Skip to main content Live Science Live Science Search Live Science View Profile Sign up to our newsletter Planet Earth Archaeology Physics & Math Human Behavior Science news Life's Little Mysteries Science quizzes Story archive Power plant on the moon Hidden source of clean energy Strongest solar flare of 2025 Milky Way visible across US Earth from space Recommended reading Artificial Intelligence AI can handle tasks twice as complex every few months. What does this exponential growth mean for how we use it? Artificial Intelligence Older AI models show signs of cognitive decline, study shows — but not everyone is entirely convinced Artificial Intelligence AI is just as overconfident and biased as humans can be, study shows Artificial Intelligence 'Math Olympics' has a new contender — Google's AI now 'better than human gold medalists' at solving geometry problems Artificial Intelligence AI creates better and funnier memes than people, study shows — even when people use AI for help Artificial Intelligence Traumatizing AI models by talking about war or violence makes them more anxious Artificial Intelligence AI models will lie to you to achieve their goals — and it doesn't take much Artificial Intelligence AI models can't tell time or read a calendar, study reveals Drew Turney 17 May 2025 Challenges in visual and spatial processing and a deficit in training data have revealed a surprising lack of timekeeping ability in AI systems When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works. AI systems read clocks correctly only 38.7% and calendars only 26.3% (Image credit: Alamy) New research has revealed another set of tasks most humans can do with ease that artificial intelligence (AI) stumbles over — reading an analogue clock or figuring out the day on which a date will fall. AI may be able to write code, generate lifelike images, create human-sounding text and even pass exams (to varying degrees of success) yet it routinely misinterprets the position of hands on everyday clocks and fails at the basic arithmetic needed for calendar dates. Researchers revealed these unexpected flaws in a presentation at the 2025 International Conference on Learning Representations (ICLR). They also published their findings March 18 on the preprint server arXiv, so they have not yet been peer-reviewed . You may like AI can handle tasks twice as complex every few months. What does this exponential growth mean for how we use it? Older AI models show signs of cognitive decline, study shows — but not everyone is entirely convinced AI is just as overconfident and biased as humans can be, study shows "Most people can tell the time and use calendars from an early age. Our findings highlight a significant gap in the ability of AI to carry out what are quite basic skills for people," study lead author Rohit Saxena, a researcher at the University of Edinburgh, said in a statement. These shortfalls must be addressed if AI systems are to be successfully integrated into time-sensitive, real-world applications, such as scheduling, automation and assistive technologies." To investigate AI's timekeeping abilities, the researchers fed a custom dataset of clock and calendar images into various multimodal large language models (MLLMs), which can process visual as well as textual information. The models used in the study include Meta's Llama 3.2-Vision, Anthropic's Claude-3.5 Sonnet, Google's Gemini 2.0 and OpenAI's GPT-4o. And the results were poor, with the models being unable to identify the correct time from an image of a clock or the day of the week for a sample date more than half the time. Related: Current AI models a 'dead end' for human-level intelligence, scientists agree Sign up for the Live Science daily newsletter now Get the world’s most fascinating discoveries delivered straight to your inbox. Contact me with news and offers from other Future brandsReceive email from us on behalf of our trusted partners or sponsorsBy submitting your information you agree to the Terms & Conditions and Privacy Policy and are aged 16 or over. However, the researchers have an explanation for AI's surprisingly poor time-reading abilities. "Early systems were trained based on labelled examples. Clock reading requires something different — spatial reasoning," Saxena said. "The model has to detect overlapping hands, measure angles and navigate diverse designs like Roman numerals or stylized dials. AI recognizing that 'this is a clock' is easier than actually reading it." Dates proved just as difficult. When given a challenge like "What day will the 153rd day of the year be?," the failure rate was similarly high: AI systems read clocks correctly only 38.7% and calendars only 26.3%. This shortcoming is similarly surprising because arithmetic is a fundamental cornerstone of computing, but as Saxena explained, AI uses something different. "Arithmetic is trivial for traditional computers but not for large language models. AI doesn't run math algorithms, it predicts the outputs based on patterns it sees in training data," he