Study Challenges 'Visual' AI Models' GPT-4o and Gemini 1.5 Pro Ability to Truly 'See'

Futurepedia - Learn to Leverage AI@futurepedia_io

2 years ago

🔍 A new study has revealed that AI models like GPT-4o and Gemini 1.5 Pro don't "see" like humans! They excel in pattern recognition but struggle with simple visual tasks. Are we overstating their visual capabilities? 🤔 #AI #MachineLearning #Vision https://t.co/6g8ENO1IIm

Reasoning skills of large language models are often overestimated - interesting study by MIT CSAIL researchers reveals surprising insights into AI capabilities 🤖💡 How do you think this impacts our understanding of AI? #AI #TechDiscussion https://t.co/2T8aOpVnWF

Talk AI Today@TalkAIToday

2 years ago

🤖🇺🇸 Can these so-called 'visual' AI models truly see? Spoiler alert: maybe not as well as we thought. Discover how models like GPT-4o and Gemini 1.5 Pro are failing simple visual tasks! https://t.co/zdM2h01f3V