Have you heard about the latest controversy surrounding AI chatbots? A recent study published in The BMJ has revealed some concerning findings about the cognitive abilities of advanced chatbots in the tech industry. It seems that these chatbots are showing signs of mild cognitive impairment, which only worsen as they age.
The study aimed to challenge the idea that AI technology is advanced enough to be used in the medical field, particularly for diagnostic purposes. Researchers discovered that leading chatbots like GPT-4 and Gemini 1.0 and 1.5 struggled with tasks related to visuospatial and executive functions, such as drawing a clock or recalling a word sequence. These limitations raise doubts about the reliability of AI models in medical diagnostics.
Interestingly, when tested with the Montreal Cognitive Assessment, GPT-4 scored the highest among the chatbots, while the Gemini family performed the worst. Despite excelling in tasks like naming and language, all the chatbots showed a lack of empathy, a key symptom of frontotemporal dementia.
The researchers emphasize the importance of not attributing human qualities to AI models and advise against holding them to the same standards as humans. However, they argue that if tech companies are marketing these models as conscious entities, they should be held accountable for their cognitive abilities.
In conclusion, the study suggests that AI models may not be ready to replace human doctors just yet. It warns that healthcare professionals may soon find themselves treating virtual patients – AI models exhibiting cognitive impairment. These findings highlight the current limitations of AI technology and raise doubts about the industry’s claims regarding the capabilities of these advanced chatbots.
