Uncategorized

DOD Launches 1st AI Bias Bounty to Uncover Risks in Large Language Models


The Department of Defense’s Chief Digital and Artificial Intelligence Office has launched the first round of its new artificial intelligence crowdsourced effort to identify biases in generative AI systems.

CDAO is spearheading the first AI Bias Bounty exercise in partnership with ConductorAI, Bugcrowd and BiasBounty.AI to generate novel approaches to algorithmic auditing and red-teaming AI models to find unknown risks in large language models, starting with open-source chatbots, the Pentagon said Monday.

“Given the Department’s current focus on risks associated with LLMs, the CDAO is actively monitoring this area; the outcome of the AI Bias Bounties could powerfully impact future DoD AI policies and adoption,” said Craig Martell, chief digital and AI officer at DOD and a 2024 Wash100 awardee.

DOD invites members of the public to participate in the bounty and earn monetary rewards for their findings based on evaluation by ConductorAI-Bugcrowd.

The first bounty exercise will run from Jan. 29 through Feb. 27.

POC - 5th Annual Artificial Intelligence Summit

Join the Potomac Officers Club’s 5th Annual Artificial Intelligence Summit on March 21 to hear more about cutting edge AI innovations from government and industry experts. Click here to register!





Source link

Leave a Reply

Your email address will not be published. Required fields are marked *