Computers can now read minds as AI advances at a rapid pace
The ability to read minds is still something that people can see only in science fiction movies, or maybe not. The University of California has developed an artificial intelligence (AI) system that can actually turn people’s thoughts into speech by analyzing brain activity. This new AI machine can decipher up to 250 words in real-time in 30 to 50 words sentences. Researchers said that an algorithm was trained to do this through neural signals via electrodes implanted in the brain.
To train the algorithm, neural signals of four women with electrodes implanted in their brains were used. These electrodes were in place already as they have been used to monitor epileptic seizures. Volunteers in the study read repeated sentences aloud while the researchers processed the data coming from the brain waves to the AI system so it could recognize certain patterns that eventually were associated with actual words. The results were fascinating as the error rate across a repeated set was as low as 3%.
“A decade after speech was first decoded from human brain signals, accuracy and speed remain far below that of natural speech” states a paper detailing the research, published in the journal Nature Neuroscience. “We trained a recurrent neural network to encode each sentence-length sequence of neural activity into an abstract representation, and then to decode this representation, word by word, into an English sentence.”
This is the first step that proves that AI is capable of reading minds; however, there is still a long way to go for systems to understand regular speech. “People could become telepathic to some degree, able to converse not only without speaking but without words,” stated the report.