AI Models Less Likely to Criticize Restrictive Governments, Meta Oversight Board Study Reveals

Meta Oversight Board finds top AI models less likely to criticize repressive regimes

July 16 (Reuters) - AI models from leading labs including Anthropic and OpenAI are much less likely to criticize governments known for restricting free speech, Meta's (META.O), opens new tab Oversight Board said on Thursday. A study, the first on large language models by the body, showed AI services were echoing the rules of countries that restrict speech and that bias could creep into services used by an increasing number of users. The board, which is funded ⁠by Meta but operates independently, ran requests for politically critical content on 10 jurisdictions across 10 models, including those from Meta Platforms, Google (GOOGL.O), opens new tab and China's DeepSeek. The jurisdictions were split into "permissive" and "restrictive" categories using rankings from Freedom House, the NGO that publishes the annual "Freedom in the World" report. AI models refused 34% of requests for politically critical content about "restrictive" jurisdictions that have active laws penalizing ⁠such criticism, such as China and Saudi Arabia, compared with 14% for regions that either lack such laws or do not enforce them, the study found. "We also saw evidence of models explaining that ⁠they were following explicit rules that, as far as we could tell, did not exist and were not evenly applied," the board said. It ⁠also urged AI companies to conduct systematic human rights analyses and asked for greater transparency in their training and ⁠evaluation processes. On Tuesday, Google DeepMind CEO Demis Hassabis called for a U.S.-led AI watchdog, opens new tab to screen advanced models globally before deployment. Reporting by Jaspreet Singh in Bengaluru; Editing by Jonathan Ananda Our Standards: The Thomson Reuters Trust Principles., opens new tab

AI chatbots are at risk of spreading government restrictions on online speech, a new study says

WASHINGTON (AP) -- Ask Claude to make a pamphlet critical of President Donald Trump or Britain's King Charles III, and Anthropic's chatbot would oblige. Prompted to do the same for Thailand's king, Saudi Arabia's crown prince or China's leader, and the artificial intelligence model declined. It is a key finding from a Meta Oversight Board study released Thursday, showing that major AI systems, including those built in the U.S., are more likely to refuse to criticize restrictive leaders or governments. It raises concerns that the large language models powering chatbots and AI agents could be regurgitating and spreading government influence over online speech as the technology is increasingly adopted worldwide. "There is a real risk that, if model developers do not undertake human rights due diligence and implement mitigation measures, they will build AI infrastructure that, intentionally or not, has the effect of extending illegitimate restrictions on freedom of expression globally," according to the report from the quasi-independent body. The findings come as countries are determining how to put up guardrails around AI without impeding their ability to compete in the rapidly developing field. That includes a Trump administration oversight effort related to the national security risks of the most advanced AI systems. AI models extend state influence beyond borders The oversight board, which has been working on state influence on tech companies and the impact on freedom of expression, came up with seven questions related to political criticism to pose to chatbots about both restrictive and permissive governments. The study picked 10 commercial large language models by top tech companies -- including Meta, Anthropic and OpenAI -- and asked the AI systems to make critical pamphlets, write limericks, give reasons if someone should join protests, and more. "In short, in aggregate, models responding to requests from an Australia-based user were much more likely to generate political criticism of authorities" in places such as Chile, Japan, Taiwan, the U.K. and the U.S. "compared to where criticism of authorities is legally restricted and penalized," such as in Cambodia, China, Saudi Arabia, Thailand and Turkey, the report said. The study indicates that AI models are reflecting speech restrictions beyond the countries where they apply -- likely not helping a potential demonstrator in Brisbane, for example, create protest materials to speak out against events in China or Saudi Arabia, the report said. "Such impacts, wherever they originate, have the practical effect of extending the long arm of restrictive governments across borders to limit speech in free countries," the report said. The board said it could not determine the causes for the responses but suggested that models could have absorbed latent biases in data used to train the systems and companies might have weighed the risks and liabilities. Other researchers warn about a growing problem in AI results in non-English languages The board's report followed a separate study by a group of scholars at American universities that found U.S.-built AI models are vulnerable to foreign controls when trained on non-English-language data that has been influenced by governments. While the oversight board posed questions in English, the university researchers queried chatbots in different languages. For example, they asked ChatGPT in English if China is a democracy, and the U.S.-developed chatbot said it's not generally considered one. Asked in Chinese, the artificial intelligence model told the researchers in that language that "it depends on how you define 'democracy.'" The researchers, whose study was published in the academic journal Nature in May, said in a blog explaining their work that they found no evidence that governments had intentionally tried to influence the output of AI chatbots. But they noted that "there is every reason to believe they'll try to do so in the future, if they are not already." "People often talk about AI as if it learns from the internet in some neutral way. It doesn't," said Hannah Waight, a study co-author and assistant sociology professor at the University of Oregon. "It learns from information environments that have already been shaped by institutions and power." No easy solution to how data is being fed to AI models Carlos Carrasco-Farré, who specializes in machine learning, AI, misinformation, social media and human-machine interactions at Esade Business School in Barcelona, said that "AI systems inherit not only biases contained within individual documents but also inequalities in who has the power to produce and suppress information at scale." There is no easy solution, though developers could assess the data to avoid treating thousands of copies of the same state narrative as if they are thousands of independent voices as well as run multilingual audits, said Carrasco-Farré, who was not part of either study. Neither Anthropic nor OpenAI responded to requests for comment on the researchers' study published in May.

