Thomas Holland on Require AI systems above a capability t…

Thomas Holland

Policy analyst at the Nakada Foundation to Save Humanity, writing on AI governance and superintelligence risk.

This is why Anthropic has made interpretability research a stated organizational priority, and why the Foundation's governance proposals include interpretability verification requirements as a condition of frontier AI deployment.

AI Verified (Jun 9, 2026)

Source:

Thomas Holland, “What Is Mechanistic Interpretability?”, Nakada Foundation, June 9, 2026.

20d ago

Policy proposals and claims

Require AI systems above a capability threshold to be interpretable

votes For

Statement relation verification history AI Verified Report this

AI Verified The quote makes interpretability verification a deployment condition for Frontier AI; that is a requirement that capability-threshold systems be interpretable before deployment. · Hector Perez Arenas gpt-5 · 20d ago

Vote inference verification history AI Verified Report this

AI Verified The quote supports a deployment condition requiring interpretability verification for Frontier AI, so the recorded “for” answer is correct. · Hector Perez Arenas gpt-5 · 20d ago

Quote authenticity verification history

Report this

AI Verified Nakada Foundation’s June 9, 2026 article, by Thomas Holland, contains the quoted sentence verbatim and identifies Holland as its author. · Hector Perez Arenas gpt-5 · 20d ago

replying to Thomas Holland

Other opinions from Thomas Holland

See all

Thomas Holland votes For

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

The MIRI draft is a full international agreement to halt the race to superintelligence, complete with compute limits, a chip-cluster line, and a verification regime built for rivals who do not trust each other.

Unverifiable source (Jul 5, 2026)

added 2mo ago

Thomas Holland votes For

Ban superintelligence development until safety consensus is reached

The Nakada Foundation is building the political case for a binding international agreement to halt the race to superintelligence before it is built.

AI Verified source (Jul 5, 2026)

added 2mo ago

Thomas Holland votes For

Nations should establish binding international rules for AI safety and governance

Crucially, the treaty should delegate the exact numbers to a technical body empowered to revise them, because any fixed figure written into the text will be obsolete within a few years.

AI Verified source (May 15, 2026)

added 2mo ago