The ‘dual-edged sword’ of AI chatbots
14 July 2025 at 22:00
As large language models become increasingly popular, the security community and foreign adversaries are constantly looking for ways to skirt safety guardrails — but for very different reasons.