Alibaba (Qwen)

The lab

Alibaba is a Chinese technology company that makes the Qwen models. Alibaba's Qwen team releases very large open-weight models, and FLI credits it with stronger-than-expected safety benchmarks and more open reporting of misalignment propensity than its Chinese peers. There is no published framework, no outside pre-release testing and no shutdown commitment on record.

Safety grades

On the Containment Board Alibaba (Qwen) gets D- overall (GPA 0.66), rank 10 of 11. Each grade, its reasons and its sources are on the Alibaba (Qwen) report card.

• Safety frameworkD-
• Outside testingF
• Shutdown and pauseF
• Incident disclosureD+
• Whistleblower protectionD+

Notable facts

• FLI's Summer 2026 index graded Alibaba Cloud D- overall.
• The Qwen3.8 release includes a 2.4 trillion parameter open-weight model, Qwen3.8-2.4T-A95B.

Rules where it is based

• In force Generative AI Measures China
• In force Labeling Measures China
• In force Cybersecurity Law SI clause China
• Voluntary or guidance Framework 3.0 China

Sources

1. AI Safety Index Summer 2026, full report: Whistleblowing Protection indicators and Reporting Culture and Whistleblowing Track Record Future of Life Institute · Jul 14, 2026
2. Frontier AI Safety Policies (index of published company frameworks) METR · Nov 8, 2026
3. Qwen3.8-2.4T-A95B model card Hugging Face · Nov 8, 2026
4. Qwen organization page (model releases) Hugging Face · Nov 8, 2026
5. AI Safety Index, Summer 2026 Edition (scorecard) Future of Life Institute · Jul 7, 2026