New challenge worth $15,000
Together we can make AI safer for everybody
Our research community has been recognised by some of the world’s largest AI organizations. Together, we are collaborating to continue to drive critical discoveries in open-source software that underpin AI and ML technologies.
Hugging Face
“At Hugging Face we take security seriously, as AI rapidly evolves, new threat vectors seemingly pop up every day.” Said Julien Chaumond, Co-Founder of Hugging Face. “We have been very impressed by the work Protect AI have been doing in the huntr community, coupled with the scanning capabilities of Guardian, they were an obvious choice to help our users responsibly experiment with and operationalize AI/ML systems and technologies”
500+
Vulnerabilities discovered in AI/ML (last 365 days)
70%
High severity vulnerabilities or above
15k+
Security researchers and open source maintainers
Spotlight
Here are a few high-profile vulnerabilities we've uncovered. Head over to our hacktivity for more.
High SEVERITY
Denial of Service via Large Character Payload in tensorflow/serving
Critical SEVERITY
Arbitrary file reading via SQL Injection in intel/neural-compressor
Critical SEVERITY
Deserialization of untrusted data leads to arbitrary code execution in apache/arrow
Critical SEVERITY
Distributed RPC Framework Remote Code Execution in pytorch/pytorch