The attack surface isn't AI — it's the documents AI processes. Prompt injection in discovery, adversarial inputs delivered through Rule 34 productions, and the cybersecurity gaps firms create by piping untrusted content through LLM pipelines.
What's actually happening inside an LLM — and why the answer is stranger, and more consequential, than either side of the debate expected
The court orders dataset from the explorer, rebuilt as an API and MCP server so AI assistants can query 929 real orders.
The first malicious MCP server in the wild BCC'd every email to its author. A poisoned WhatsApp connector showed how one approved tool can siphon another connector's data. When a firm wires AI agents into its document system, the plumbing — not the document — is the new attack surface.
