How AI guardrails are impeding the work of offensive cybersecurity researchers
For months, AI giants have devised special vetted programs and strict guardrails to limit the use of their models by malicious hackers. But these limits are now hindering the work…