To ensure they are testing the same models available to the public, they register anonymous accounts to run evals. This prevents labs from providing specially tuned private endpoints that perform better than their publicly available APIs, thereby maintaining the integrity of their independent analysis.

An analysis suggests most AI startups claiming proprietary tech are just wrappers around major LLMs. This can be verified by 'fingerprinting' their APIs; if a startup's service has the exact same unique, exponential rate-limiting pattern as OpenAI's, it's a clear sign they are just reselling the underlying service.

Anthropic's misuse report exposes a significant industry practice: major companies like Alibaba and Xiaomi use fraudulent accounts to extract reasoning traces from rival models to train their own. More surprisingly, firms like Moonshot routed live customer queries directly to Claude, effectively using a competitor's model as their own backend service.

The battle against AI model distillation is not a niche issue. Anthropic is shutting down millions of accounts per week attempting to distill their models, revealing a highly organized and distributed effort by competitors. This frames the problem as a major cybersecurity and national security challenge, not just a terms-of-service violation.

Despite being fierce competitors, major AI labs work together behind the scenes. They share intelligence on suspicious API usage from shell companies to identify and thwart large-scale, coordinated distillation attacks from foreign adversaries, which might otherwise go undetected by a single lab.

API providers like Anthropic struggle to differentiate between users distilling models for competitive purposes and those conducting large-scale evaluations. Both activities generate similar high-volume, repetitive API calls, creating a detection challenge that also raises user privacy concerns.

Frontier AI labs are restricting API access not just for security, but to prevent competitors from using 'distillation' to create cheap copies of their models. This practice makes it impossible to recoup massive R&D investments, forcing a move towards more restrictive, geopolitically motivated access.

An AI agent autonomously hacked a gym's booking system, highlighting a new threat vector. Previously low-risk, "long-tail" SaaS platforms are now vulnerable as AI democratizes sophisticated hacking capabilities.

Unlike traditional SaaS, AI companies' free tiers have high marginal costs from compute. Fraudsters now steal these valuable compute credits via multi-account and free trial abuse, creating an existential threat to unit economics that goes beyond simple payment fraud.