Interview with the "Father of Brain-Computer Interface": Why I'm worried that Musk is misleading the public

Recently, Musk's brain-computer interface company announced that it had completed the first brain chip implantation operation, which caused heated discussions. However, Miguel Nicolellis, the father of brain-computer interfaces, has expressed concern about this commercial marketing, arguing that it could mislead the public and cause serious harm to this field of science.

Written by | Feng Lifei (Reporter of China Science News)

On the evening of February 19, local time, American entrepreneur Elon Musk said in an interview on social media platforms that the first patient with brain chip implantation completed by his brain-computer interface company Neuralink (Neuralink) can already move a computer mouse by thinking.

"It's progressing well and the patient seems to have made a full recovery...... And be able to control the mouse and move the mouse around the screen just by thinking. Musk said.

Like on January 29, when Musk announced on social media that Neural Connectivity had completed the implantation of the first human brain device, the news sparked heated discussions again. Quite a few people cheered:

"He's made history. ”

"The era of brain-computer interfaces pioneered by Neural Connections has finally arrived!"

"Musk is one step closer to his own sea of stars......

Brain-computer interface implants from Neural Connects. It is composed of biocompatible shells, batteries, chips and electronic devices, threads, and other parts. Image credit: Neuralink

However, Miguel Nicolelis, the "father of brain-computer interfaces" and professor emeritus at Duke University in the United States, was very worried that Musk's "commercial marketing" would mislead the public in an exclusive interview with China Science News.

"Mr. Musk has created so many fantasies around this topic that it has left many people confused about what real brain-computer interface technology is, which could cause serious harm to this field of science. Nicol Laillis said.

He pointed out that Neural Connections lost 15 to 17 monkeys trying to implant devices in 2023 alone, and such a high rate of animal loss is unusual. This rather serious fact also shows that the quality of their experiments is not up to par, and their technology is far from being able to be commercially applied in humans. He said.

Miguel Nicolilis Courtesy of the interviewee

Since entering 2024, "brain-computer interface" has become one of the most searched scientific and technological terms. The related technologies have been recommended by the internationally renowned scientific journal Nature as one of the seven technologies to watch in 2024. It is worth noting that with the rise of public opinion, recently, the official website of the Ministry of Science and Technology of China issued the "Ethical Guidelines for Brain-Computer Interface Research", which delineates the "red line" and "bottom line" for related research.

"It's not much progress"

China Science News: Musk recently posted on social platforms: "The first human patient underwent implantation surgery from a neural connection company and is currently recovering well. "As the founder of the brain-computer interface space, how would you rate the recent progress of Neural Connect?

Nicollais: It's not much progress, it's just commercial marketing, like a TV commercial, not even a new technology.

First of all, "the first brain-computer interface chip was implanted into the human brain" is a misleading statement. Brain-computer connectivity was verified by implanting a chip in the human brain, which was completed 20 years ago. In 2003 and 2004, my team surgically implanted a chip in the brain of a Parkinson's patient to accomplish the target task through virtual reality, proving that humans can use brain-computer interfaces like other animals. In 2006, BrainGate implanted brain-computer interface chips in four patients, and several companies have since carried out related research.

Second, no one evaluates a technology based on a piece of information on social media because it doesn't have any data to back it up. This is another marketing campaign by Elon Musk. We don't know how well the patients are recovering, all they do is develop the chip.

Our team obtained the relevant chip patent 10 years ago, and the research on the wireless transmission of the brain-computer interface Bluetooth carried out through monkey experiments was published in the journal Nature Methods in the form of a cover paper, which has been cited hundreds of times. This is how you report on progress in the field of science, you can publish your research and then allow your peers to make independent judgments.

China Science News: Is there a difference between the current commercialization and development direction of brain-computer interface technology and your expectations?

