Despite its rapidly increasing popularity, ChatGPT is not free from the typical problems that all software experiences. Tasks like user experience may suffer from a few glitches and bugs.
However, on a more serious level, malicious and nefarious actors could exploit these issues and expose OpenAI’s 100 million users to potential security risks. In an effort to safeguard its audience, the offshore software development company is now offering a reward of up to $20,000 for anyone able to identify and detect any programming faults.
Bug Hunting with OpenAI’s Hefty Rewards Program – Get Paid for Finding Issues
On Tuesday, April 11th, OpenAI kicked off its new Bug Bounty Program, extending an invitation to investigative security researchers, ethical hackers, and technology-savvy individuals to tackle its products, like ChatGPT, to check for susceptibilities, errors, or issues. The reward depends on the gravity of the issue uncovered, ranging from $200 for small-scale discoveries to as high as $20,000 if a remarkable identification is made.
Companies are now turning to externally sourced bug bounty programs as a sort of virtual beta test to ensure that their releases are running smoothly. Rather than using an internal network of developers to check for issues, these programs open up their testing procedures to anyone willing to join, greatly increasing the breadth and depth of potential issues that can be identified.
As far as bug bounties go, it’s a high-stakes affair; companies are primarily attempting to discover scenarios where potential vulnerabilities exist in their software, thus putting their consumers under the specter of potential security risks.
If you’re eager to join OpenAI’s lucrative Bug Bounty Program, the sign-up process is simple and straightforward. First and foremost, you need to register an account on OpenAI’s website and confirm your email address.
After creating your account, simply fill out the short registration form by providing a few basic details, such as your name, address, and type of bug bounty program you would like to participate in.
Finally, check the confirmation box to signify that you’ve read and agreed to OpenAI’s terms of service for the Bug Bounty Program, and then click “Submit” – you’re all done! Within no time, you’ll have access to a host of great rewards for helping OpenAI identify pesky bugs.
OpenAI and Bugcrowd, which specialize in finding bugs, have innovatively formed a partnership to launch a Bug Bounty Program. Through Bugcrowd’s platform, willing hackers have the chance to get paid for discovering vulnerabilities. Currently, there have been 24 lucrative rewards ruled out, with the standard reward at an average of $983.33.
OpenAI is adamant that safety issues do not factor into this program. Should you discover that the model isn’t performing as it’s supposed to, please submit your feedback via the model behavior form rather than entering the Bug Bounty Program. Submissions regarding malicious instructions or written code and hallucinations will not be eligible for rewards.
If you’re planning on submitting any bugs for OpenAI, be sure to double-check their Bugcrowd page for a comprehensive overview of accepted – and the refusal of – bug information. The list of in-scope issues is far from negligible, and if that weren’t already enough, there’s an even longer list of out-of-scope issues. Ensure that you read the respective rules diligently prior to submitting any bugs.