The first results of the project, called "Artificial Intuition," are reported on the website of the biotech lab Neiry. Its ultimate goal is to develop a neural interface that will help connect the human brain with the knowledge of all of humanity. At the moment, scientists have shown the first implementation on a laboratory rat model.
The neural interface was implanted in a rat named Pythia. The implant works in tandem with AI. When the animal is asked a question, the AI formulates an answer and then starts stimulating a certain area of the animal's brain. As a result, the rat always correctly answers any question asked by pressing the keyboard.
For example, scientists asked Pythia about the structure of proteins, the advisability of choosing a foreign currency mortgage, the function of myelin sheaths, the Python programming language, Albert Einstein's birthday, and hundreds of other questions on various topics. In all cases, Pythia gave the correct answer.
"We wanted to demonstrate the concept of combining natural and artificial intelligence. Now we have plans for much more complex and interesting ideas for the interaction of AI and the brain," said Vasily Popkov, head of the laboratory for the development of invasive neurointerfaces at the Institute of Artificial Intelligence at Moscow State University.
In the future, the developers plan to create a popular user product. "Our user will be able to feel the truth or falsity of any statement or intuitively understand which answer in a test is correct. All this will increase the efficiency of human activity," they said.
The scientists are currently preparing to publish their results in scientific journals and are planning further research.
Previously, US company Synchron integrated generative AI into its brain-computer interface to help paralyzed patients communicate and interact with others. The new real-time chat feature offers suggestions based on the user’s context and emotions to make typing easier.