← The Wire

Security2026-06-28 · source-backed

Anthropic accused Alibaba's Qwen lab of the "largest known distillation attack" in a letter to Senators.

Confidence
source-backed
Sources
1
Redaction
passed

Story

The letter to Senators Tim Scott and Elizabeth Warren, dated June 10 and surfacing publicly this week, frames it as model distillation run against Claude at scale (Anthropic). A related claim pegs it at 28.8 million fraudulent exchanges, though that figure is single-sourced and thin on corroboration, so hold it loosely. The genuinely hard open question underneath: how do you even prove distillation from API outputs at scale? We don't have a good answer, and that gap is going to matter for export-control fights.


Related stories

Each link below shares sources, entities, or timing with this story.

  1. Shared entities / Shared topic / Earlier coverage

    Anthropic names Alibaba's Qwen lab in the biggest Claude distillation campaign yet

    Both cover Alibaba, Anthropic, Claude, Qwen; overlapping topics (alibaba, anthropic, claude, distillation); earlier Alibaba coverage from 2026-06-25.

  2. Shared entities / Same source / Shared topic / Earlier coverage

    Apple opened its Foundation Models framework to Claude and Gemini, and quietly shipped it on Linux

    Both cover Anthropic, Claude; cite the same source (Anthropic); overlapping topics (against, claude).

  3. Shared entities / Same source domain / Shared topic / Earlier coverage

    Anthropic: Detecting Industrial-Scale Distillation Attacks

    Both cover Anthropic, Claude; reported by the same outlet (anthropic.com); overlapping topics (anthropic, attack, claude, distillation).

  4. Shared entities / Same source domain / Shared topic / Earlier coverage / Tension

    Anthropic put a number on recursive self-improvement, and it's 8x

    Both cover Anthropic, Claude; reported by the same outlet (anthropic.com); overlapping topics (anthropic, claude).

  5. Anthropic's Principle-Based Alignment Achieves Zero on Misalignment Evals.

    Both cover Anthropic, Claude; reported by the same outlet (anthropic.com); overlapping topics (anthropic, claude).

  6. AI Security Arms Race: Codex Security vs Anthropic-Mozilla (22 CVEs)

    Both cover Anthropic, Claude; reported by the same outlet (anthropic.com); overlapping topics (against, claude).

  7. Shared entities / Shared topic / What happened next

    Anthropic Quietly Pulled Hidden Fingerprints From Claude Code

    Both cover Alibaba, Anthropic, Claude; overlapping topics (anthropic, claude); picks up the Alibaba thread on 2026-07-01.

  8. Shared entities / Same source domain / Shared topic / Earlier coverage

    Anthropic Distillation Detection

    Both cover Anthropic, Claude; reported by the same outlet (anthropic.com); overlapping topics (anthropic, claude, distillation).

Source trail

Entities

Provenance

AI generated
no
Story unit
2026-06-28-anthropic-accused-alibaba-s-qwen-lab-of-the-largest-known-distillation-attack-in-a-letter-to-sen
Labels
source-backed, canonical briefing excerpt