Today’s AI models have a poor grasp of world history

May Be Interested In:Trump defends Gaza proposal, says territory would be ‘turned over’ to U.S. by Israel



Today’s AI models do a poor job of providing accurate information about world history, according to a new report from the Austrian research institute Complexity Science Hub (CSH).

In an experiment, OpenAI’s GPT-4, Meta’s Llama, and Google’s Gemini were asked to answer yes or no to historical questions — and only 46% of the answers were correct. GPT-4, for example, answered “yes” to the question of whether Ancient Egypt had a standing army, likely because the AI model chose to extrapolate data from other empires such as Persia.

“If you are told A and B 100 times and C one time, and then asked a question about C, you might just remember A and B and try to extrapolate from that,” researcher Maria del Rio-Chanona told Techcrunch.

share Share facebook pinterest whatsapp x print

Similar Content

Former Planned Parenthood president, activist Cecile Richards dies at 67
Former Planned Parenthood president, activist Cecile Richards dies at 67
FIFA’s Saudi Arabia World Cup host pick sparks ‘sportswashing’ accusations by rights groups
FIFA’s Saudi Arabia World Cup host pick sparks ‘sportswashing’ accusations by rights groups
Video: From Fighters to Policemen in a Post-Assad Damascus
Video: From Fighters to Policemen in a Post-Assad Damascus
Middle East Tensions Highlights: Trump Officials Try to Walk Back Gaza Takeover Plan
Middle East Tensions Highlights: Trump Officials Try to Walk Back Gaza Takeover Plan
In one fell swoop, Meta enters its ‘post-truth’ era
In one fell swoop, Meta enters its ‘post-truth’ era
Ousted Syrian President Bashar al-Assad's wife Asma files for divorce in Russia, claims she is unhappy with...
Ousted Syrian President Bashar al-Assad’s wife Asma files for divorce in Russia, claims she is unhappy with…
Trending Stories: What's Shaping the World Today | © 2025 | Daily News