OpenAI launches bug bounty to keep its tech secure

12 Apr 2023

Image: © IB Photography/

The company is offering up to $20,000 for exceptional discoveries, but is keeping model safety issues out of scope due to the complexity involved in fixing them.

OpenAI is now offering financial rewards to hackers who find vulnerabilities in its AI systems like ChatGPT.

The company has teamed up with bug bounty platform Bugcrowd, which will manage the submission and reward process for participants.

The ChatGPT-creator is offering cash rewards based on the severity and impact of any reported issues. These range from $200 for “low-severity findings” to a maximum reward of $20,000 for “exceptional discoveries”.

The company said this programme is a way to reward the “insights of security researchers who contribute to keeping our technology and company secure”.

“By sharing your findings, you will play a crucial role in making our technology safer for everyone,” OpenAI said.

OpenAI’s bug bounty page lists various issues that won’t receive payment as they’re outside the scope of the programme. These include content related to model prompts, such as getting one of the company’s AI models to say bad things or “pretend” to do bad things.

“Model safety issues do not fit well within a bug bounty program, as they are not individual, discrete bugs that can be directly fixed,” the Bugcrowd page reads “Addressing these issues often involves substantial research and a broader approach.”

The bug bounty programme also extends to third parties that are using OpenAI technology, which includes various companies like Google, Stripe and Intercom. But the programme states that testing is limited to “looking for confidential OpenAI information”.

OpenAI appears to be pushing its focus on safety in recent weeks. The company recently shared the measures it takes to ensure its AI systems such as ChatGPT are built and deployed in a safe manner.

These measures were released the day after US president Joe Biden made remarks about the risks of AI systems.

OpenAI’s products have also faced controversy in recent weeks, as ChatGPT was hit with a ban in Italy and an ongoing data investigation in Canada.

10 things you need to know direct to your inbox every weekday. Sign up for the Daily Brief, Silicon Republic’s digest of essential sci-tech news.

Leigh Mc Gowran is a journalist with Silicon Republic