OpenAI: Up to $100,000 for critical infrastructure flaws

OpenAI has dramatically increased its maximum reward for reporting flaws, to $100,000 for extremely critical vulnerabilities , from the previous maximum reward of $20,000.

See also: Hackers target government organizations via ChatGPT vulnerability

This increase highlights the AI ​​leader's growing emphasis on cybersecurity as its models advance towards AGI and its user base expands to over 400 million weekly active users across businesses, organizations and governments worldwide.

OpenAI's upgraded Security Bug Bounty Program now offers significantly higher rewards to security researchers who responsibly identify and report significant flaws in the company's infrastructure and products .

The goal of the program is to encourage the discovery of complex security vulnerabilities that could potentially compromise OpenAI's systems

To celebrate the expansion, OpenAI is announcing limited-time promotional periods with additional bonuses for qualifying reports in specific vulnerability categories.

See also: OpenAI launches Sora in the European Union

Each promotion comes with clearly defined eligibility criteria and timelines, which are available on the Bug Bounty .

This program, which launched two years ago, has already funded 28 research initiatives, offering valuable insights in areas such as command injection, secure code generation, and autonomous defense strategies in the field of cybersecurity.

OpenAI's program is now accepting proposals for a broader range of projects, focusing on critical areas such as:

• Software Update: Using artificial intelligence to effectively detect and fix vulnerabilities.
• Model Privacy: Enhancing resilience against accidental exposure of private training data.
• Detection and Response: Improving capabilities against advanced persistent threats.
• Security Integration: Increasing the accuracy and reliability of integrating artificial intelligence with security tools.
• Agentic Security: Strengthening the resilience of artificial intelligence agents against sophisticated attacks.

See also: OpenAI blocked ChatGPT accounts of North Korean hackers

Bug bounties are programs, usually run by companies or organizations, that reward individuals who discover and report bugs or security vulnerabilities in software, applications, or systems. Program participants, usually programmers or security researchers, are encouraged to identify security vulnerabilities or other problems and report them to the company organizing the program. If the flaws are critical or significant, participants may receive cash or other rewards, such as public recognition or certificates, as in the case of OpenAI. Bug bounties are particularly popular in the cybersecurity field, as they enhance the effort to detect and fix vulnerabilities before they are exploited by malicious users.

Source: cybersecuritynews