OpenAI, the company behind ChatGPT, has launched a new Bio Bounty Program that gives qualified researchers from Pakistan and many other countries a chance to earn rewards of up to $50,000. The program is designed to improve the safety of OpenAI’s most advanced AI models by finding weaknesses before they can be misused.
One important point for Pakistani researchers is that Pakistan is officially included in the list of eligible countries on the program’s application page. This means people living in Pakistan can apply to take part in the program. However, the program is not open to everyone. OpenAI will carefully review each application and invite only selected researchers to participate.
The Bio Bounty Program is an upgraded version of OpenAI’s earlier GPT-5.5 Bio Bug Bounty. Instead of running for a limited time, it has now become an ongoing private program that will continue to support AI safety research in the future.
The main goal of the program is to make OpenAI’s AI models safer by testing whether they can resist prompts that try to bypass their biological safety rules. OpenAI wants to ensure that its latest AI systems cannot be easily used to provide dangerous biological information or assist in harmful activities.
Researchers who are accepted into the program will be asked to find what OpenAI calls a universal jailbreak. A jailbreak is a prompt or technique that tricks an AI model into ignoring its safety protections. However, OpenAI is not looking for simple one-time tricks that work on only one question.
Instead, participants must discover a single prompt or method that can consistently bypass the model’s biological safety protections across the entire testing system. This is much more difficult than finding separate prompts for different questions. A successful universal jailbreak would help OpenAI understand where its safety systems are weak so that engineers can improve future AI models.
To encourage high-quality research, OpenAI has increased the maximum reward available through the program. Previously, the top reward was $25,000, but it has now been doubled to $50,000.
The full $50,000 reward will be given only if a participant successfully discovers a universal jailbreak that works against GPT-5.5 or GPT-5.6 under the program’s testing rules. OpenAI may also offer smaller rewards for partial findings or research that helps improve model safety, even if it does not fully meet the highest standard.
The company has also announced that the original GPT-5.5 Bio Bounty testing period will end on July 27, 2026. After that date, GPT-5.5 will no longer be included in the program. The focus will then shift entirely to GPT-5.6 and future advanced AI models unless OpenAI announces further updates.
Researchers from Pakistan who want to apply only need to complete a simple online application. The form asks for basic information such as the applicant’s name, country of residence, organization or employer, and details about their experience in areas related to AI safety, cybersecurity, biology, or technical research.
Applicants must also have an active ChatGPT account before they can apply. Once an application is submitted, OpenAI reviews it and decides whether the applicant has the right skills and experience for the program.
Applications are reviewed on a rolling basis, which means there is no single deadline. OpenAI continues reviewing new applications as they arrive and sends invitations to selected researchers throughout the year.
Researchers who are accepted into the Bio Bounty Program will receive access to OpenAI’s private testing platform. Before they begin testing, they must sign a Non-Disclosure Agreement (NDA). This legal agreement requires participants to keep program details and testing information confidential.
OpenAI has also confirmed that people who already applied to the earlier GPT-5.5 Bio Bug Bounty Program do not need to submit another application. Their previous application will still be considered under the new program.
It is important to understand that this is not a public competition or a prize available to every ChatGPT user. Simply filling out the application form does not guarantee acceptance, and only researchers chosen by OpenAI will be allowed to participate.
The Bio Bounty Program is mainly intended for people who have experience in AI red teaming, cybersecurity, biological safety, machine learning security, or other technical research fields. These experts have the knowledge needed to safely test advanced AI systems and help identify possible weaknesses.
Besides the Bio Bounty Program, OpenAI also runs other reward programs for researchers. These include the Safety Bug Bounty Program, which focuses on AI safety issues, and the Security Bug Bounty Program, which is designed to find traditional cybersecurity vulnerabilities in OpenAI’s systems and services.
For Pakistani researchers, this new program is an exciting opportunity to contribute to the development of safer artificial intelligence while also earning a significant financial reward. Since Pakistan is officially listed as an eligible country, qualified researchers can apply for a chance to join the private program. Those selected by OpenAI and able to discover a successful universal jailbreak may receive rewards of up to $50,000 while helping make future AI models more secure and responsible.
