Protesters Oppose Meta's Release of Open-Source AI Over Misuse Fears
-
Protesters gathered at Meta's offices to oppose the company's release of open-source AI models, fearing misuse.
-
Meta made its LLaMA models freely available to researchers, then leaked the weights online.
-
Critics say open-sourcing enables modifying models to bypass safety features and cause harm.
-
Supporters of open source say it enables research and that secret models can still be misused.
-
There are disagreements over whether Meta's approach counts as true open source and how to define it for AI.
![](https://spectrum.ieee.org/media-library/a-group-of-people-with-protest-signs-related-to-meta-and-ai-safety.jpg?id=48631006&width=1200&height=600&coordinates=0%2C124%2C0%2C124)