MIND YOUR THOUGHTS

Technologies with potential to read one’s thoughts have grown tremendously in recent years driven largely by artificial intelligence, a develooment that raises ethical and privacy issues

- ROHINI KRISHNAMURTHY

IF YOU believe that no one can know what you are thinking, chances are you are not up-to-date with latest advancements in brain-computer interface (BCI). The innocuous-sounding technology has other-worldly effects.

This was on display in December 2023, when scientists from the University of Technology Sydney, Australia, released a video of a man reading lines silently in his head and a machine repeating his words. "Good afternoon! I hope you're doing well. I'll start with a cappuccino, please, with an extra shot of espresso," read the man from a screen, without uttering a sound. A robotic voice from a speaker announced: "Afternoon! You well? Cappuccino, Xtra shot. Espresso," states the university website.

As we think, or perform activities, such as picking up a cup of coffee and dancing, or even while resting, the activity of our nerve cells fluctuates. This can be picked up by neuroimaging devices such as EEG (measures electric activity by placing sensors on the scalp and was used in the University of Technology Sydney research); functional magnetic resonance imaging or fMRI (measures brain activity by detecting changes associated with blood flow); and magnetoencephalography or MEG (measures the magnetic fields produced by the brain's electrical currents).

These devices have been around for decades now, but use of artificial intelligence (AI) models, which can be trained to match a pattern of brain signals to thoughts or activities and to translate them into sentences, has enhanced brain-reading technologies. A particular branch of AI called machine learning gives computers the ability to learn without being explicitly programmed to do so. Researchers first train the algorithm by feeding brain data corresponding to different phrases or words. Once it learns to see patterns of what words can be mapped to complex neural data, it can decode the inner language of the brain.