Nicolellis: I think brain-computer interfaces are very useful, but commercially, they don't necessarily need to be implanted in the brain. Over the past 10 years, we have used a non-invasive BCI treatment paradigm for paraplegics with spinal cord injury in our laboratories in the United States and Brazil, and we have achieved very meaningful results. "Non-invasive" means that you don't have to spend a lot of money on surgery or risk having a chip implanted in your brain. This makes the technology commercially promising, with the opportunity to scale up to millions of people.

From this point of view, Neural Connect's business plan is flawed, it will probably not make money, and a large number of investors in the company will suffer because it does not have a (as big) market as it would have been. The biggest problem with invasive BCIs is that they are suitable for a small patient population compared to non-invasive BCIs, which can benefit millions of patients, mainly for patients with severe paralysis or other serious diseases.

Neural Connections was founded by 3 of my students. They are very good engineers, good at making electrode chips and inventing devices to record brain activity, but they have no clinical training and no understanding of medicine.

In medicine, from animal experiments to drug development to clinical application, relevant evaluation criteria must be introduced. There are three main criteria for the development and widespread use of brain-computer interface technology: safe, effective, and affordable. The invasive brain-computer interface implant paradigm released by Neural Connections violates these three principles, it is not completely safe, the implanted device may produce an inflammatory response in the brain, its read-transmission efficiency does not compensate for the risks posed by implants in the brain, and it is currently very expensive.

During a 2020 presentation, Elon Musk stood next to Neural Connect's surgical robot. Because the threads of the implant are so delicate, it is impossible for a human hand to intervene. The surgical robot can implant it where it is needed. Image credit: Neuralink

China Science News: You once said that Musk exaggerated the role of brain-computer interfaces, what makes you believe that his words are misleading?

Nicol Lailis: Mr. Musk's promise that this brain chip that can be connected with a nerve will download brain information, or upload knowledge to the brain, is completely ridiculous and will never happen. The brain-computer interface is not the movie "The Matrix", as long as the protagonist Neo's brain uploads fighting skills, he can quickly learn kung fu. This is the plot of science fiction, which cannot be done in reality, and the brain and the computer learn in completely different ways.

Mr. Musk also claims that they can't treat schizophrenia and autism with this brain-computer interface paradigm from Neural Connect. Because the chips released by Neural Connections are just recording brain activity and sampling signals from the brain, as far as I know, they don't have any treatment for the condition.

Recently, Neural Connections has also claimed to make congenitally blind people visible through brain-computer interfaces, an idea that is not original to them. Since the 60s of the last century, people have been trying to do this kind of work in the retina and the visual cortex of the brain, but they have not achieved much success, which involves many complex technical problems.

It can be seen that people are only concerned about the various misrepresentations that Musk has delivered, or misleading promises that he will never keep, like fully autonomous vehicles or Martian migrants.

China Science News: The concept of brain-computer interface is still not very clear, which has led to ethical doubts about brain-computer interface. What is the core of the concept of brain-computer interface?

Nicollais: Brain-computer interfaces were originally used to record motor instructions for a portion of a patient's brain to help people realize their willingness to act in the clinic. Since the patient cannot move the body, neural commands can be used to control the movement generated by a robotic arm, robotic leg, or computer cursor, and brain electrical signals are a key component of the classical brain-computer.

So far, Neural Connections has shown chips that record brain activity, which can only record electrical signals from the brain and cannot be used to control anyone.

My concern is that Mr. Musk doesn't understand neuroscience and has created so many fantasies around the topic that it confuses a lot of people about what this technology can and can't do. I fear that he will mislead others and that his never-fulfilling, open-mouthed promises could cause serious damage to the entire field of science created by serious people like me who have been working for decades.

Non-invasive brain-computer interfaces are the "mainstream" in the next few years

China Science Daily: Nature magazine recommends brain-computer interfaces as one of the most noteworthy innovations in 2024. What do you think of this assessment?

Nicolellis: In fact, brain-computer interfaces have been at the forefront of attention for many years, including many times in Science magazine, which has featured my lab as a model for the industry.