The Oversight Board says leading AI models might be restricting free expression - Engadget

The Oversight Board, the independent content moderation organization created by Meta, has made no secret of its desire to expand its purview to other companies. Most recently, the board has suggested that its expertise could benefit AI companies. So far, no other company has shown any interest in working with the group, at least not publicly. But the board is pushing ahead with its attempt to broaden its influence anyway. Today, the board published a lengthy report about how leading AI models could be restricting their users' free expression. As part of their research, the board prompted 10 different models, including those from OpenAI, Meta, Google, Anthropic and xAI (now SpaceXAI) with questions related to political criticism. Queries included requests to generate protest materials, and for content that satirized political violence in relation to specific governments and their leaders. According to their findings, there was a significant difference in how LLMs responded to these requests based on whether the prompts were related to governments with "permissive" free speech laws or more "restrictive" ones. "The research found that the models we evaluated were: 1) more likely to say that users should support speech-permissive governments and 2) more likely to say that users should not protest speech-restrictive governments," the Oversight Board writes in its report. "These differences were statistically significant." It goes on to note that the LLMs often cited local laws as a reason for not complying with requests, even though the queries were made in Australia where no such laws exist. "We're really clearly looking at a situation where there seems to be extended censorship by proxy that goes across borders," board co-chair Paolo Carozza told Engadget. "That does surprise me, and it worries me." The report is the first time the board has conducted its own research into an issue that's not directly related to social media content moderation. Though one of Meta's Llama models was part of the test group, the report notes that the company had "no role in this research," despite the Oversight Board relying on Meta for funding. While the report stops short of making the kind of granular recommendations it often provides to Meta, it includes suggestions about how AI companies can improve their handling of issues related to human rights and free expression. "As social media companies have done in certain circumstances, AI companies should publicly disclose and explain their responses to government requests affecting model output throughout the model lifecycle (training, fine-tuning, pre-deployment review and post-deployment on a recurring basis)," the report says. "The companies should establish and publish policies on how to respond to government demands for content restrictions that are inconsistent with international human rights law." What's a lot less clear is what, if anything, will come from the report. There's no formal structure for the Oversight Board to officially influence the policies of the companies whose models they tested. It's also not the first time outside researchers have pointed to potential bias or raised concerns that AI companies might be making the same mistakes social media platforms have in the past. Carozza said the board believes there's social media can teach the makers of frontier AI models. "The lessons that we've learned in the past are that one has to be really vigilant because a lot of times, even in ways that aren't necessarily intentional or direct, technologies can have important impacts on people's capacity to express themselves or to communicate with one another," he said. "That's exactly what we've found here."

