Backdooring Open-Weight AI Models
This is a problem: Any open model whose weights have been edited can carry a backdoor, whether it’s a task-specific fine-tune, a merged adapter or an abliterated build. We have abl…
Aggregated from vendor advisories, security research, and industry publications.
This is a problem: Any open model whose weights have been edited can carry a backdoor, whether it’s a task-specific fine-tune, a merged adapter or an abliterated build. We have abl…
Last week, Anthropic released a long and detailed document describing current misuses of their Claude models. I’m still reading it, but I wanted to flag this: We identified a cell …
Interesting research: “Stealing Reasoning Traces from Proprietary LLM APIs“: Abstract: Leading large language model providers now conceal their models’ step-by-step reasoning, or c…
Not sure it’s practical, but it’s certainly striking.
Someone hid AI instructions into a legal filing. Alternate link.
A usage policy for Flock license plate reader cameras tells police not to talk about the cameras: When cops use Flock to arrest someone in Wapello County, Iowa, they don’t want the…
Yet another Israeli mass surveillance company: Made by Israeli surveillance company Cognyte, the tech simulates a mobile phone tower, which forces nearby phones to connect to it. T…