While it has in-house research and engineering teams to test its AI systems, it now turns to the community to help spot vulnerabilities and security flaws that could be critical to its proper functioning. And for this, it has opened a rewards program.
OpenAI opens a rewards program to detect vulnerabilities in its AI systems
We have already seen that many technology companies resort to the famous bounty programs, so that experts from the community can detect errors or serious vulnerabilities in their systems and products.
A dynamic that Open AI is now implementing so that those interested can look for these vulnerabilities in their systems, for example, in ChatGPT:
[…] we invite the global community of security researchers, ethical hackers, and technology enthusiasts to help us identify and address vulnerabilities in our systems.
For this, OpenAI has the collaboration of Bugcrowd, a cybersecurity company that implements this work dynamic to identify important vulnerabilities. So everything will be managed through this company. And of course, in return, users who report critical bugs will be rewarded with money.
How to participate in the OpenAI bounty program
This dynamic that is established through the bounty program not only applies to ChatGPT, but also to the OpenAI API, plugins, and other related services.
And of course, when we talk about vulnerabilities, we don’t mean weird things you might encounter in your interaction with ChatGPT, like what happens with the ChatGPT Diablo prompt. It refers to real problems or potential vulnerabilities that could affect OpenAI systems.
The rewards that OpenAI gives to users who contribute by reporting these bugs will depend on a number of factors, for example, the impact of the bug, the priority of the vulnerability, or the criticality of the situation. And as they have expressed, the incentives range from 200 to 20,000 dollars.
In this link you will find a long list of requirements and details to take into account if you want to participate in this rewards program. It is important that you read this carefully, as OpenAI is very specific about what is allowed in this program.
Let’s remember that OpenAI has had several shocks with ChatGPT for exposing information, such as the personal data of a small group of subscribers or the history of chats that were exchanged between users. And not forgetting to mention the blocking that ChatGPT is receiving in some countries.
