Challenges with defined objectives
Each challenge provides rules and objectives for participants to review before they begin testing the AI.
AI security challenges for researchers to test guardrails and compete for cash prizes.
huntr is an AI/ML security platform where participants compete in timed hacking challenges. Its challenge workflow asks researchers to test AI defenses, meet defined objectives, and compete for leaderboard positions that determine cash prizes.
Participants choose a challenge, review its rules, and attempt tasks such as bypassing guardrails or safety filters. The site also publishes technical articles about AI exploits, LLM jailbreaks, and vulnerabilities in machine learning software.
Each challenge provides rules and objectives for participants to review before they begin testing the AI.
Participants work toward higher rankings by achieving challenge objectives before time runs out.
Challenges award real cash prizes to participants who finish in the top ranks.
The blog covers AI exploits, LLM jailbreaks, and zero-day intelligence, with articles on Keras model deserialization, archive slip bugs, and TensorFlow model exploits.
Use a challenge's objectives and rules to guide attempts to bypass AI guardrails and safety filters.
Work toward a ranked finish against other participants within a challenge's time limit, with cash prizes for top placements.
Read technical posts about model deserialization and other AI/ML exploits to prepare for future security challenges.
Choose a challenge, review its rules, and begin working toward its stated objectives.
The site describes tasks such as bypassing AI guardrails, outsmarting safety filters, and achieving challenge-specific objectives.
The stated prize workflow rewards participants who finish a challenge in the top ranks. Taking part alone does not guarantee winnings.
Os dados de tráfego são apenas para referência.