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.
Voluntary commitments only matter if someone checks them, and simple, well-sourced tracking is one way to do that.
Output: A small tracker: company, commitment, evidence, date, status.
Open the Frontier AI Safety Commitments page on GOV.UK. List the specific things companies agreed to do.
Mix large and smaller signatories so the picture is fair.
For each company, find its published framework or statement. Record the link and date, or “not found”.
Use three simple labels: done, partly done, not found. Write one sentence justifying each.
Note your search method and date so others can update it. Share it with your group.
All steps done
Share what you made, then try another project. The PDF version has room for notes.
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.
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.
UK public bodies publish records about the algorithmic tools they use under the Algorithmic Transparency Recording Standard. Read five records and assess what is disclosed and what is missing.
Pick one verification mechanism, such as network taps, trusted execution environments, chip telemetry or satellite monitoring, and write a short, clear analysis of how it works and how it could fail.