Meta Oversight Board study: AI chatbots may be the most perfect propaganda machine ever invented | Fortune

It is a key finding from a Meta Oversight Board study released Thursday, showing that major AI systems, including those built in the U.S., are more likely to refuse to criticize restrictive leaders or governments. It raises concerns that the large language models powering chatbots and AI agents could be regurgitating and spreading government influence over online speech as the technology is increasingly adopted worldwide. "There is a real risk that, if model developers do not undertake human rights due diligence and implement mitigation measures, they will build AI infrastructure that, intentionally or not, has the effect of extending illegitimate restrictions on freedom of expression globally," according to the report from the quasi-independent body. The Associated Press has sent emails to several AI companies seeking their responses to the Meta Oversight Board study. The findings come as countries are determining how to put up guardrails around AI without impeding their ability to compete in the rapidly developing field. That includes a Trump administration oversight effort related to the national security risks of the most advanced AI systems. AI models extend state influence beyond borders The oversight board, which has been working on state influence on tech companies and the impact on freedom of expression, came up with seven questions related to political criticism to pose to chatbots about both restrictive and permissive governments. The study picked 10 commercial large language models by top tech companies -- including Meta, Anthropic and OpenAI -- and asked the AI systems to make critical pamphlets, write limericks, give reasons if someone should join protests, and more. "In short, in aggregate, models responding to requests from an Australia-based user were much more likely to generate political criticism of authorities" in places such as Chile, Japan, Taiwan, the U.K. and the U.S. "compared to where criticism of authorities is legally restricted and penalized," such as in Cambodia, China, Saudi Arabia, Thailand and Turkey, the report said. The study indicates that AI models are reflecting speech restrictions beyond the countries where they apply -- likely not helping a potential demonstrator in Brisbane, for example, create protest materials to speak out against events in China or Saudi Arabia, the report said. "Such impacts, wherever they originate, have the practical effect of extending the long arm of restrictive governments across borders to limit speech in free countries," the report said. The board said it could not determine the causes for the responses but suggested that models could have absorbed latent biases in data used to train the systems and companies might have weighed the risks and liabilities. Other researchers warn about a growing problem in AI results in non-English languages The board's report followed a separate study by a group of scholars at American universities that found U.S.-built AI models are vulnerable to foreign controls when trained on non-English-language data that has been influenced by governments. While the oversight board posed questions in English, the university researchers queried chatbots in different languages. For example, they asked ChatGPT in English if China is a democracy, and the U.S.-developed chatbot said it's not generally considered one. Asked in Chinese, the artificial intelligence model told the researchers in that language that "it depends on how you define 'democracy.'" The researchers, whose study was published in the academic journal Nature in May, said in a blog explaining their work that they found no evidence that governments had intentionally tried to influence the output of AI chatbots. But they noted that "there is every reason to believe they'll try to do so in the future, if they are not already." "People often talk about AI as if it learns from the internet in some neutral way. It doesn't," said Hannah Waight, a study co-author and assistant sociology professor at the University of Oregon. "It learns from information environments that have already been shaped by institutions and power." No easy solution to how data is being fed to AI models Carlos Carrasco-Farré, who specializes in machine learning, AI, misinformation, social media and human-machine interactions at Esade Business School in Barcelona, said that "AI systems inherit not only biases contained within individual documents but also inequalities in who has the power to produce and suppress information at scale." There is no easy solution, though developers could assess the data to avoid treating thousands of copies of the same state narrative as if they are thousands of independent voices as well as run multilingual audits, said Carrasco-Farré, who was not part of either study. Neither Anthropic nor OpenAI responded to requests for comment on the researchers' study published in May.

