Airlock launches human authorization product for AI agents
Understand this story
Airlock has launched an enterprise-grade "human-in-the-loop" authorization product for AI agents, along with the HARP (Human Authorization & Review Protocol) specification. Its process is: after an agent generates a sensitive command, it is intercepted by the Host Enforcer and submitted encrypted with AES-256-GCM to a zero-knowledge Gateway; a human reviews it via a mobile Approver and then signs with Ed25519; the command executes only if the signature is verified locally. Airlock says the product is deployed within the customer's own infrastructure, and the mobile Approver is also deployed inside the organization; authorized reviewers can receive push notifications, view encrypted content, and sign decisions on any approved device.
AI overview based on source reports · Updated 2 days ago
Coverage timeline
Follow the coverage to explore different perspectives.
- Hacker News AIHuman-in-the-loop authorization for AI agents
Airlock is a human-in-the-loop authorization system built on the HARP specification that intercepts sensitive AI agent commands, encrypts them with AES-256-GCM, and routes them to a mobile approver who signs a decision with Ed25519 that is verified locally before execution. It features a zero-knowledge gateway, replay protection via nonces and expiry timestamps, and enterprise deployment within an organization's own infrastructure. The HARP protocol is described as an open, tool-agnostic standard covering canonicalization, artifact hashing, decision signing, and replay protection.
Activity for this story
- Current comparable activity
- 3
- Peak comparable activity
- 10Nov 5 00:00
- Change in the past 24 hours
- -50%
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.