AI is transforming cybersecurity on both sides of the battle — and the organizations that embrace openness in threat intelligence, collaboration, and transparent AI systems will be the ones best equipped for the future. This article explores why openness matters and what practical steps defenders should take now.
Anthropic has withheld its most powerful AI model, Claude Mythos Preview, from public release after it discovered thousands of cybersecurity vulnerabilities across major platforms. Through Project Glasswing, the company is sharing the model exclusively with leading tech organisations and committing $100 million to help secure critical infrastructure.
