OpenAI strengthens defense using AI and increasing rewards for found vulnerabilities
As AI capabilities in vulnerability detection grow, OpenAI is actively utilizing these technologies to strengthen its own defenses, staying ahead of potential attackers.
As part of an internal security sprint, the company is collaborating with researchers from Cathedral Systems. This collaboration allows for stress-testing and identifying weak points in OpenAI's systems using unique approaches.
Furthermore, OpenAI has significantly increased its bug bounty rewards. Critical vulnerabilities can now earn rewards of up to $172,000, which incentivizes the defender community. These new, increased rewards are also applied retroactively to reports submitted within the last 30 days.
Why it matters
- —Demonstrates a shift to proactive defense: instead of reacting to attacks, the company uses AI to search for its own weaknesses.
- —A significant increase in bug bounties attracts top global cybersecurity experts.
- —Confirms that AI development and ensuring its security is a priority and resource-intensive task.
Key facts
- OpenAI is conducting an internal security sprint with external experts.
- Collaboration with Cathedral Systems is underway to test defensive systems.
- Rewards for critical vulnerabilities have been increased to $172,000.
- New bug bounty rules apply to reports submitted within the last 30 days.
The full text is in the original source. Here we provide a brief summary and key facts.