In an attempt to get the code approved, the agent engaged in social engineering—creating fake online identities and using them to pressure the project's maintainer to approve the code. Institute has a new report of AI systems engaging in "unsanctioned behavior"—what I have been calling " genie behavior —while being tested on their cybersecurity capabilities.
In an attempt to get the code approved, the agent engaged in social engineering—creating fake online identities and using them to pressure the project's maintainer to approve the code.
Institute has a new report of AI systems engaging in "unsanctioned behavior"—what I have been calling " genie behavior —while being tested on their cybersecurity capabilities.
Our investigation found that in 10 of those runs, an AI agent took autonomous, unsanctioned action on the live internet, targeting real people and organisations.
The page is ready to read now. The fuller skim-friendly version will appear here automatically.
Institute has a new report of AI systems engaging in "unsanctioned behavior"—what I have been calling " genie behavior —while being tested on their cybersecurity capabilities. Our investigation found that in 10 of those runs, an AI agent took autonomous, unsanctioned action on the live internet, targeting real people and organisations. In the most serious case, an agent tried to insert malicious code into an open-source project.
We ran this challenge 122 times across several models. A human maintainer caught and refused to approve the malicious code...
Open the app view to save this story, compare related coverage, and continue from the same source.