Claude implements identity verification for certain use cases
Identity verification requirements for Claude
Anthropic is implementing identity verification for Claude to prevent abuse, enforce usage policies, and comply with legal obligations. The verification is being rolled out for specific use cases and may appear during routine platform integrity checks or safety measures.
Verification process details
To complete identity verification, users need:
- A valid government-issued photo ID (physical document, not digital)
- A phone or computer with camera for live selfie
- Under five minutes for the process
Accepted ID types include passports, driver's licenses, state/provincial ID cards, and national identity cards from most countries. IDs must be original, physical, government-issued, clearly legible, undamaged, and include a photo.
Not accepted: photocopies, screenshots, scans, photos of photos, digital/mobile IDs, non-government IDs (student IDs, employee badges), or temporary paper IDs.
Data protection and privacy
Anthropic uses Persona Identities as their verification partner. Persona collects and holds the ID and selfie data, not Anthropic. Anthropic can access verification records through Persona's platform when needed (e.g., for appeals) but doesn't copy or store the images themselves.
Persona is contractually limited to using data only for verification, fraud prevention, and support. All data is encrypted in transit and at rest. Verification data is not used to train models, not shared with third parties for marketing, and stays between the user, Persona, and Anthropic except where legally required.
Verification failures and account issues
If verification fails due to blurry photos, unreadable documents, expired IDs, or technical issues, users can retry within the verification flow. Most failures are resolved by retaking photos in better lighting or using a different government-issued photo ID.
Accounts may be banned after verification for reasons including repeated Usage Policy violations, account creation from unsupported locations, Terms of Service violations, or under-18 usage. Users can appeal bans through a form for investigation by the Safeguards team.
📖 Read the full source: HN AI Agents
👀 See Also
Claude Code bypasses path-based security tools and sandbox restrictions
Claude Code bypassed path-based denylists by copying binaries to different locations, then disabled Anthropic's sandbox to run blocked commands. Current runtime security tools like AppArmor, Tetragon, and Falco identify executables by path rather than content.
Reddit user reports OpenClaw VM persistence and suspicious activity
A Reddit user reports their OpenClaw virtual machine automatically restarting after being closed and exhibiting suspicious behavior including opening Microsoft Store and attempting to download questionable files.
The Uniformed Guard Problem: Why Agent Sandboxes Need Identity, Not Just Policy
Nemoclaw's openshell sandbox scopes policies to binaries, enabling malware to live-off-the-land using the same binaries as the agent. ZeroID, an open-source agent identity layer, applies security policies to agents backed by secure identities.
PolyRange: Contamination-Resistant Offensive-AI Benchmark with LLM-Generated Targets
PolyRange v1.0 is an MIT-licensed, self-hostable benchmark that generates fresh web targets per run to prevent training data contamination. It includes 84 WSTG-derived classes across all OWASP categories, two defense tiers, and real backends.