Weizmann Institute AI reconstructs images from brain scans

This digest was compiled by AI from multiple sources — links to the originals are below.
Researchers at Israel's Weizmann Institute developed an AI tool that reconstructs what a person sees from brain scans with high accuracy. The system also predicts brain activity from viewed images, and scientists say it could aid brain research and help people with locked-in syndrome.
Key Facts
- The AI tool was developed at the Weizmann Institute in Rehovot, Israel.
- Eight volunteers each viewed about 9,000 images inside a high-resolution fMRI scanner to train the model.
- Neuroethicist Judy Illes of the University of British Columbia called the work 'magnificent' and said the approach is 'incredibly inspiring' for therapeutic help.
- Neuroscientist Tommy Sprague of UC Santa Barbara said the results look impressive but warned that covert extraction of thoughts could be 'disturbing in many ways'.
The Technology
The system reconstructs what a person sees by analyzing brain scans, and it also works in reverse to predict brain activity from viewed images. The researchers started with publicly available fMRI data and then trained the model on high-resolution scans from eight volunteers who each viewed about 9,000 images. Earlier attempts to reconstruct images from fMRI data produced blurry results, but the new approach achieves markedly better accuracy.
Scientific Reaction
Judy Illes, a neuroethicist and professor of neurology at the University of British Columbia, described the work as 'magnificent' and said the idea of using the approach for therapeutic help is 'incredibly inspiring'. Tommy Sprague, a neuroscientist at the University of California, Santa Barbara, said the results look very impressive but warned that a way to covertly extract information about what a person is thinking could make 150 years of science fiction real 'at any moment', which is 'disturbing in many ways'.
Potential Applications
The scientists say the discovery could help study the human brain and be useful for people with locked-in syndrome. The technology could also be used to recreate the content of dreams, according to the researchers.