Read two frontier AI companies’ published safety frameworks and compare them side by side using METR’s list of the elements these policies usually share.
These frameworks are the main promises companies make about dangerous capabilities. Clear comparisons show who has committed to what, and where promises are vague.
Output: A comparison table with a one-paragraph verdict.
Read METR’s “Common Elements of Frontier AI Safety Policies”. Turn its elements into the columns of a table.
Choose two companies’ published frameworks (METR links to them). Note each document’s date and version.
For each element, quote what each company actually commits to. Write “not mentioned” where it is missing.
Highlight promises that would be hard to check from outside, such as undefined thresholds or “we may” language.
In one paragraph: which framework is more concrete, and what one change would most improve the weaker one?
All steps done
Share what you made, then try another project. The PDF version has room for notes.
Next: Did they keep their promise? Check the Seoul commitments Next steps
At the 2024 AI Seoul Summit, frontier AI companies signed voluntary safety commitments, including publishing a safety framework. Pick five signatories and check what they actually published.
The Future of Life Institute’s AI Safety Index grades leading companies on public evidence. Apply its indicators to a company it does not cover, or re-score one with newer evidence.
AI biology tools (protein design, CRISPR design and more) sometimes publish biosecurity statements. Add new tools to the repository and compare what they promise.
Butlin and Lappas set out principles for research objectives, procedures, knowledge sharing and public communication. Score three AI developers’ published policies against them, with Suleyman’s essay as a deliberately contrasting position.
If an employee at an AI company sees something dangerous, what protects them if they speak up? Compare protections across two or three jurisdictions.
