YouCongress
Explore About Sign up

Comment by Adam Hadley

Human avatar
Executive Director of Tech Against Terrorism, a nonprofit countering terrorist exploitation of technology.
Open weights are not the problem in themselves. The acute and least reversible risk is the abliterated build, an open model with its safety training stripped out: the two we tested complied with 89% and 100% of requests, and once downloaded they cannot be recalled.
Unverified (Jul 2026)
Source:
Adam Hadley, “Counter-Terrorism AI (CT-AI) Benchmark,” Tech Against Terrorism, July 2026.
Like Share on X 1h ago

Policy proposals and claims

Open-source AI is more dangerous than closed-source AI
votes Against
Statement relation verification history Unverified Report this
No statement relation verification comments yet.
Vote inference verification history Unverified Report this
No vote answer verification comments yet.
replying to Adam Hadley