said. So while it may answer arithmetic questions correctly some of the time, its reasoning isn't consistent or rule-based, and our work highlights that gap." The project is the latest in a growing body of research that highlights the differences between the ways AI "understands" versus the way humans do. Models derive answers from familiar patterns and excel when there are enough examples in their training data, yet they fail when asked to generalize or use abstract reasoning. "What for us is a very simple task like reading a clock may be very hard for them, and vice versa," Saxena said. RELATED STORIES —Scientists discover major differences in how humans and AI 'think' — and the implications could be significant —If any AI became 'misaligned' then the system would hide it just long enough to cause harm — controlling it is a fallacy —Researchers gave AI an 'inner monologue' and it massively improved its performance The research also reveals the problem AI has when it's trained with limited data — in this case comparatively rare phenomena like leap years or obscure calendar calculations. Even though LLMs have plenty of examples that explain leap years as a concept, that doesn't mean they make the requisite connections required to complete a visual task. The research highlights both the need for more targeted examples in training data and the need to rethink how AI handles the combination of logical and spatial reasoning, especially in tasks it doesn't encounter often. Above all, it reveals one more area where entrusting AI output too much comes at our peril. "AI is powerful, but when tasks mix perception with precise reasoning, we still need rigorous testing, fallback logic, and in many cases, a human in the loop," Saxena said. Drew Turney Drew is a freelance science and technology journalist with 20 years of experience. After growing up knowing he wanted to change the world, he realized it was easier to write about other people changing it instead. As an expert in science and technology for decades, he’s written everything from reviews of the latest smartphones to deep dives into data centers, cloud computing, security, AI, mixed reality and everything in between. You must confirm your public display name before commenting Please logout and then login again, you will then be prompted to enter your display name. AI can handle tasks twice as complex every few months. What does this exponential growth mean for how we use it? Older AI models show signs of cognitive decline, study shows — but not everyone is entirely convinced AI is just as overconfident and biased as humans can be, study shows 'Math Olympics' has a new contender — Google's AI now 'better than human gold medalists' at solving geometry problems AI creates better and funnier memes than people, study shows — even when people use AI for help Traumatizing AI models by talking about war or violence makes them more anxious Latest in Artificial Intelligence Scientists use AI to encrypt secret messages that are invisible to cybersecurity systems 'Murder prediction' algorithms echo some of Stalin's most horrific policies — governments are treading a very dangerous line in pursuing them AI researchers ran a secret experiment on Reddit users to see if they could change their minds — and the results are creepy 'Annoying' version of ChatGPT pulled after chatbot wouldn't stop flattering users US Air Force wants to develop smarter mini-drones powered by brain-inspired AI chips AI is just as overconfident and biased as humans can be, study shows Latest in News 150,000-year history of Earth's magnetic field reveals clues about the climate when early humans were spreading out of Africa Scientists solve mystery of how orange cats got their coats — and why so many are male 'Yet another miracle save': NASA engineers complete nail-biting maneuver to resurrect Voyager 1's long-dead thrusters Living lunch box? Iceland orcas are unexpectedly swimming with baby pilot whales, but it's unclear why. 'Very rare' African ebony figurines found in 1,500-year-old Christian burials in Israeli desert NASA spacecraft snaps eerie image of eclipsed sun with an extra moon overhead. What's going on? LATEST ARTICLES Scaly-foot snail: The armor-plated hermaphrodite with a giant heart that lives near scalding deep-sea volcanoes and never eats Our moon may have once been as hellish as Jupiter's super volcanic moon Io Why do elephants have big ears? Science news this week: 2025's biggest solar flare and Yosemite's ghost volcano 150,000-year history of Earth's magnetic field reveals clues about the climate when early humans were spreading out of Africa Live Science is part of Future US Inc, an international media group and leading digital publisher. Visit our corporate site. Contact Future's experts Terms and conditions Privacy policy Cookies policy Accessibility Statement Advertise with us Web notifications Editorial standards How to pitch a story to us Future US, Inc. Full 7th Floor, 130 West 42nd Street, Please login or signup to comment Please wait...