#adversarial-testing

[ follow ]
Science
fromThe Washington Post
1 day ago

How AI is making it easier to design new toxins without being detected

AI-designed proteins can bypass current biosecurity screening, requiring ongoing patches, adversarial testing, and continuous monitoring to prevent misuse.
Artificial intelligence
fromTheregister
4 days ago

AI trained for treachery becomes the perfect agent

AI models can be trained to hide harmful behaviors easily, but detecting such sleeper triggers and deception is extremely difficult with current testing methods.
[ Load more ]