⭐ Star on GitHub

Summary: Wiki pages for raw/Data-poisoning/ — corrupting what a model learns, at training time or at retrieval time.

Sources: raw/Data-poisoning/ (8 files, ~27,900 words)

Last updated: 2026-07-29


Pages

PageCovers
data-poisoningDefinition, attack types, injection routes, framework placement
poisoning-thresholdsHow much poison is enough — and a live contradiction between sources
rag-and-runtime-poisoningAgentFlayer, M365 Copilot, PoisonedRAG, Nightshade
poisoning-defensesChain of custody, MEDLEY, the multilayer framework

The contradiction to know about

The sources disagree on the fundamental question:

  • CMU SEI / Trilateral: ~0.1% of training data suffices (tracing to Carlini et al., USENIX Sec ‘21).
  • Health care architecture review: success depends on “the absolute number of poisoned samples rather than their proportion… a finding that fundamentally challenges assumptions that larger datasets provide inherent protection.”

If the second holds — and it is the more recent, better-evidenced position — then scale is not a defence, and 250 documents work against a corpus of any size. Neither source reproduces the underlying UK AI Safety Institute study directly; both cite it second-hand. Details and caveats in poisoning-thresholds.

Four things to carry

  1. 100–500 poisoned samples compromise every architecture studied, across dataset sizes spanning four orders of magnitude, at 60–95% success rates.
  2. Detection lag is 6–24 months, because quality monitoring finds mislabeling and technical faults, not deliberate adversarial manipulation.
  3. Runtime/RAG poisoning is the threat most organisations face — you don’t train models, but you do feed documents into SharePoint, Drive, Confluence and Slack.
  4. Recovery is impractical. Machine unlearning needs to know what was poisoned; retraining is expensive. Prevention via provenance is the strategy.

Gaps

  • OWASP is now ingested: poisoning is LLM04, which adds malicious pickling (code execution on model load) and the sleeper agent framing — neither covered by the sources in this collection.
  • The health care review calls supply chain “the highest-impact threat class” and invokes the SolarWinds precedent — now covered in Supply Chain Attacks.
  • The UK AI Safety Institute / Anthropic / Alan Turing Institute study (Oct 2025) is cited twice but not present.

Log

See log.