Hey PH community. I'm Jason, a solo founder building AIR Blackbox, an open-source CLI that scans Python AI projects for EU AI Act compliance.
Quick background: I spent 13+ years as a technical sourcer at Meta and AWS, where I watched companies move fast on AI adoption and slow on governance. When the EU AI Act passed with an August 2, 2026 enforcement deadline, I realized most teams had no idea what the technical requirements actually look like in code.
Your AI agents make thousands of calls but can you prove what they said? AIR Blackbox is an open-source reverse proxy that records every LLM call with HMAC-SHA256 chaining. If anyone tampers with a record, the chain breaks instantly.
Unlike observability tools, AIR focuses on provability:
- Cryptographic audit chain for tamper-proof integrity
- EU AI Act compliance (Articles 9-15)
- PII detection and prompt injection scanning
- Shadow AI detection via model registry
pip install air-blackbox
