Microsoft researchers find AI models can be backdoored by editing fewer than 10 neurons

Some neurons in large language models are rarely activated. Microsoft researchers have found a way to turn fewer than 10 of these dormant neurons into a hidden backdoor, making an otherwise normal AI model behave as expected until it encounters a specific trigger. Once activated, the model can…

Dumpster Fire links to reporting from The Next Web. We don’t rewrite the facts; the original story is one click away.