OpenAI's GPT-5.6 Sol and Google's Gemini 4 Argon have independently uncovered a critical security flaw in PHP that affects millions of websites, according to The Decoder. This discovery highlights the growing role of advanced AI in cybersecurity efforts.
The vulnerability was first found by OpenAI's GPT-5.6 Sol as part of the Patch the Planet initiative, while the security firm Wiz confirmed the same flaw using Google's Gemini 4 Argon, The Decoder reported. Both AI models demonstrated significant capabilities in identifying complex security issues.
OpenAI also cautioned that the same AI techniques used to detect vulnerabilities could be exploited by malicious actors, raising concerns about the dual-use nature of AI in cyber threats. For Japanese markets, where web security is critical for maintaining trust in digital services and e-commerce, this development underscores the importance of proactive vulnerability management.