Indonesian student reports security vulnerability in Claude AI
An Indonesian university student discovered a High-severity security vulnerability in Claude AI and reported it through Anthropic's bug bounty program on HackerOne. The company validated the finding and awarded the researcher a $3,700 bounty for the responsible disclosure.
What happened
Muhamad Arga Reksapati, a student at Universitas Teknologi Bandung and recipient of the Indonesian government's KIP Kuliah scholarship, discovered a security vulnerability in Claude AI through independent security analysis of how the system processes context, input, and supporting data.
Arga reported the vulnerability through Anthropic's official security reporting mechanism on HackerOne platform. After technical verification to confirm the vulnerability was reproducible and had real security impact, Anthropic validated the report with a High severity rating of 7.7.
Governance controls
Anthropic operates a bug bounty program through HackerOne, which accepted and validated Arga's security report.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.