The AI Now Institute reported that Heidy Khlaaf is calling for alarming claims about AI safety to be supported by evidence and specific examples. This approach should help verify how well-founded warnings about potential risks are.
The impetus was statements and essays warning of threats to humanity from AI. The available description does not provide details of these materials, the positions of the other experts, or specific verified cases.
The significance of the publication is currently limited to the scope of expert discussion: the source offers a criterion for verifying claims but does not present new empirical data in the package. Assessing risks requires described scenarios, facts confirming them, and independent verification.
editorial commentary
Why it matters
The likely consequence is increased demands for evidentiary support in public statements about extreme AI risks. The next observable signal will be the emergence of specific scenarios, data, and independent checks related to such claims. Substantial uncertainty remains: the available package lacks the full text of the publication and supporting materials.