While GPT-4 performs well in structured reasoning tasks, a new study shows that its ability to adapt to variations is weak—suggesting AI still lacks true abstract understanding and flexibility in decision-making.
Artificial Intelligence (AI), particularly large language models like GPT-4, has shown impressive performance on reasoning tasks. But does AI truly understand abstract concepts, or is it just mimicking patterns? A new study from the University of Amsterdam and the Santa Fe Institute reveals that while GPT models perform well on some analogy tasks, they fall short when the problems are altered, highlighting key weaknesses in AI’s reasoning capabilities.
Analogical reasoning is the ability to draw a comparison between two different things based on their similarities in certain aspects. It is one of the most common methods by which human beings try to understand the world and make decisions. An example of analogical reasoning: cup is to coffee as soup is to ??? (the answer being: bowl)
Large language models like GPT-4 perform well on various tests, including those requiring analogical reasoning. But can AI models truly engage in general, robust reasoning, or do they over-rely on patterns from their training data? This study by language and AI experts Martha Lewis (Institute for Logic, Language and Computation at the University of Amsterdam) and Melanie Mitchell (Santa Fe Institute) examined whether GPT models are as flexible and robust as humans in making analogies. ‘This is crucial, as AI is increasingly used for decision-making and problem-solving in the real world,’ explains Lewis.
Comparing AI models to human performance
Lewis and Mitchell compared the performance of humans and GPT models on three different types of analogy problems:
- Letter sequences – Identify patterns in letter sequences and complete them correctly.
- Digit matrices – Analyzing number patterns and determining the missing numbers.
- Story analogies – Understanding which of two stories best corresponds to a given example story.
A system that truly understands analogies should maintain high performance even on variations
In addition to testing whether GPT models could solve the original problems, the study examined how well they performed when the problems were subtly modified. ‘A system that truly understands analogies should maintain high performance even on these variations’, state the authors in their article.
GPT models struggle with robustness
Humans maintained high performance on most modified versions of the problems, but GPT models, while performing well on standard analogy problems, struggled with variations. ‘This suggests that AI models often reason less flexibly than humans, and their reasoning is less about true abstract understanding and more about pattern matching,’ explains Lewis.
In digit matrices, GPT models showed a significant performance drop when the missing number’s position changed. Humans had no difficulty with this. In story analogies, GPT-4 tended to select the first given answer as correct more often, whereas humans were not influenced by answer order. Additionally, GPT-4 struggled more than humans when key elements of a story were reworded, suggesting a reliance on surface-level similarities rather than deeper causal reasoning.
When tested on modified versions, GPT models showed a decline in performance on simpler analogy tasks, while humans remained consistent. However, both humans and AI struggled with more complex analogical reasoning tasks.
Weaker than human cognition
This research challenges the widespread assumption that AI models like GPT-4 can reason in the same way humans do. ‘While AI models demonstrate impressive capabilities, this does not mean they truly understand what they are doing,’ conclude Lewis and Mitchell. ‘Their ability to generalize across variations is still significantly weaker than human cognition. GPT models often rely on superficial patterns rather than deep comprehension.’
This is a critical warning about using AI in important decision-making areas such as education, law, and healthcare. While AI can be a powerful tool, it is not yet a replacement for human thinking and reasoning.
- Lewis, Martha, and Melanie Mitchell. “Evaluating the Robustness of Analogical Reasoning in Large Language Models.” Transactions on Machine Learning Research, 2025, openreview.net/forum?id=t5cy5v9wp
News
Scientists Unlock a New Way to Hear the Brain’s Hidden Language
Scientists can finally hear the brain’s quietest messages—unlocking the hidden code behind how neurons think, decide, and remember. Scientists have created a new protein that can capture the incoming chemical signals received by brain [...]
Does being infected or vaccinated first influence COVID-19 immunity?
A new study analyzing the immune response to COVID-19 in a Catalan cohort of health workers sheds light on an important question: does it matter whether a person was first infected or first vaccinated? [...]
We May Never Know if AI Is Conscious, Says Cambridge Philosopher
As claims about conscious AI grow louder, a Cambridge philosopher argues that we lack the evidence to know whether machines can truly be conscious, let alone morally significant. A philosopher at the University of [...]
AI Helped Scientists Stop a Virus With One Tiny Change
Using AI, researchers identified one tiny molecular interaction that viruses need to infect cells. Disrupting it stopped the virus before infection could begin. Washington State University scientists have uncovered a method to interfere with a key [...]
Deadly Hospital Fungus May Finally Have a Weakness
A deadly, drug-resistant hospital fungus may finally have a weakness—and scientists think they’ve found it. Researchers have identified a genetic process that could open the door to new treatments for a dangerous fungal infection [...]
Fever-Proof Bird Flu Variant Could Fuel the Next Pandemic
Bird flu viruses present a significant risk to humans because they can continue replicating at temperatures higher than a typical fever. Fever is one of the body’s main tools for slowing or stopping viral [...]
What could the future of nanoscience look like?
Society has a lot to thank for nanoscience. From improved health monitoring to reducing the size of electronics, scientists’ ability to delve deeper and better understand chemistry at the nanoscale has opened up numerous [...]
Scientists Melt Cancer’s Hidden “Power Hubs” and Stop Tumor Growth
Researchers discovered that in a rare kidney cancer, RNA builds droplet-like hubs that act as growth control centers inside tumor cells. By engineering a molecular switch to dissolve these hubs, they were able to halt cancer [...]
Platelet-inspired nanoparticles could improve treatment of inflammatory diseases
Scientists have developed platelet-inspired nanoparticles that deliver anti-inflammatory drugs directly to brain-computer interface implants, doubling their effectiveness. Scientists have found a way to improve the performance of brain-computer interface (BCI) electrodes by delivering anti-inflammatory drugs directly [...]
After 150 years, a new chapter in cancer therapy is finally beginning
For decades, researchers have been looking for ways to destroy cancer cells in a targeted manner without further weakening the body. But for many patients whose immune system is severely impaired by chemotherapy or radiation, [...]
Older chemical libraries show promise for fighting resistant strains of COVID-19 virus
SARS‑CoV‑2, the virus that causes COVID-19, continues to mutate, with some newer strains becoming less responsive to current antiviral treatments like Paxlovid. Now, University of California San Diego scientists and an international team of [...]
Lower doses of immunotherapy for skin cancer give better results, study suggests
According to a new study, lower doses of approved immunotherapy for malignant melanoma can give better results against tumors, while reducing side effects. This is reported by researchers at Karolinska Institutet in the Journal of the National [...]
Researchers highlight five pathways through which microplastics can harm the brain
Microplastics could be fueling neurodegenerative diseases like Alzheimer's and Parkinson's, with a new study highlighting five ways microplastics can trigger inflammation and damage in the brain. More than 57 million people live with dementia, [...]
Tiny Metal Nanodots Obliterate Cancer Cells While Largely Sparing Healthy Tissue
Scientists have developed tiny metal-oxide particles that push cancer cells past their stress limits while sparing healthy tissue. An international team led by RMIT University has developed tiny particles called nanodots, crafted from a metallic compound, [...]
Gold Nanoclusters Could Supercharge Quantum Computers
Researchers found that gold “super atoms” can behave like the atoms in top-tier quantum systems—only far easier to scale. These tiny clusters can be customized at the molecular level, offering a powerful, tunable foundation [...]
A single shot of HPV vaccine may be enough to fight cervical cancer, study finds
WASHINGTON -- A single HPV vaccination appears just as effective as two doses at preventing the viral infection that causes cervical cancer, researchers reported Wednesday. HPV, or human papillomavirus, is very common and spread [...]