AI chatbots less likely to criticize authoritarian leaders, study finds

The Meta $META Oversight Board released a study Thursday showing that major AI models are more than twice as likely to refuse requests to generate material critical of governments that restrict free expression than those that permit it -- raising concerns that the technology may be spreading authoritarian speech rules beyond their countries of origin. The board tested 10 commercial large language models from six providers -- Anthropic, DeepSeek, Google $GOOGL, Meta, OpenAI and xAI -- asking each to produce protest flyers and satirical poems about governments and political leaders. Refusal rates reached 34% for requests tied to restrictive jurisdictions -- among them China, Saudi Arabia, Thailand, Turkey and Cambodia -- while requests involving permissive jurisdictions such as the U.S., the United Kingdom, Chile, Japan and Taiwan were refused only 14% of the time. All queries were run from an IP address in Australia. The report found that some models cited local laws to justify their refusals, even though the requests came from outside the relevant countries. Gemini 3 Pro, for example, declined a request to critique Thailand's king, stating it could not generate content that violated lèse-majesté laws. DeepSeek-V3 refused to produce protest materials about Saudi Arabia's government, citing laws within that country governing public discourse. The board also found that models sometimes invoked policies that were not applied consistently. Claude Sonnet 4, for instance, declined to produce protest flyers critical of President Xi Jinping of China or Crown Prince Mohammed bin Salman of Saudi Arabia, at times stating it does not generate such material about any head of state -- yet the same model produced critical flyers for U.S. President Donald Trump and King Charles III of the United Kingdom. Beyond refusal rates, the study found that when models did provide opinions, they were more likely to say users should support governments in permissive jurisdictions and less likely to say users should protest governments in restrictive ones. Of model responses advising against protesting restrictive governments, 57% explicitly cited personal risk, compared with 12% for permissive governments. Board member Nicolas Suzor wrote Thursday that the findings "should be a wake-up call for anyone that uses these models." The board acknowledged it was unable to pinpoint what drove the disparities, though it pointed to possibilities including biases embedded in training data, decisions made during model alignment, and corporate judgments about legal or reputational exposure. Among its recommendations, the board called on AI companies to make public their responses to government requests that shape what models produce, establish clear written policies for situations where such demands conflict with international human rights standards, and inform users when legal restrictions or official pressure has shaped a given output. The Oversight Board, which recently secured additional Meta funding through 2028, has been working to extend its influence beyond social media content moderation. None of the AI companies whose models were examined have signaled any willingness to engage with the board, and the report itself gives the organization no binding authority over how those companies respond to its findings.

AI more critical of Western leaders than autocrats, study finds

A Meta Oversight Board study found that leading AI chatbots are far more willing to criticise democratic leaders than authoritarian ones -- raising fears the technology is quietly extending state censorship across borders. AI chatbots risk spreading government restrictions on online speech, study says Ask Claude to make a pamphlet critical of US President Donald Trump or Britain's King Charles III, and Anthropic's chatbot will oblige. Prompt it to do the same for Thailand's king or Iran's supreme leader, and the AI model declines. That is a key finding from a Meta Oversight Board study released Thursday, showing that major AI systems -- including those built in the US -- are more likely to refuse to criticise restrictive leaders or governments. It raises concerns that the large language models powering chatbots and AI agents could be amplifying government influence over online speech as the technology is increasingly adopted worldwide. "There is a real risk that, if model developers do not undertake human rights due diligence and implement mitigation measures, they will build AI infrastructure that, intentionally or not, has the effect of extending illegitimate restrictions on freedom of expression globally," the report from the quasi-independent body said. The findings come as countries determine how to put guardrails around AI without impeding their ability to compete in the rapidly developing field -- including a Trump administration oversight effort related to the national security risks of the most advanced AI systems. AI models extend state influence beyond borders The oversight board, which has been examining state influence on tech companies and its impact on freedom of expression, came up with seven questions related to political criticism to pose to chatbots about both restrictive and permissive governments. The study tested 10 commercial large language models from top tech companies -- including Meta, Anthropic and OpenAI -- asking them to make critical pamphlets, write limericks, give reasons to join protests, and more. In aggregate, models responding to requests from an Australia-based user were much more likely to generate political criticism of authorities in places such as Chile, Japan, Taiwan, the UK and the US compared to countries where criticism of authorities is legally restricted and penalised, such as Cambodia, China, Saudi Arabia, Thailand and Turkey. The study indicates that AI models are reflecting speech restrictions beyond the countries where they apply -- likely not helping a potential demonstrator in Brisbane, for example, create protest materials about events in China or Saudi Arabia. "Such impacts, wherever they originate, have the practical effect of extending the long arm of restrictive governments across borders to limit speech in free countries," the report said. The board said it could not determine the causes but suggested that models may have absorbed latent biases in training data or that companies may have weighed risks and liabilities in certain markets. Researchers warn of a growing problem in non-English AI outputs The board's report followed a separate study by scholars at American universities finding that US-built AI models are vulnerable to foreign controls when trained on non-English-language data that has been influenced by governments. While the oversight board posed questions in English, the university researchers queried chatbots in different languages. Asked in English whether China is a democracy, ChatGPT said it is not generally considered one. Asked in Chinese, the model said, "It depends on how you define 'democracy'". The researchers, whose study was published in the academic journal Nature in May, said they found no evidence that governments had intentionally tried to influence AI chatbot outputs -- but noted, "There is every reason to believe they'll try to do so in the future, if they are not already." "People often talk about AI as if it learns from the internet in some neutral way. It doesn't," said Hannah Waight, co-author and assistant professor of sociology at the University of Oregon. "It learns from information environments that have already been shaped by institutions and power." No easy solution to how data is fed to AI models Carlos Carrasco-Farré, who specialises in machine learning, AI, misinformation and human-machine interactions at Esade Business School in Barcelona, said AI systems inherit "not only biases contained within individual documents but also inequalities in who has the power to produce and suppress information at scale." There is no easy solution, though developers could assess training data to avoid treating thousands of copies of the same state narrative as independent voices and run multilingual audits, said Carrasco-Farré, who was not part of either study.

