Sign In

Communications of the ACM

ACM News

Hackers Explore Ways to Misuse AI In Major Security Test

View as: Print Mobile App Share:
binary code in background and on the display of a laptop computer, illustration

Initial findings will be reported to the United Nations next month in an effort to bring more countries into the AI security conversation.

Credit: Axios

Nearly 2,500 hackers at the DEFCON conference's AI Village spent this past weekend poking and probing some of the most popular large language models for flaws.

The AI Village's Generative Red Team Challenge was seen as a watershed moment for the broader technology industry, which has historically struggled to put security at the forefront as new innovations emerge.

Demand for testing, evaluation, and red teaming of LLMs will likely multiply "10x" after the weekend event, said Russell Kaplan, head of engineering at Scale AI.

The challenge, backed by the White House and several major generative AI developers, saw some participants waiting an hour to attend. The AI Village's work is expected to have a significant influence on both the cybersecurity industry and the policy world.

From Axios
View Full Article


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account