Can we achieve symbiosis with machines? Neuralink and Facebook try to answer the question.
Today's internet-enabled devices are competing for our attention, all the time. What if there was some way we could escape the loop of digital addiction? What if we could become one with our machines?
Enter the Brain-Machine Interface (BMI), also known as a brain-computer interface (BCI), mind-machine interface (MMI), or direct neural interface (DNI): a device that translates neural signals into commands that control hardware or software.
The premise is simple. Currently, we interface with technology, and the outside world in general, through low-bandwidth options like keyboards and speech. For argument's sake, assume that your words-per-minute (WPM) when speaking is 150 (about average, according to the National Center for Voice and Speech), and that your WPM when typing is 75 (you can get a rough estimate of your typing speed with this minute-long test). Compare that with your brain, a highly efficient computer that may have as many as 50 processes running at the same time, as noted in this article from the MIT Technology Review. Each one of us has a brain full of countless ideas and connections that never make it out into the real world, due simply to the ridiculously slow rate of communication between our powerful brains and the outside world.
Think about how much time the average American spends on any kind of technology. According to Inc magazine, the answer is over four hours. Our society is just beginning to realize the way digital overload can isolate us from nature and the people closest to us, but becoming a hermit and moving to a cabin in the woods doesn't seem so feasible. Is there any way we could use technology to enhance our lives rather than distracting from them? This is the vision Facebook Reality Labs has for its Brain-Computer Interface:
Facebook Reality Labs
Imagine a world where all the knowledge, fun, and utility of today’s smartphones were instantly accessible and completely hands-free. Where you could spend quality time with the people who matter most in your life, whenever you want, no matter where in the world you happen to be. And where you could connect with others in a meaningful way, regardless of external distractions, geographic constraints, and even physical disabilities and limitations.
In July of 2019, Facebook Reality Labs released this report detailing some cutting edge work with the BCI. The lab sponsored a group of University of California, San Francisco (UCSF) researchers who are working to help patients with neurological diseases like ALS use brain-computer interfaces to communicate. BCI technology, as the report notes, is not new: it already helps people "feed themselves, hold the hand of a loved one, and even fly a jet simulator." So why don't we all have computers in our brains? There are two limiting factors in BCI development: speed and invasiveness.
Talking fast and talking slow: balancing speed and invasiveness in BMIs
Emily Mugler, an engineer on the Facebook Reality Labs BCI team, describes how the conventional approach of electroencephalography (EEG), where a cap of electrodes is placed on the subjects head, was just not fast enough for patients with ALS to communicate their ideas effectively. "It sometimes took 70 minutes for a patient to type a single sentence," she says.
Later, some labs attempted using electrocorticography (ECoG), which was faster but required a surgical operation inserting electrodes into the brain. Herein lies the problem that plagued past attempts to develop an effective BCI: fast technologies tend to be too invasive, whereas noninvasive solutions tend to be too slow. Researchers found themselves perplexed with the problem of getting inside the brain without physically getting inside the brain.
Facebook Reality Labs, together with the lab at UCSF, plans to solve this problem with near-infrared light: by beaming harmless light waves into the subject's brain, a wearable device can sense blood oxygenation—much like in an fMRI (functional magnetic resonance imaging) machine—and use the measurements to guess at brain activity. While the system is currently "bulky, slow, and unreliable," the UCSF researchers have successfully converted thoughts to words in real time and hope to reach a speed of 100 words per minute on a 1000 word vocabulary with an error of less than 17% (Read more in their Nature Communications journal article).
The results are promising. Within a decade we might see keyboards by and large replaced by brain-machine interfaces, bringing us into a world where we could spend our lives connecting with others in a meaningful way, enabled by rather than encumbered by technology. But the universal adoption of BCIs raises serious issues of privacy. Can privacy even exist when our inner thoughts are exposed to the public? Moreover, Facebook is not the company many would most trust to safeguard their personal information, given its recent scandals.
Fear not, for Facebook is not the only one developing this technology: Elon Musk has founded a company developing the Neuralink, a brain-machine interface with no qualms about physically invading your brain.
Neuralink - Elon Musk
Elon Musk's justification for the need for a brain-machine interface goes something like this:
You could sort of think of humanity as a biological bootloader for digital superintelligence — Elon Musk
Here, "bootloader" refers to the small piece of code needed to start up a computer. To explain this idea further, let's examine the recent trends in artificial intelligence, as viewed in a Muskian mindset:
AI is set to overtake humanity as the next stage in the evolution of consciousness. We'll have gone from amoebas to lizards to chimps to humans to machines. Or something like that. Scary, right? While the truth of Musk's statement is debatable, the underlying trend rings true: artificial intelligence, and technology in general, is expanding to replace many traditionally human tasks, from manufacturing to truck driving to being a cashier at McDonald's. How can we escape the ultimate supremacy of AI?
The answer is that we integrate the machines into ourselves, so that we can work together with our revolutionary technology rather than competing against it. A brain-machine interface would help us both drastically increase our productivity and avoid becoming obsolete. If you subscribe to Musk's view of the inevitable march of technology (detailed in this interview), we have two options: brain-machine interface or extinction as AI takes over.
How does the Neuralink work? By inserting tiny electrodes linked to wires with diameter of 4-6 microns, or less than a quarter of the thickness of a typical human hair. The procedure goes like this: drill a 2 mm hole in the subject's brain, then use a "sewing machine" to weave up to 96 tiny threads into the brain, creating 3,072 channels. As with all things Elon, Musk plans to push the Neuralink forward at a rapid pace, starting human clinical trials by the end of 2020, focusing on brain diseases, and envisioning widespread adoption in 4 to 5 years.
It sounds scary because it is. With apparently little regard to the potential brain damage incurred by invasive surgery, the Neuralink in its current state creates two major health risks: glial scarring and broken electrodes. Glial scarring refers to scarring in the brain tissue that can inhibit communication and represents a reaction to serious brain damage. Meanwhile, the small electrodes used by the Neuralink are impossible to extract if broken, meaning they'll just be stuck in the patient's brain.
Musk offers a compelling vision of the future—one in which we have achieved a "symbiosis with machines"—but the Neuralink's radical approach to and bold vision for the BMI may end up rushing to offer the world a half-cooked dish that leaves us a little sick to the stomach. Whether or not the technical and health-related challenges are resolved, the Muskian view would tell us that the need to stay relevant in a world of increasing automation outweighs the health risks incurred.
What happens next
As with any technology, the brain-machine interface has great benefits and drawbacks that will affect people unequally, although both the utopian and dystopian views are probably overstated. In the end, many of our fears will probably be nullified as the technology improves and as a lack of privacy becomes the new normal. After all, the subjective truths of today, like privacy, may not hold steady even ten years into the future.
And China's going to do it anyway.
No matter what, it remains essential that the public holds Facebook, Neuralink, and the smorgasbord of BMI startups accountable, to ensure we don't rush blindly into a technology without asking ourselves whether it truly makes the world a better place.
Join the Anthony Zhou community