Garbage in, garbage out — and it may not take much garbage to do real harm! This is why data curation and model testing is so important.
Reposted from
Scott McGrath
🧪 NYU researchers show AI models can be easily poisoned with medical misinformation, increasing risks of false outputs. Their study also suggests strategies to intercept and mitigate harmful content. 🩺💻 #MLSky
Comments
https://bsky.app/profile/chrisoffner3d.bsky.social/post/3lbk4ezr3gk2y