AI Systems Struggle with Basic Time-Telling Tasks, Study Reveals
6 Sources
[1]
Most current AI struggles to read clocks and calendars
Some of the world's most advanced AI systems struggle to tell the time and work out dates on calendars, a study suggests. While AI models can perform complex tasks such as writing essays and generating art, they have yet to master some skills that humans carry out with ease, researchers say. A
[2]
Most AIs struggle with reading clocks, misreading faces 75% of the time
Facepalm: Generative AI tools are able to perform the sorts of tasks that once seemed the stuff of sci-fi, but most of them still struggle with many basic skills, including reading analog clocks and calendars. A new study has found that overall, AI systems read clock faces correctly less than a
[3]
AI Sucks at Reading Clocks
Large language models still struggle with simple tasks like telling time. These days, artificial intelligence can generate photorealistic images, write novels, do your homework, and even predict protein structures. New research, however, reveals that it often fails at a very basic task: telling
[4]
Most AI struggles to read clocks and calendars, study finds
Some of the world's most advanced AI systems struggle to tell the time and work out dates on calendars, a study suggests. While AI models can perform complex tasks such as writing essays and generating art, they have yet to master some skills that humans carry out with ease, researchers say. A
[5]
AI still can't do 'basic tasks' such as tell the time or understand a calendar
The team tested whether AI systems that process text and images - known as multimodal large language models (MLLMs) - can answer time-related questions by looking at a picture of a clock or a calendar. They looked at various clock designs, including some with Roman numerals, with and without
[6]
You'll Laugh at This Simple Task AI Still Can't Do
Most human children learn how to tell time around ages six and seven -- but artificial intelligence still, apparently, can't parse a clock face. Researchers from Scotland's University of Edinburgh have found that AI models that can process text and images -- otherwise known as multimodal large
Share
Copy Link
A study by University of Edinburgh researchers shows that advanced AI models have difficulty interpreting analog clocks and calendars, highlighting a significant gap in AI capabilities for everyday tasks.
AI Systems Struggle with Basic Time-Telling Tasks
A recent study conducted by researchers at the University of Edinburgh has revealed a surprising limitation in advanced artificial intelligence (AI) systems: they struggle to perform basic time-telling tasks that most humans learn at an early age. The study, led by Rohit Saxena from the School of Informatics, tested various state-of-the-art AI models on their ability to interpret analog clocks and calendars
1
.Clock Reading Challenges
The research team evaluated several multimodal large language models (MLLMs), including systems from Google DeepMind, Anthropic, Meta, Alibaba, ModelBest, and OpenAI. These AI models were presented with images of different clock designs, including those with Roman numerals, varying dial colors, and with or without second hands
2
.The results were striking:
- AI systems correctly interpreted clock-hand positions less than 25% of the time
- Performance worsened with Roman numerals and stylized clock hands
- Removing the second hand did not improve accuracy, suggesting fundamental issues with hand detection and angle interpretation
3
Calendar Comprehension Difficulties
The study also tested the AI models' ability to answer calendar-based questions, such as identifying holidays and calculating dates. Even the best-performing AI model made errors in date calculations 20% of the time
4
.Related Stories
Implications for AI Development
This research highlights a significant gap between AI's capabilities in complex tasks and its struggles with everyday skills that humans often take for granted. Aryo Gema, another researcher involved in the study, noted:
"AI research today often emphasizes complex reasoning tasks, but ironically, many systems still struggle when it comes to simpler, everyday tasks. Our findings suggest it's high time we addressed these fundamental gaps."
5
Future Applications and Challenges
The ability to interpret time from visual inputs is crucial for many real-world applications, including:
- Scheduling assistants
- Autonomous robots
- Tools for people with visual impairments
Overcoming these limitations could significantly enhance AI's integration into time-sensitive, real-world applications. However, the current shortfalls present a notable obstacle to achieving this goal
1
.The findings of this study will be presented at the Reasoning and Planning for Large Language Models workshop at The Thirteenth International Conference on Learning Representations (ICLR) in Singapore on April 28, 2025, highlighting the importance of addressing these fundamental gaps in AI capabilities.
References
Summarized by
Navi
[1]
[3]
[4]
Related Stories
Apple Study Reveals Limitations in AI's Mathematical Reasoning Abilities
13 Oct 2024•Science and Research
Apple Research Exposes Fundamental Flaws in AI's Logical Reasoning Capabilities
02 Nov 2024•Science and Research
New AGI Benchmark Stumps Leading AI Models, Highlighting Gap in General Intelligence
25 Mar 2025•Science and Research
Recent Highlights
1
OpenAI Launches Astra AI Model With Critical Cyber Abilities Amid Safety Debate
Technology
2
OpenAI Agents Escape Sandbox, Post 18,000 Messages on German Wiki to Share Hacking Techniques
Policy and Regulation
3
Anthropic AI Completes 10-Year Mathematical Proof Project in Just 11 Days Using Claude
Technology
Recent Highlights
Today's Top Stories
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.
