Back in February, the American AI startup brand – Anthropic had introduced Claude Code Security, allowing users to detect flaws in applications and vulnerabilities that could lead to security threats. Now, almost three months later, OpenAI has also joined the Cybersecurity race and announced its own security-focused tool called Daybreak.
Read more about it below.
OpenAI Announces New Security Tool Called Daybreak
As per what has been shared by OpenAI, Daybreak is not just designed to find bugs or vulnerabilities in older systems, but rather uses AI to assist developers by continuously identifying risks and vulnerabilities during development.
Speaking more about this tool, with the help of OpenAI’s Codex, it can detect subtle vulnerabilities that human developers may miss. Also, instead of just identifying issues, it can suggest fixes and later validate whether the fixes actually work or not. Additionally, it is also capable of analysing complex and unfamiliar codebases. While these same capabilities could also potentially be misused by bad actors, OpenAI says that Daybreak comes with expanded defensive capabilities involving trust, verification, proportional safeguards, and accountability.
Regarding availability, the company has stated that it is currently working with both industry and government partners as part of its iterative deployment approach for increasingly more cyber-capable AI models.
Stay tuned for more such updates.
