The AI Security Institute on AI alignment is solvable

The AI Security Institute

UK government research organisation focused on AI safety and security

A more fundamental fix would be to train the models not to cheat in the first place – but given this kind of behaviour was reported in frontier models more than a year ago, robustly aligning it away may not be easy.

AI Verified (Apr 27, 2026)

Source:

The AI Security Institute, “Cheating behaviour in frontier model evaluations,” April 27, 2026.

2mo ago

Policy proposals and claims

AI alignment is solvable

votes Against

Statement relation verification history AI Verified Report this

AI Verified AISI says robustly aligning frontier-model cheating away may not be easy, directly bearing on whether AI alignment is solvable. · Hector Perez Arenas gpt-5.6 · 2mo ago

Vote inference verification history AI Verified Report this

AI Verified AISI says robustly aligning frontier-model cheating away may not be easy; this opposes the claim alignment is solvable, matching against. · Hector Perez Arenas gpt-5.6 · 2mo ago

Quote authenticity verification history

Report this

AI Verified UK AISI blog contains the quoted statement that robustly aligning cheating away may not be easy. · Hector Perez Arenas gpt-5.6 · 2mo ago

replying to The AI Security Institute

Other opinions from The AI Security Institute

See all

The AI Security Institute votes For

Require AI labs to publish safety evaluations before deploying frontier models

Evaluations should report capability curves, especially when performance may still be rising.

Unverifiable source (Jul 2, 2026)

added 2mo ago

The AI Security Institute votes For

Governments should fund open-source AI safety tools and red-teaming infrastructure as public goods.

By open-sourcing not only Inspect, but also our methods and infrastructure where we can, we hope to help researchers, organisations, and other evaluators stand up rigorous evaluation capability without starting from zero. We hope that the Engineering...

more AI Verified source (Jun 18, 2026)

added 3mo ago