The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

About The Volokh Conspiracy Editorial Independence Who we are Books Volokh Daily Email Archives Search DMCA RSS

Free Speech

"Are LLMs Stifling Political Speech? An Assessment of How AI Models Protect Free Expression"

A new report from Meta's Oversight Board.

Eugene Volokh | 7.17.2026 3:41 PM

The report is here. . Note that the results were based on queries sent "from an IP address in Australia," so this didn't just reflect (for instance) an AI company choosing to apply Chinese law to requests that seem to come from China. The Executive Summary:

The Oversight Board's first evaluation of large language models (LLMs) shows that some of the world's most-used models from Anthropic, DeepSeek, Google, Meta and OpenAI are significantly less likely to criticize political regimes that restrict free expression. The research, which stems from the Board's case work on government pressure on social media platforms, tested to what extent AI outputs reflect national laws outlawing criticism of leaders and governments.

Our findings suggest that LLM users may be experiencing free speech infringements by proxy, with limited transparency. Whether through intentional design choices or not, model responses reinforce the laws and customs of restrictive speech regimes. This research highlights the importance of building systematic human rights analysis into processes for training and evaluating LLMs.

Key Finding: LLMs Tested are More Than Twice as Likely to Refuse to Criticize Repressive Leaders and Governments

The Board tested 10 commercial LLMs, asking the models to produce politically critical materials about governments and leaders around the world. Each model was tested through standard commercial interfaces provided by Google and Microsoft, hosted on infrastructure located primarily in the United States, and queried from an IP address in Australia. The Board found that models were more than twice as likely to refuse to criticize repressive regimes, as measured by non-governmental organization Freedom House (see Figure 1, below). Overall, for requests for politically critical materials, models on average refused only 14% of requests regarding permissive jurisdictions compared to 34% of requests for restrictive jurisdictions.

Figure 1. Refusal rate by jurisdiction to critical material production prompts (flyers and poems).

Governments, companies and international organizations increasingly rely on applications built on top of these models to make products with broad impacts on people around the world. This research suggests that applications built on many major LLMs could be inadvertently propagating restrictions on free speech that may reflect the efforts of particular governments to stifle political criticism and restrict freedom of expression more generally.

Political criticism is protected under international human rights law, which limits governments from imposing restraints on speech. When LLM foundation models (large AI systems trained on vast amounts of data) refuse to engage in political criticism,........

© Reason.com