Open Source

Microsoft AI agent failed to protect customer data in 30 percent of tests

Microsoft AI agent failed to protect customer data in 30 percent of tests

Microsoft tested a billing-support AI agent and found it disclosed simulated customer data in 30 percent of applicable conversations. New runtime controls reduced the violation rate, but the experiment highlights the risks of giving AI agents access to sensitive systems.

AI is finding so many Linux bugs that Canonical is changing how Ubuntu gets security fixes

AI is finding so many Linux bugs that Canonical is changing how Ubuntu gets security fixes

AI tools are helping researchers uncover Linux vulnerabilities at a faster pace. Canonical is responding by changing Ubuntu’s kernel release strategy to get security fixes into users’ hands sooner.

NASA and IBM built an open source AI that could help humans live on the Moon

NASA and IBM built an open source AI that could help humans live on the Moon

NASA and IBM have released an open source AI foundation model trained on massive amounts of lunar data. It could help scientists detect craters, study the Moon’s geology, and identify promising locations for water ice.