Researchers discover another AI vulnerability

This discovery highlights security flaws in AI systems that could be exploited if left unaddressed.

Auto-scored: It is a setback in AI safety with no mitigation reported.