1.8 · Legality of data

Explore 2 source-supported AI security cases for 1.8 · Legality of data in Databricks DASF 3.0, with evidence, scope, and defensive analysis.

Risk category

Training data is used without adequately addressing ownership, intellectual-property restrictions, or applicable legal requirements, including deletion requests. Removing a source record does not necessarily remove its influence from a trained model because training encodes information in parameters and other internal representations. Tracking the data used and, where needed, retraining on verified lawful data are therefore part of managing this risk.

Official framework source ↗ · Open interactive explorer ↗

Incident · · Practical overlap

Model Distillation Campaigns Target Google AI Services

Google reported campaigns that used compromised credentials, fraudulent accounts and proxies to collect model outputs for distillation: training another model from a provider’s responses. A February 2026 reasoning-focused campaign exceeded 100,000 prompts; later campaigns included some above 100 million. Those are campaign query volumes, not model weights stolen or a measured replication success rate. Google says it disrupted accounts and reduced the usefulness of collected outputs.

Why it relates: The provider explicitly reports collecting model outputs for unauthorized competitor training in violation of its terms. The shared issue is whether the training-data user has permission to use those outputs; lawful distillation of owned or authorized models is a different situation. This is the provider-reported contractual and intellectual-property concern about training inputs. It is not a court ruling, a finding about privacy-law deletion rights, or a claim that all distillation is unlawful.

Google’s compromised-account and fraudulent-account campaigns; capability distillation is distinguished from stealing original weights. The prompt totals are collection volumes, not measured replication success or model weights. No autonomous agent deployment or protected-reasoning recovery is established.

SecureFlow · Read the full case ↗

Incident · · Practical overlap

Model Distillation Campaigns Targeting Anthropic Claude

Anthropic attributed more than 16 million exchanges through roughly 24,000 fraudulent accounts to DeepSeek, Moonshot AI and MiniMax. The flow distinguishes proxy access, authenticated accounts, distributed routing, inference-response collection and downstream model training. Provider-side hunting and verified limits act before outputs leave; potential harms from models lacking safeguards remain Anthropic’s risk assessment, not observed downstream victim counts.

Why it relates: The provider explicitly reports collecting model outputs for unauthorized competitor training in violation of its terms. The shared issue is whether the training-data user has permission to use those outputs; lawful distillation of owned or authorized models is a different situation. This is the provider-reported contractual and intellectual-property concern about training inputs. It is not a court ruling, a finding about privacy-law deletion rights, or a claim that all distillation is unlawful.

Anthropic’s DeepSeek, Moonshot AI and MiniMax campaigns; capability distillation is distinguished from stealing original weights. The 24,000-account and 16-million-exchange figures do not quantify recovered weights or observed downstream harms. Potential harms remain Anthropic’s assessment; an agent deployment is not established.

SecureFlow · Read the full case ↗