OpenAI Introducing an application security agent designed to build deep context about your projects identify complex vulnerabilities and minimize False positive.
The agent is codex securityis currently in the research preview stage, the company announced on Friday (March 6). blog post. Previously known as aardvark When it was in private beta in October.
codex security is designed flag real safety risk and to help security review The post says it will keep up with software development accelerated by agents.
“The combination of agent inference and automated verification from our frontier models provides reliable discovery and actionable fixes, allowing teams to focus on critical vulnerabilities and ship secure code faster,” OpenAI said in the post.
We want to be your favorite news source.
Add us to your preferred sources list to see our news, data, and interviews in your feed. thank you!
OpenAI has begun deploying Codex Security in research preview. ChatGPT Enterprise, work and education It will be made available to customers via Codex Web on Friday and will be available for free next month, according to the post.
PYMNTS reported in July that a new category of tools is emerging. AI first Threat prevention Rather than waiting for alerts, the platform looks for weaknesses in your code, configuration, or behavior and automatically takes defensive actions.
Advertisement: SCROLL TO CONTINUE
The solution is: under development As AI-enabled tools such as agent AI systems and polymorphic malware accelerate cyber attacklowering the barrier to entry for fraudsters and exposing the gap between traditional incident response and forensic models.
of world economic forum said It was discovered in January that A.I. is expected The factor that most influences the formation cyber security This year’s strategy. that said 94% of executives surveyed cite technology as a force multiplier for defense and offense. WEF also highlighted how generative AI technologies are expanding the attack surface.
In this environment, companies and investors are moving towards self-sustaining remediation. Pimunto Reported in February. in some scenario, defense agent remove The need for human intervention above of a particular class vulnerability. Triage and coordination may also be compressed, forcing engineers to focus on higher-level decisions. In both cases, the common denominator is speed, as human remediation is no longer sufficient when AI-driven attackers operate in a continuous loop.
For all of our coverage of PYMNTS AI, subscribe to our daily subscription AI Newsletter.
