How can AI 'read minds'?

Can artificial intelligence read our minds? Researchers from National University of Singapore say they have developed technology that can do just that. AI research participant Li Ruilin is getting ready to jump into an MRI machine for a brain scan. It will allow a computer to quote, read his thoughts and recreate the visuals he is looking at. I'm also interested in what happens in my brain and what my brain can output and what I'm thinking. So I try to participate in it to see what really happens on my brain. Li is one of potentially up to 58 participants who are providing the brain scan data sets to researchers working on developing a mind-reading AI. The technique relies on scanning the brains of volunteers in a magnetic resonance imaging MRI machine where they are exposed to 160,000 different images. The mind-reading AI, called MindVis, then associates the brain scans with the images and creates an individual AI model for each participant. Jiaxin Qin is one of the lead researchers on the study. So after we collect enough training data for you, we can create an individual AI model for you. And this AI model is kind of a translator. It can understand your brain activities just like Chai-Chi-Pi-Ti understands the natural languages of humans. The mind-reading kicks in when the same participant comes in again for a brain scan. So next time you come in, you will do the scan, right? And in the scan, you will see the visual stimuli like this and then we'll record your brain activities at the same time. And your brain activities will go into our AI translator and this translator will translate your brain activities into a special language that a stable diffusion can understand. And then it will generate the images you are seeing. So that's basically how we can read your mind in this sense. The technology is not able to mind-read the general public yet. Unlike scenes from science fiction movies, this AI system is modelled to individual participants. And there are also the possible risks of the datasets being assessed without consent, relative lack of legislations in AI research. Joanne Helen Zhao is the Associate Professor at NUS Medicine. The privacy concerns is the first important thing. And then people might be worried, right, whether the information we provided here might be assessed or shared without prior consent. So the thing to address this is we should have very strict guidelines, ethical and law in terms of how to protect the privacy. Still, the scientists are excited about the possibilities of a mind-reading AI system. One of the lead researchers, Chen Zijiao, explains. Say for some subject, some patient without like a motor ability, maybe we can help him to like control their robots and their phone, like communicate with others, like just using their thoughts instead of speech. In fact, that person couldn't speak at that time. So, that's the thing.