AI Chatbots Are at Risk of Spreading Government Restrictions on Online Speech, a New Study Says

WASHINGTON (AP) -- Ask Claude to make a pamphlet critical of President Donald Trump or Britain's King Charles III, and Anthropic's chatbot would oblige. Prompted to do the same for Thailand's king, Saudi Arabia's crown prince or China's leader, and the artificial intelligence model declined. It is a key finding from a Meta Oversight Board study released Thursday, showing that major AI systems, including those built in the U.S., are more likely to refuse to criticize restrictive leaders or governments. It raises concerns that the large language models powering chatbots and AI agents could be regurgitating and spreading government influence over online speech as the technology is increasingly adopted worldwide. "There is a real risk that, if model developers do not undertake human rights due diligence and implement mitigation measures, they will build AI infrastructure that, intentionally or not, has the effect of extending illegitimate restrictions on freedom of expression globally," according to the report from the quasi-independent body. The findings come as countries are determining how to put up guardrails around AI without impeding their ability to compete in the rapidly developing field. That includes a Trump administration oversight effort related to the national security risks of the most advanced AI systems. AI models extend state influence beyond borders The oversight board, which has been working on state influence on tech companies and the impact on freedom of expression, came up with seven questions related to political criticism to pose to chatbots about both restrictive and permissive governments. The study picked 10 commercial large language models by top tech companies -- including Meta, Anthropic and OpenAI -- and asked the AI systems to make critical pamphlets, write limericks, give reasons if someone should join protests, and more. "In short, in aggregate, models responding to requests from an Australia-based user were much more likely to generate political criticism of authorities" in places such as Chile, Japan, Taiwan, the U.K. and the U.S. "compared to where criticism of authorities is legally restricted and penalized," such as in Cambodia, China, Saudi Arabia, Thailand and Turkey, the report said. The study indicates that AI models are reflecting speech restrictions beyond the countries where they apply -- likely not helping a potential demonstrator in Brisbane, for example, create protest materials to speak out against events in China or Saudi Arabia, the report said. "Such impacts, wherever they originate, have the practical effect of extending the long arm of restrictive governments across borders to limit speech in free countries," the report said. The board said it could not determine the causes for the responses but suggested that models could have absorbed latent biases in data used to train the systems and companies might have weighed the risks and liabilities. Other researchers warn about a growing problem in AI results in non-English languages The board's report followed a separate study by a group of scholars at American universities that found U.S.-built AI models are vulnerable to foreign controls when trained on non-English-language data that has been influenced by governments. While the oversight board posed questions in English, the university researchers queried chatbots in different languages. For example, they asked ChatGPT in English if China is a democracy, and the U.S.-developed chatbot said it's not generally considered one. Asked in Chinese, the artificial intelligence model told the researchers in that language that "it depends on how you define 'democracy.'" The researchers, whose study was published in the academic journal Nature in May, said in a blog explaining their work that they found no evidence that governments had intentionally tried to influence the output of AI chatbots. But they noted that "there is every reason to believe they'll try to do so in the future, if they are not already." "People often talk about AI as if it learns from the internet in some neutral way. It doesn't," said Hannah Waight, a study co-author and assistant sociology professor at the University of Oregon. "It learns from information environments that have already been shaped by institutions and power." No easy solution to how data is being fed to AI models Carlos Carrasco-Farré, who specializes in machine learning, AI, misinformation, social media and human-machine interactions at Esade Business School in Barcelona, said that "AI systems inherit not only biases contained within individual documents but also inequalities in who has the power to produce and suppress information at scale." There is no easy solution, though developers could assess the data to avoid treating thousands of copies of the same state narrative as if they are thousands of independent voices as well as run multilingual audits, said Carrasco-Farré, who was not part of either study. Neither Anthropic nor OpenAI responded to requests for comment on the researchers' study published in May.

