Felix Pinkston
Mar 25, 2026 17:33
OpenAI expands its safety efforts with a brand new Security Bug Bounty program targeted on agentic dangers, immediate injection assaults, and information exfiltration in AI merchandise.
OpenAI has launched a public Security Bug Bounty program geared toward figuring out AI abuse and security dangers throughout its product suite, marking a big enlargement of the corporate’s strategy to securing more and more autonomous AI programs. This system, introduced March 25, 2026, particularly targets vulnerabilities in agentic AI merchandise that might result in real-world hurt.
The brand new initiative enhances OpenAI’s present Safety Bug Bounty by accepting submissions that pose significant abuse and security dangers even after they do not qualify as conventional safety vulnerabilities. Researchers who determine points can have their submissions triaged by each Security and Safety groups, with reviews routed between packages primarily based on scope.
Agentic Dangers Take Heart Stage
This system’s scope reveals OpenAI’s rising concern about AI brokers working with rising autonomy. Key focus areas embrace third-party immediate injection assaults the place malicious textual content can hijack a consumer’s agent—together with Browser, ChatGPT Agent, and related merchandise—to carry out dangerous actions or leak delicate data. To qualify for rewards, such assaults have to be reproducible a minimum of 50% of the time.
Different in-scope vulnerabilities embrace agentic merchandise performing disallowed actions on OpenAI’s web site at scale, publicity of proprietary data associated to mannequin reasoning, and bypasses of anti-automation controls or account belief alerts.
What’s Out of Scope
Normal jailbreaks will not qualify for this program. OpenAI explicitly excludes common content-policy bypasses with out demonstrable security impression—getting a mannequin to make use of impolite language or return simply searchable data would not rely. Nonetheless, the corporate runs periodic personal campaigns targeted on particular hurt sorts, together with latest packages concentrating on biorisk content material in ChatGPT Agent and GPT-5.
The corporate will think about edge circumstances on a case-by-case foundation if researchers determine flaws that create direct paths to consumer hurt with actionable remediation steps.
Trade Implications
This launch alerts that main AI builders are taking agentic security severely as these programs acquire capabilities to browse the online, execute code, and work together with exterior providers. The Mannequin Context Protocol (MCP) dangers talked about in this system scope counsel OpenAI is especially targeted on how brokers work together with third-party instruments and information sources.
For the broader AI ecosystem, this program establishes a framework that different firms could comply with as autonomous brokers develop into extra prevalent. Researchers eager about collaborating can apply by way of OpenAI’s Bugcrowd portal, with the corporate emphasizing its dedication to working alongside moral hackers to safe AI programs earlier than vulnerabilities may be exploited at scale.
Picture supply: Shutterstock
