Recent research from Edinburgh University highlights that AI still struggles with fundamental tasks, such as reading analog clocks. Despite advancements in artificial intelligence, particularly in multimodal large language models, the ability to interpret time from visual inputs remains a challenge. The study tested several well-known AI models, revealing significant shortcomings in their performance on time-related questions.
The findings indicate that AI systems read the time on analog clocks correctly less than 25% of the time, showcasing a gap in their cognitive abilities. While Google's Gemini 2.0 performed best in clock tasks, even the top models made errors, emphasizing the need for improvement in AI's temporal reasoning skills. Addressing these deficiencies is crucial for integrating AI into time-sensitive applications like scheduling and automation.
• AI struggles to read analog clocks and calendars effectively.
• Most AI models scored poorly on basic time-related tasks.
These AI models can interpret and generate various types of media, yet struggle with temporal reasoning.
This refers to the ability to understand and reason about time, which remains underexplored in AI.
Reading clocks and calendars involves intricate cognitive steps that AI currently fails to perform well.
OpenAI develops advanced AI models like GPT-4o, which were tested in the study.
0 was noted for achieving the highest score in clock reading tasks among the tested models.
Analytics India Magazine 14month
People on MSN.com 8month
Tech Xplore on MSN.com 12month
Isomorphic Labs, the AI drug discovery platform that was spun out of Google's DeepMind in 2021, has raised external capital for the first time. The $600
How to level up your teaching with AI. Discover how to use clones and GPTs in your classroom—personalized AI teaching is the future.
Trump's Third Term? AI already knows how this can be done. A study shows how OpenAI, Grok, DeepSeek & Google outline ways to dismantle U.S. democracy.
Sam Altman today revealed that OpenAI will release an open weight artificial intelligence model in the coming months. "We are excited to release a powerful new open-weight language model with reasoning in the coming months," Altman wrote on X.