iStockphoto
While many artificial intelligence models have shown that they have the ability to possess intelligence far beyond most humans, some of the world’s most advanced AI systems still cannot interpret the hands on a clock and don’t understand how perform the arithmetic required to read a calendar correctly. At least that’s what new research claims to have revealed.
According to researchers at the University of Edinburgh, “Unlike simply recognizing shapes, understanding analog clocks and calendars requires a combination of spatial awareness, context and basic math – something that remains challenging for AI.”
This is how we will beat them when the inevitable AI robot uprising begins.
“Researchers tested various clock designs, including some with Roman numerals, with and without second hands, and different colored dials,” read a press release announcing the results of the study. “Their findings show that AI systems, at best, got clock-hand positions right less than a quarter of the time. Mistakes were more common when clocks had Roman numerals or stylized clock hands. AI systems also did not perform any better when the second hand was removed, suggesting there are deep-seated issues with hand detection and angle interpretation.”
When it came to reading a calendar, the researchers revealed that the AI models often failed to correctly answer a range of calendar-based questions, such as identifying holidays and working out past and future dates. “Even the best-performing AI model got date calculations wrong one-fifth of the time,” they stated.
“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,” said study lead author Rohit Saxena. “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.”
The study, published on the preprint server arXiv, showed that AI models including Meta’s Llama 3.2-Vision, Anthropic’s Claude-3.5 Sonnet, Google’s Gemini 2.0 and OpenAI’s GPT-4o read clocks correctly only 38.7% of the time and calendars only 26.3% of the time.
“Arithmetic is trivial for traditional computers but not for large language models,” Saxena added. “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.”
So, while AI might be able to create things that humans can’t even understand and be better than humans at creating dank memes, we still have the edge when it comes to telling time. Point: humanity.
Content shared from brobible.com.