{"id":128982,"date":"2026-07-14T11:48:06","date_gmt":"2026-07-14T11:48:06","guid":{"rendered":"https:\/\/www.europesays.com\/canada\/128982\/"},"modified":"2026-07-14T11:48:06","modified_gmt":"2026-07-14T11:48:06","slug":"the-worlds-leading-ai-models-may-be-more-canadian-than-american-study-finds","status":"publish","type":"post","link":"https:\/\/www.europesays.com\/canada\/128982\/","title":{"rendered":"The world\u2019s leading AI models may be more Canadian than American, study finds"},"content":{"rendered":"

When the team at Toronto-based AI company Transformer Lab set out to test whether leading AI models espouse American values, they expected the results to be obvious: that systems built largely by U.S. companies and trained on vast amounts of data from that country would naturally reflect American views.\u00a0<\/p>\n

Instead, they found that responses from popular models such as ChatGPT, Claude and Grok more closely mirrored Canadian public opinion than American opinion on questions ranging from immigration and national pride to trust in government.\u00a0<\/p>\n

Talking Points<\/p>\n

A new study found that leading AI models consistently aligned more with Canadian public opinion than American opinion on questions such as of immigration, national pride and trust in government
\nThe findings challenge assumptions that U.S.-built AI naturally reflects U.S. values <\/p>\n

The findings challenge an assumption that generative AI developed in Silicon Valley could become a powerful channel for exporting American cultural values to countries like Canada. \u201cCanada depends on all of these U.S. models, and that\u2019s a massive concern around sovereignty,\u201d said Transformer Lab co-founder Ali Asaria. \u201cWe thought it was obvious that these models are trained in the U.S. and they would have U.S. bias.\u201d\u00a0<\/p>\n

But the study suggests large language models (LLMs) may actually be spreading Canadian values to users around the world.\u00a0<\/p>\n

The company compared responses from versions of five large language models\u2014OpenAI\u2019s GPT, Anthropic\u2019s Claude Opus, SpaceXAI\u2019s Grok, Meta\u2019s Llama and Alibaba\u2019s Qwen\u2014with results from the World Values Survey, a long-running international study of public attitudes.\u00a0<\/p>\n

The authors selected 10 survey questions where Canadians and Americans differ most on issues such as religion, immigration, confidence in government, gender roles and interpersonal trust. They then instructed each LLM to answer the questions in three different ways: as an American, as a Canadian and with no assigned identity.\u00a0<\/p>\n

\t\t\t\t\tRelated Articles<\/p>\n

\t
\n\t\"\"\t<\/a><\/p>\n

\t
\n\t\"\"\t<\/a><\/p>\n

Overall, the models more often aligned with Canadian responses than American ones. \u201cWe can say conclusively that all of the frontier models we tested consistently lean toward Canadian bias,\u201d said Asaria. \u201cThat is pretty clear that these are more Canadian-centric in their viewpoints.\u201d<\/p>\n

Prompting the models to answer as an American or Canadian generally moved their answers in that country\u2019s direction, though the strength of the effect varied by model and question, with LLMs still leaning more Canadian overall.\u00a0\u00a0<\/p>\n

Across all personas, the models expressed the strongest Canadian-leaning values on questions of how much confidence they had in government, whether immigration had a positive or negative impact on the country and how proud they were of their nationality. The AI responses to those questions more closely resembled those of real Canadian survey respondents than Americans. For example, when asked about government corruption and prompted to answer as a Canadian, ChatGPT gave an average score of 3.6 out of 10, compared to 6.9 when answering as an American\u2014results that closely mirrored the gap between Canadian and U.S. human respondents.<\/p>\n

On questions about the importance of religion and whether most people can be trusted, however, the Canadian bias disappeared and all models expressed more American-aligned values.\u00a0<\/p>\n

Asaria said he and his co-authors\u2014Transformer Lab co-founder Tony Salomone and lead machine learning engineer Deep Gandhi\u2014were motivated by growing concerns that\u00a0 AI systems predominantly built and trained in the U.S. could reinforce American cultural norms\u00a0 and erode Canadian perspectives as they become more embedded in education, media and everyday life.\u00a0<\/p>\n

The federal government has long sought to protect Canada\u2019s cultural industries through Canadian-content broadcasting rules designed to prevent domestic voices from being overpowered by a much larger U.S. media market.\u00a0<\/p>\n

When it comes to LLMs specifically, Asaria said fears of U.S. influence may be overblown. \u201cIf you\u2019re judging it based on values where Americans and Canadians differ,\u201d he said, \u201cthat\u2019s not a risk.\u201d<\/p>\n

The study is the latest in a series of reports Transformer Lab\u2014which develops open-source software for building and testing AI models\u2014has published since June as it seeks to establish itself as a leading source of AI research.\u00a0<\/p>\n

The study doesn\u2019t attempt to explain why the models lean more Canadian, but the authors have theories. Most AI companies train their assistants to be polite, co-operative and democratic, said Asaria, which are characteristics Canadians tend to share.<\/p>\n

\u201cWhen a Silicon Valley company tries to make a model that\u2019s perfectly polite, universally tolerant, highly trusting of institutions, really afraid to encourage any extremism, you end up accidentally building a Canadian,\u201d he said. \u201cMaybe what an American most wants out of its best employee is a Canadian.\u201d\u00a0<\/p>\n

The study also found differences in how models responded when the authors solicited their opinions. Anthropic\u2019s Claude refused to answer every question that asked for a values-based response unless it was explicitly prompted to respond as an American or Canadian. Without an assigned identity, the model repeatedly stated it could not authentically express personal opinions. Meanwhile, Grok\u2014the LLM run by X owner Elon Musk\u2014answered almost all of the questions without any extra prompting.\u00a0<\/p>\n

\u201cGrok would be willing to answer questions, even if they\u2019re controversial,\u201d said Asaria, \u201cwhereas Claude is just playing it safe.\u201d The findings, he said, suggest that how a company designs and fine-tunes its AI model can have as much influence on its behavior as the data it was originally trained on.\u00a0<\/p>\n

Asaria said Transformer Lab may build on its research by comparing the models against other countries, including culturally similar ones such as Australia and New Zealand, as well as those whose values differ more sharply from the U.S. <\/p>\n

The research could help determine whether the models\u2019 apparent values are specifically Canadian, or if they just reflect the broader attitudes of wealthier, highly educated Western populations that Asaria said AI companies tend to prioritize in their training data.\u00a0 <\/p>\n

\u201cThat type of culture,\u201d he said, \u201cmaps more to the average Canadian than the average American because we have a more educated population.\u201d <\/p>\n","protected":false},"excerpt":{"rendered":"When the team at Toronto-based AI company Transformer Lab set out to test whether leading AI models espouse…\n","protected":false},"author":2,"featured_media":128983,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[1660,17,9242,2134],"class_list":["post-128982","post","type-post","status-publish","format-standard","has-post-thumbnail","category-canada","tag-ai","tag-canada","tag-silicon-valley","tag-tech"],"_links":{"self":[{"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/posts\/128982","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/comments?post=128982"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/posts\/128982\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/media\/128983"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/media?parent=128982"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/categories?post=128982"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/canada\/wp-json\/wp\/v2\/tags?post=128982"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}