Openai Bug Bounty Program Increases Top Reward To $100,000

OpenAI is prioritizing information pinch a awesome bug bounty programme summation and caller AI information investigation grants. Find retired really they’re collaborating pinch researchers and experts to protect their AI platforms from emerging threats

OpenAI is enhancing its information infrastructure, focusing connected a forward-looking attack towards AI, by expanding information initiatives crossed assistance programs, bug bounties, and soul defences.

In its latest blog post, OpenAI has unveiled a suite of caller cybersecurity initiatives, signalling a bold push towards artificial wide intelligence (AGI). A cardinal constituent of this strategical move is simply a important summation successful nan maximum reward offered done its bug bounty program, now reaching $100,000 for captious findings.

As antecedently reported by HackRead.com, OpenAI launched its bug bounty programme successful April 2023 successful business pinch Bugcrowd. This programme initially focused connected finding flaws successful ChatGPT AI chatbot to heighten its information and reliability, pinch rewards starting astatine $200 for low-severity findings and reaching $20,000 for exceptional discoveries.

Now, OpenAI has confirmed that this programme is seeing a important overhaul pinch nan maximum pay-out accrued from $20,000 to $100,000 and nan scope of programme being broadened substantially, a move OpenAI states reflects its committedness to ensuring users’ spot successful its systems.

“This summation reflects our committedness to rewarding meaningful, high-impact information investigation that helps america protect users and support spot successful our systems,” nan institution noted successful nan announcement.

To further incentivize participation, OpenAI is introducing limited-time prize promotions, pinch nan first focusing connected IDOR entree power vulnerabilities. This promotion, moving from March 26th to April 30th, 2025, besides increases nan baseline bounty scope for these types of vulnerabilities.

The institution besides plans to grow its Cybersecurity Grant Program, which has already funded 28 investigation projects focused connected some violative and protect information strategies. These projects person explored areas specified arsenic autonomous cybersecurity defenses, unafraid codification generation, and punctual injection. The assistance programme is now seeking proposals for 5 caller investigation areas: package patching, exemplary privacy, discovery and response, information integration, and agentic AI security.

OpenAI is besides introducing microgrants successful nan shape of API credits to facilitate accelerated prototyping of innovative cybersecurity ideas. Furthermore, it plans to prosecute successful open-source information research, collaborating pinch experts from academic, government, and commercialized labs to place vulnerabilities successful open-source package code.

This displacement is aimed astatine improving nan expertise of OpenAI’s AI models to find and spot information flaws. The institution plans to merchandise information disclosures to applicable open-source parties arsenic vulnerabilities are discovered.

In addition, OpenAI is integrating its ain AI models into its information infrastructure to heighten real-time threat discovery and response. To fortify its defences, nan institution has established a caller reddish squad business pinch SpecterOps, a cybersecurity firm. This collaboration will impact laborious, simulated attacks crossed OpenAI’s infrastructure, including corporate, cloud, and accumulation environments.

As OpenAI’s personification guidelines expands, now serving complete 400 cardinal play progressive users, nan institution acknowledges its increasing work to safeguard personification information and systems. While it focuses connected processing precocious AI agents, nan institution is besides addressing nan unsocial information challenges associated pinch these technologies. This includes defending against punctual injection attacks, implementing precocious entree controls, broad information monitoring, and cryptographic protections, reinforcing their dedication to building unafraid and trustworthy AI.