The technology that connects brains to computers is becoming increasingly advanced. Will the emerging neurorights movement be able to keep pace?
By
Fletcher Reveley / Undark
|
Published Jan 3, 2024 3:00 PM EST
brain-computer interfaces. ” decoding=”async” data-dimension=”landscape”/>
Some scientists worry that not enough attention is being paid to the ethics of brain-computer interfaces. Chen Xiaogen/VCG via Getty Images
SHARE
This article was originally featured on Undark.
Back in May 2020, Jerry Tang, a Ph.D. student in computer science at the University of Texas at Austin, was amazed by a strange string of words on his computer screen:
“I am not finished yet to start my career at twenty without having gotten my license I never have to pull out and run back to my parents to take me home.”
Despite its lack of coherence, Tang saw it as a significant achievement: a computer had managed to extract a thought, no matter how disjointed, from a person’s mind.
For weeks, Tang had been refining a semantic decoder—a brain-computer interface, or BCI that produces text from brain scans. While the university was closed due to the pandemic, participants provided data to train the decoder by listening to hours of storytelling podcasts while their brain responses were recorded by a functional magnetic resonance imaging (fMRI) machine. Afterward, they heard a new story not included in the decoder’s training, and based on those fMRI scans, the decoder, which utilized GPT1, an AI chatbot similar to ChatGPT, produced a text prediction of what it believed the participant had heard. For a particular snippet, Tang compared it to the original story:
“Although I’m twenty-three years old I don’t have my driver’s license yet and I just jumped out right when I needed to and she says well why don’t you come back to my house and I’ll give you a ride.”
The decoder not only captured the essence of the original but also matched specific words exactly—twenty, license. When Tang showed the results to his advisor, UT Austin neuroscientist Alexander Huth, who had spent nearly a decade working on building such a decoder, Huth was astounded. “Holy shit,” he exclaimed. “This is actually working.” By the fall of 2021, the scientists were testing the device with no external stimuli at all—participants simply imagined a story, and the decoder produced a recognizable, if somewhat foggy, description of it. Huth commented, “What both of those experiments kind of point to, is the fact that what we’re able to read out here was really like the thoughts, like the idea.”
The scientists were thrilled about the potentially life-changing medical uses of such a device—restoring communication to people with