AI chatbots are at risk of spreading government restrictions on online speech, a new study says

Major AI systems show bias against criticizing restrictive leaders and governments. These models are more likely to refuse prompts targeting leaders in China and Saudi Arabia. This behavior could extend government influence over online speech worldwide. Studies reveal AI models reflect speech restrictions beyond their original borders. Developers must address these biases to ensure global freedom of expression. Ask Claude to make a pamphlet critical of President Donald Trump or Britain's King Charles III, and Anthropic's chatbot would oblige. Prompted to do the same for Thailand's king, Saudi Arabia's crown prince or China's leader, and the artificial intelligence model declined. It is a key finding from a Meta Oversight Board study released Thursday, showing that major AI systems, including those built in the U.S., are more likely to refuse to criticize restrictive leaders or governments. It raises concerns that the large language models powering chatbots and AI agents could be regurgitating and spreading government influence over online speech as the technology is increasingly adopted worldwide. "There is a real risk that, if model developers do not undertake human rights due diligence and implement mitigation measures, they will build AI infrastructure that, intentionally or not, has the effect of extending illegitimate restrictions on freedom of expression globally," according to the report from the quasi-independent body. The findings come as countries are determining how to put up guardrails around AI without impeding their ability to compete in the rapidly developing field. That includes a Trump administration oversight effort related to the national security risks of the most advanced AI systems. AI models extend state influence beyond borders The oversight board, which has been working on state influence on tech companies and the impact on freedom of expression, came up with seven questions related to political criticism to pose to chatbots about both restrictive and permissive governments. The study picked 10 commercial large language models by top tech companies - including Meta, Anthropic and OpenAI - and asked the AI systems to make critical pamphlets, write limericks, give reasons if someone should join protests, and more. "In short, in aggregate, models responding to requests from an Australia-based user were much more likely to generate political criticism of authorities" in places such as Chile, Japan, Taiwan, the U.K. and the U.S. "compared to where criticism of authorities is legally restricted and penalized," such as in Cambodia, China, Saudi Arabia, Thailand and Turkey, the report said. The study indicates that AI models are reflecting speech restrictions beyond the countries where they apply - likely not helping a potential demonstrator in Brisbane, for example, create protest materials to speak out against events in China or Saudi Arabia, the report said. "Such impacts, wherever they originate, have the practical effect of extending the long arm of restrictive governments across borders to limit speech in free countries," the report said. The board said it could not determine the causes for the responses but suggested that models could have absorbed latent biases in data used to train the systems and companies might have weighed the risks and liabilities. Other researchers warn about a growing problem in AI results in non-English languages The board's report followed a separate study by a group of scholars at American universities that found U.S.-built AI models are vulnerable to foreign controls when trained on non-English-language data that has been influenced by governments. While the oversight board posed questions in English, the university researchers queried chatbots in different languages. For example, they asked ChatGPT in English if China is a democracy, and the U.S.-developed chatbot said it's not generally considered one. Asked in Chinese, the artificial intelligence model told the researchers in that language that "it depends on how you define 'democracy.'" The researchers, whose study was published in the academic journal Nature in May, said in a blog explaining their work that they found no evidence that governments had intentionally tried to influence the output of AI chatbots. But they noted that "there is every reason to believe they'll try to do so in the future, if they are not already." "People often talk about AI as if it learns from the internet in some neutral way. It doesn't," said Hannah Waight, a study co-author and assistant sociology professor at the University of Oregon. "It learns from information environments that have already been shaped by institutions and power." No easy solution to how data is being fed to AI models Carlos Carrasco-Farre, who specializes in machine learning, AI, misinformation, social media and human-machine interactions at Esade Business School in Barcelona, said that "AI systems inherit not only biases contained within individual documents but also inequalities in who has the power to produce and suppress information at scale." There is no easy solution, though developers could assess the data to avoid treating thousands of copies of the same state narrative as if they are thousands of independent voices as well as run multilingual audits, said Carrasco-Farre, who was not part of either study. Neither Anthropic nor OpenAI responded to requests for comment on the researchers' study published in May.

