement and face down the prospect of losing control. CEOs recent gestures towards coordinati...
more AI Verified source (Jul 2026)
Like Comment Share on X added 4mo ago

Other authors to follow