Home / Science / New AI Translates Brain Activity to Images
New AI Translates Brain Activity to Images
8 Oct
Summary
- AI tool decodes brain scans to recreate viewed images with high detail.
- Researchers developed a universal brain encoder, reducing scan time to one hour.
- Breakthrough raises significant concerns regarding mental privacy and future applications.
A groundbreaking artificial intelligence system, dubbed Brain-IT, has been developed by researchers at the Weizmann Institute of Science, capable of recreating images viewed by individuals through the analysis of their brain scans. The technology decodes neural signals recorded during functional magnetic resonance imaging (fMRI) scans, translating them into visual representations with remarkable detail and accuracy, surpassing previous methods that could only identify general image categories.
The innovation lies in its two-part AI decoder, which first maps the scene's layout, shapes, and colors, and then decodes its meaning. A significant hurdle in prior research was the extensive time, up to 40 hours, required for an AI to learn an individual's brain patterns. The Weizmann team overcame this by creating a 'universal brain encoder,' trained on numerous scans, allowing adaptation to new individuals with just one hour of data.
While the technology shows immense promise, it is not without its limitations, having produced humorous errors such as mistaking a dog for a goat. More profoundly, this advancement has sparked widespread discussions concerning mental privacy. Researchers acknowledge the significant challenges in decoding dynamic processes like dreams, but the potential for future advancements in understanding brain activity remains vast.
