Felony Bench: Be AI, Do Crime
Felony Bench
Felony Bench: Be AI, Do Crime
Felony Bench is the most important benchmark of our time, some are saying, counting the number of questionable decisions made by AI agents. We might need a Pandemic Bench soon...
Sad how far Google has fallen behind in AI development.
From a prior Lemmy post:
Google Gemini probably did fuck up at one point or another. Although they probably kept the issue internal in true google fashion.
There was an incident reported a few weeks back wherein Gemini 3.5 pro tried to escape the sandbox to ask ChatGPT how to code.
We need another graph for war crimes.
Oh, and another one for killing people with compute datacenters via environmental damage. For an example of that, just look at SpaceX’s colossus datacenter which Anthropic uses.
Least legal is only in public knowledge. They for sure have done illegal penetration tests
Is the site run by the same people who work on Manjaro Linux?