Wiz Deploys AI Red Agent Across 200 Environments a Week as Defenders Claim Edge
Updated
Updated · CTech · Jul 26
Wiz Deploys AI Red Agent Across 200 Environments a Week as Defenders Claim Edge
3 articles · Updated · CTech · Jul 26
Summary
Wiz says its AI-powered Red Agent now probes about 200 customer environments weekly, proactively exploiting exposed weaknesses before attackers can and surfacing thousands of critical vulnerabilities.
The system works by mapping an organization’s full cloud attack surface, ranking the assets most likely to be vulnerable, then adapting its attack path in real time as servers respond.
Gal Nagli, Wiz’s head of offensive security, said AI has compressed work that once required elite researchers, while also lowering the bar for attackers who can now ask models to find sensitive data or hack websites.
Wiz argues that the same shift may finally favor defenders because companies hold far more internal context than outsiders; about 30% of its customers, including 65% of the Fortune 100, already use the automated testing.
The company also warns that AI is creating new risks inside firms, especially from non-engineers building chatbots or websites since 2023, while low-quality AI bug reports are flooding disclosure programs.
As AI floods bug bounty programs with fake reports, will the cybersecurity industry abandon human hackers for autonomous defense agents?
With marketing teams launching vulnerable chatbots overnight, are citizen developers becoming the biggest insider threat to enterprise cybersecurity?
If AI allows attackers to find vulnerabilities instantly, can defenders truly rely on hallucination-prone models to secure their systems within an hour?