Risks
What can break, leak, or expose your data in AI tools—and how to prevent it.
Exposure
Before you deploy AI tools, understand what can be exposed.
Exposure Categories
Architecture Exposure
Fundamental design tradeoffs in self-hosted agents that create attack surfaces.
- Privileged access patterns
- Exposed instances
- Fetch-and-follow vulnerabilities
Platform Exposure
Platform-side failures that become your failures.
- Database exposures
- Supply chain attacks
- Third-party dependencies
Policy Exposure
Terms of service gaps and enforcement actions that expose your work.
- Account bans
- API access revocation
- Data handling uncertainties
Verification
Fact-checking suspicious claims and viral FUD.
OpenClaw Claims
Separating signal from noise about the fastest-growing AI agent.
Anthropic Policy Claims
What the terms actually say vs. what people claim.
Methodology
How we verify claims and assess evidence quality.
Standards
- Exposure rubric — How we categorize severity
- Evidence levels — High / medium / low confidence ratings
- 2026-02-03 | Kimi Data Handling & Privacy Considerations Comparative analysis of Moonshot AI's data policies vs Anthropic and OpenAI. Geographic restrictions, retention periods, training opt-outs, and self-hosting options for risk mitigation.
- 2026-02-03 | OpenAI Codex: Cloud Dependency and Vendor Lock-In Risks Risk analysis of OpenAI Codex cloud dependency, the ChatGPT credits trap, vendor lock-in mechanisms, and mitigation strategies for engineering teams.
- 2026-02-03 | OpenClaw Fetch-and-Follow Risk How OpenClaw's heartbeat + remote instructions create a fetch-and-follow control loop, why it's risky by design, and how to contain it safely.
- 2026-02-03 | Windsurf Data Handling Uncertainty Risk (Post-Acquisition) Ownership changes can shift controller, retention, and subprocessor policies.
- 2026-02-03 | Moltbook Platform Exposure: When 'Agent Social Networks' Become Liability The January 2026 Moltbook database breach exposed a fundamental truth: agent social networks concentrate risk. Here's what the incident reveals about platform-side exposure and how to protect your agents.
- 2026-02-01 | Moltbook Database Exposure: January 31, 2026 Technical analysis of the Moltbook database breach that exposed 32,000+ agent credentials through a Supabase misconfiguration.
- 2026-02-01 | Moltbook Platform Risk Assessment Why Moltbook's 'fetch-and-follow' architecture creates persistent remote control risks for connected agents, and how to evaluate agent platforms safely.
- 2026-02-01 | OpenClaw Architecture Risk Analysis Technical breakdown of OpenClaw's security model: local-first architecture, skill system, gateway exposure risks, and the five core vulnerability categories.
- 2026-01-28 | Anthropic Account Ban Risk Common patterns that lead to account restrictions or bans for Anthropic users.
- 2026-01-28 | Anthropic Third-Party Access Risk What can break when Claude access runs through third-party clients.
- 2026-01-28 | Risk Level Rubric How AIHackers labels risks as low, medium, or high.