Skip to content
IFAI

HackerOne

Company4 stories

  1. AI modelsReach 48

    Anthropic reveals Fable 5's cyber safeguards and jailbreak framework

    Anthropic just detailed the cybersecurity protections baked into its globally redeployed Claude Fable 5. The model relies on dedicated safety classifiers to sort tasks into four risk tiers, intentionally widening its safety margin to block high-risk dual-use actions like exploit creation. Alongside the update, Anthropic introduced a draft jailbreak severity framework and launched a HackerOne bug bounty program to crowdsource vulnerabilities.

  2. AI toolsReach 39

    OpenAI shifts AI security from finding bugs to auto-patching them

    OpenAI just expanded its Daybreak initiative, launching an updated Codex Security plugin alongside its specialized GPT-5.5-Cyber model. The cyber model reached 85.6% on the CyberGym benchmark, beating GPT-5.5's 81.8%. Why it matters: AI has made surfacing vulnerabilities trivial, overwhelming defenders with alerts—now OpenAI wants models to validate issues, draft targeted patches, and fix code at machine speed.

  3. IndustryReach 39

    Anthropic offers $25k to break Claude's guardrails

    Anthropic just partnered with HackerOne to launch an invite-only bug bounty program, offering up to $25,000 for verified universal jailbreaks against its unreleased safety classifiers. The initiative lets red teamers test Claude 3.7 Sonnet specifically against CBRN threats to meet its ASL-3 deployment standards. Why it matters: Stress-testing frontier models before public rollout is becoming an essential line of defense.

  4. IndustryReach 38

    Anthropic offers $15k bounties to crack its AI safeguards

    Anthropic just expanded its model safety bug bounty program in partnership with HackerOne, inviting security researchers to probe its unreleased next-generation safety mitigations. The initiative focuses on uncovering novel "universal jailbreaks" in high-risk domains like CBRN and cybersecurity, offering rewards of up to $15,000. It is currently invite-only, with applications open through August 16.