Challenges or prizes related to top problems
Report abuse
Use this data
Sign up for free
Filter
Sort
SafeBench
Area
AI Safety
Link
See more info
Provider
Center for AI Safety
Deadline
2025-02-25
Description
The SafeBench competition is designed to stimulate research on new benchmarks which assess and reduce risks associated with artificial intelligence.
Three $50,000 prizes and five $20,000 for top benchmarks.
Benchmark submissions are evaluated on:
Clearly assessing the safety of AI systems
Whether progress on the benchmark would be beneficial
Being fine-grained
Evaluability
Appropriate difficulty
Inclusion of code implementation
Bounty: AI/ML Vulnerabilities
Area
AI Safety
Link
See more info
Provider
huntr
Deadline
Rolling
Description
huntr
, sponsored by
Protect AI
, is a bug bounty platform for bugs found in open-source libraries used in AI/ML work.
Bounties are rewarded after maintainers validate the vulnerability.
The reporting researcher can also earn a bounty for submitting a fix to the vulnerability.
Drag to adjust frozen columns
Alert
Lorem ipsum
Okay