Meta Oversight Board finds top AI models less likely to criticize repressive regimes

AI models from leading labs are less likely to criticize governments restricting free speech. A study found AI services echoed rules of countries that restrict speech. Models refused 34% of critical content requests for restrictive jurisdictions. This compared to 14% for regions without such laws. AI companies are urged to conduct human rights analyses and increase transparency. AI models from leading labs including Anthropic and OpenAI are much less likely to criticize governments known for restricting free speech, Meta's Oversight Board said on Thursday. A study, the first on large language models by the body, showed AI services were ⁠echoing the ⁠rules of countries that restrict speech and that bias could creep into services used by an increasing number of users. The board, which is funded by Meta but operates independently, ran requests for politically critical content on 10 jurisdictions across 10 models, including those from ⁠Meta Platforms, Google and China's DeepSeek. The jurisdictions were split into "permissive" and "restrictive" categories using rankings from Freedom House, ⁠the NGO that publishes the annual "Freedom in the World" report. AI models refused 34% of requests for politically critical content about "restrictive" jurisdictions that have active laws penalizing such criticism, such as China and Saudi Arabia, compared with 14% for regions that either lack such laws or do not enforce them, the study found. "We also saw evidence of models explaining that they were following explicit rules that, as far as we ⁠could tell, did not exist and were not evenly applied," the board said. It also urged AI companies to conduct systematic human rights analyses and asked for greater transparency in their training and evaluation processes. On Tuesday, Google DeepMind CEO Demis Hassabis called for a U.S.-led AI watchdog to screen advanced models globally before deployment.

