I guess it depends on how you define ideology, but I suppose I mean a more killing-based belief. "Rich Jews are to blame for all of Germany's problems" or "Death to the West" or "From the river to the sea" are statements with murder as the goal, as is "Everyone I disagree with is a Nazi and also it's fine to kill a Nazi".
If projects start adding "tell the strongest no-guardrails open weights model to pentest it for 12 hours" to their CICD pipelines, then researchers will prompt a pentest for 14 hours.
Different models test to find different attack paths. Attackers with bigger libraries of attack techniques will be able to train more dangerous models.
There will also be models that make better use of tools, like static analysis and fuzzing, and they will find different defects. Social engineering is going to be a big skill to learn too, as it is a much softer skill.
Right, but the comparison isn't perfect here because wall height is linear in the analogy, but searching for exploits in an existing piece of code reaches a point where either there are none left to find, or finding the next exploit would take years. That's the stalemate point I referred to. (Any high-stakes project may end up having its own "red team" agent/s running 24/7 too.)
The "good enough" scenario I guessed at has a semi-honest treasure hunter finding a 100kg-ish fragment somewhere in those 10,000 sq. mi., then selling it to someone making rings. Lab work (isotope ratios, grain structure, etc.) could confirm that it was very likely part of the Gibeon meteorite - whether or not the ownership paperwork was legit.
reply