🛡 AI Does Not Make Exploiting Vulnerabilities Easier
A VulnCheck study showed that vulnerabilities found using AI are exploited in only 1.3% of cases. This is comparable to the results of traditional bug-hunting methods.
🌍 At the current stage, AI is more likely to help defenders find and close holes faster than it gives attackers a decisive advantage. The main focus is shifting toward protecting the AI stack infrastructure itself.
👤 The hype surrounding an 'AI apocalypse' in cybersecurity is not yet supported by the numbers — automating bug discovery does not mean the automatic automation of their exploitation.