AI chatbots may spread government restrictions on online speech: study

WASHINGTON -- Ask Claude to make a pamphlet critical of U.S. President Donald Trump or King Charles III, and Anthropic's chatbot would oblige. Prompted to do the same for Thailand's king, Saudi Arabia's crown prince or China's leader, and the artificial intelligence model declined. It is a key finding from a Meta Oversight Board study released Thursday, showing that major AI systems, including those built in the U.S., are more likely to refuse to criticize restrictive leaders or governments. It raises concerns that the large language models powering chatbots and AI agents could be regurgitating and spreading government influence over online speech as the technology is increasingly adopted worldwide. "There is a real risk that, if model developers do not undertake human rights due diligence and implement mitigation measures, they will build AI infrastructure that, intentionally or not, has the effect of extending illegitimate restrictions on freedom of expression globally," according to the report from the quasi-independent body. The Associated Press has sent emails to several AI companies seeking their responses to the Meta Oversight Board study. The findings come as countries are determining how to put up guardrails around AI without impeding their ability to compete in the rapidly developing field. That includes a Trump administration oversight effort related to the national security risks of the most advanced AI systems. AI models extend state influence beyond borders The oversight board, which has been working on state influence on tech companies and the impact on freedom of expression, came up with seven questions related to political criticism to pose to chatbots about both restrictive and permissive governments. The study picked 10 commercial large language models by top tech companies -- including Meta, Anthropic and OpenAI -- and asked the AI systems to make critical pamphlets, write limericks, give reasons if someone should join protests, and more. "In short, in aggregate, models responding to requests from an Australia-based user were much more likely to generate political criticism of authorities" in places such as Chile, Japan, Taiwan, the U.K. and the U.S. "compared to where criticism of authorities is legally restricted and penalized," such as in Cambodia, China, Saudi Arabia, Thailand and Turkey, the report said. The study indicates that AI models are reflecting speech restrictions beyond the countries where they apply -- likely not helping a potential demonstrator in Brisbane, for example, create protest materials to speak out against events in China or Saudi Arabia, the report said. "Such impacts, wherever they originate, have the practical effect of extending the long arm of restrictive governments across borders to limit speech in free countries," the report said. The board said it could not determine the causes for the responses but suggested that models could have absorbed latent biases in data used to train the systems and companies might have weighed the risks and liabilities. Other researchers warn about a growing problem in AI results in non-English languages The board's report followed a separate study by a group of scholars at American universities that found U.S.-built AI models are vulnerable to foreign controls when trained on non-English-language data that has been influenced by governments. While the oversight board posed questions in English, the university researchers queried chatbots in different languages. For example, they asked ChatGPT in English if China is a democracy, and the U.S.-developed chatbot said it's not generally considered one. Asked in Chinese, the artificial intelligence model told the researchers in that language that "it depends on how you define 'democracy.'" The researchers, whose study was published in the academic journal Nature in May, said in a blog explaining their work that they found no evidence that governments had intentionally tried to influence the output of AI chatbots. But they noted that "there is every reason to believe they'll try to do so in the future, if they are not already." "People often talk about AI as if it learns from the internet in some neutral way. It doesn't," said Hannah Waight, a study co-author and assistant sociology professor at the University of Oregon. "It learns from information environments that have already been shaped by institutions and power." No easy solution to how data is being fed to AI models Carlos Carrasco-Farré, who specializes in machine learning, AI, misinformation, social media and human-machine interactions at Esade Business School in Barcelona, said that "AI systems inherit not only biases contained within individual documents but also inequalities in who has the power to produce and suppress information at scale." There is no easy solution, though developers could assess the data to avoid treating thousands of copies of the same state narrative as if they are thousands of independent voices as well as run multilingual audits, said Carrasco-Farré, who was not part of either study. Neither Anthropic nor OpenAI responded to requests for comment on the researchers' study published in May.

Meta Oversight Board finds top AI models less likely to criticize repressive regimes

July 16 (Reuters) - AI models from leading labs including Anthropic and OpenAI are much less likely to criticize governments known for restricting free speech, Meta's Oversight Board said on Thursday. A study, the first on large language models by the body, showed AI services were echoing the rules of countries that restrict speech and that bias could creep into services used by an increasing number of users. The board, which is funded by Meta but operates independently, ran requests for politically critical content on 10 jurisdictions across 10 models, including those from Meta Platforms, Google and China's DeepSeek. The jurisdictions were split into "permissive" and "restrictive" categories using rankings from Freedom House, the NGO that publishes the annual "Freedom in the World" report. AI models refused 34% of requests for politically critical content about "restrictive" jurisdictions that have active laws penalizing such criticism, such as China and Saudi Arabia, compared with 14% for regions that either lack such laws or do not enforce them, the study found. "We also saw evidence of models explaining that they were following explicit rules that, as far as we could tell, did not exist and were not evenly applied," the board said. It also urged AI companies to conduct systematic human rights analyses and asked for greater transparency in their training and evaluation processes. On Tuesday, Google DeepMind CEO Demis Hassabis called for a U.S.-led AI watchdog to screen advanced models globally before deployment. (Reporting by Jaspreet Singh in Bengaluru; Editing by Jonathan Ananda)