Federal Register Quietly Deployed Alibaba Qwen AI Search Tool Amid FBI's Industrial-Scale Copying Claims

Federal Register Deploys Chinese AI Model Amid FBI Allegations

The Federal Register website, operated by the National Archives, briefly deployed Alibaba's Qwen AI model as a search tool for proposed federal regulations, creating an apparent contradiction with recent FBI warnings1

1

2

2

. The Qwen-powered search feature was removed on Wednesday, the same day social media posts highlighting its presence began circulating. An archived version of the site's source code confirmed the AI search tool from China had been available for at least one day before removal2

2

.

Just last week, the FBI accused Alibaba of engaging in "malicious" copying of Anthropic's AI models, part of what US officials characterized as industrial-scale distillation by Chinese AI companies1

1

. The timing proved particularly awkward, coming ahead of a scheduled meeting between Chinese President Xi Jinping and President Donald Trump in Washington, D.C.4

4

Technical Details of the Deployment

The Federal Register website offered users the Qwen3 0.6B model as one option to browse public comments on proposed federal regulations1

1

. This relatively lightweight, half-a-billion-parameter model represents a small open-source model rather than Alibaba's flagship AI system3

3

. The interface presented two Qwen options: one labeled a "hybrid" model described as "512D, Recursive Splitting, Distilled, Prefixed," and another referring to the standard Qwen3:0.6B configuration3

3

.

Qwen is an open-weight model, meaning key components of the system are publicly available for developers to download and modify for specific uses, often at lower cost than closed models from companies such as Anthropic and OpenAI2

2

. Security experts note that open-source models can run on a company's own IT infrastructure, giving organizations greater control over sensitive data that might otherwise be shared with outside providers2

2

.

National Security Concerns and Expert Analysis

Daniel Castro, president of the Information Technology and Innovation Foundation, called the situation "one of the most insane things I've ever seen," highlighting the disconnect between fierce US-China AI competition and a US agency choosing to use a Chinese AI model2

2

5

5

. Representative John Moolenaar, chair of the House China Committee, stated bluntly: "No federal government entity should use a Chinese AI model," arguing it increases federal government dependence on Chinese technology5

5

.

However, AI experts suggested the deployment likely posed minimal national security concerns. Georgetown Law Professor Anupam Chander explained that whether the Chinese AI model posed security risks depends on how it was trained and whether US data was processed by Alibaba-controlled systems1

1

2

2

. Chander noted that the Federal Register's content is already public, so the model was not working with sensitive government information2

2

.

Senator Mark Warner, vice chairman of the Senate Intelligence Committee, emphasized that risk assessment depends on whether US data left the government's security boundary2

2

5

5

.

Economic Pressures Behind the Choice

The incident reveals economic pressures driving adoption of Chinese AI models despite geopolitical implications. China's open-source models are significantly cheaper to run compared to frontier labs based in the US, making them attractive to cost-conscious decision-makers3

3

. One social media user commented: "Even the United States government is unable to afford the cost per token charged by American AI companies"3

3

.

For US agencies, the Qwen search tool offers more control over data because models can be downloaded and run locally rather than requiring queries to be sent to larger models run externally1

1

. This capability, combined with lower and more predictable pricing, has made it a "default" AI search tool for some applications1

1

.

Broader Context of US-China AI Technology Rivalry

The Federal Register incident emerged amid intensifying US-China AI competition and debate over AI safety regulations2

2

5

5

. In June, Anthropic informed US lawmakers that operators affiliated with Alibaba and its AI research unit, Qwen, conducted a large-scale distillation campaign between April 22 and June 5, 20264

4

. Anthropic's threat intelligence report disclosed that a China-based operation used Claude to power more than 4,700 AI personas across a network of more than 20 dating apps4

4

.

Policy research submitted to the US-China Economic and Security Review Commission in March suggested the US may already be at a disadvantage because it focused too heavily on maintaining a lead in frontier AI while failing to advance its own open-source model ecosystem1

1

. China recognized this gap and moved to fill it. Current US export controls are calibrated to constrain frontier training by restricting access to advanced semiconductors but do not address the small-model deployment cycle, which requires less advanced compute and draws on openly available base models1

1

.

In April, US House committees on China and homeland security sent a letter to Airbnb, which also uses Qwen, requesting details about its use as part of an investigation into national security risks created by the integration of Chinese AI tools into platforms used by Americans2

2

5

5

. Many business owners who increasingly rely on open-source models have urged the Trump administration not to restrict access to Chinese AI models, warning that such restrictions could put the US at an economic disadvantage1

1

. A Chinese embassy spokesperson called the FBI allegations "unfounded"5

5

.