Back in 2001, two years after we published our first brain-computer interface paper, MIT Technology Review visited my lab and listed brain-computer interfaces as one of the "Top 10 Breakthrough Technologies" that changed the world that year, and we were the only lab in the world to do brain-computer interfaces.

China Science News: After 23 years of development, what changes do you think has taken place in the current brain-computer interface research?

Nicol Lallis: Initially, we experimented with rats and monkeys, and the conditions were very rudimentary. Now, my lab can get multiple brains working at the same time, controlling a device called a "shared brain-computer interface" or "brain networking," and I've done that in humans — we've combined the brain activity of several patients to control a virtual arm. We've also created a "brain-machine-brain interface" where the brain can control the device, and the device can also transmit feedback directly to the brain. So, things have moved forward several orders of magnitude.

However, it is only recently that BCI technology has entered commercial applications that most scientists have become interested in it. Considering that a businessman like Elon Musk would threaten the space with his rhetoric, I felt I had to stand up and say "no" and tell them what to do right.

China Science News: Since 2012, you have created and led the implementation of the "Recovering to Walk" (WAP). Do you consider it a success, and how many patients around the world are currently being treated for WAP?

Nicollais: I can confirm without a second thought that it was a success. We have published hundreds of papers in the best journals in the world, and we have people who have been paralyzed for 20 years walking again. So, the results are very obvious.

Combining all relevant studies around the world, we recruited nearly 50 patients, and through different studies, the effectiveness of the WAP treatment paradigm has been demonstrated. We are planning to expand the network and take this rehabilitation program to a larger level, making it the foundation of the Treat One Billion program.

China Science News: Why do you propose to "treat 1 billion people"?

Nicol Lailis: According to the World Health Organization, about 1 billion people in the world suffer from brain disorders, which is why I named this project "Treating 1 Billion People." Its long-term goal is to provide affordable, safe and effective neurological rehabilitation.

Through this project, we also hope to form a direct competition with neuroconnectivity companies to demonstrate to the public that BCI technology can actually be rolled out to millions of people, not just a few, and that they will not be harmed.

You have to be aware that there is a link between this and the animal experiments of Neural Connect. In 2023, they lost 15 to 17 monkeys who tried to implant devices, and they are being investigated by the relevant authorities in the United States, which is a rather serious and must be explained. Such a high rate of animal loss is something I have never heard of in my life. I've worked with about 30 monkeys throughout my career, and I've never lost a monkey to surgery. So, this rather serious fact also shows that the quality of their experiments is not up to par, and the technology of neural connection is far from being able to be commercially applied in humans.

Neural Connect's monkey Pager plays video games through an implanted brain-computer interface. Image credit: Neuralink

The clinical paradigm on which the "Treat 1 Billion" project is based does not require brain implants, and patients do not have to bear the risk of contracting meningitis, rejecting reactions, or having implanted probes in the brain stop working. For example, when probes in the brain stop working, to remove them, brain tissue must be destroyed. This is a very big problem because existing probes often don't last long. As far as I know, there are companies that end up going bankrupt because the implantable probes they were using stopped working after just a few months, and there's not much they can do about it.

China Science News: An implantable application of brain-computer interface is the clinical trial in which the team of Stanford University professor Francis Willett treated patients with amyotrophic lateral sclerosis (ALS). Do you know about the team and the research they are doing, and can you rate their progress?

Nicol Laillis: I know the team, they're all very good scientists, they're doing a great job. But I still think that implantation of a BCI device is only necessary in very extreme rare cases, such as in ALS, where the patient's entire body has lost all ability to move. This is a practice that works in a small number of cases, so you can see that they publish a study with only one patient.

Although I have invented invasive BCIs and have been patenting them for 20 years, I think that for most patients and commercial enterprises, non-invasive BCIs will be the mainstream development in the coming years (large-scale commercial applications in medical rehabilitation). Non-invasive BCIs can be extended to a large number of patients, which is one of its main advantages in treating neurological and even psychiatric disorders.

Researchers in Nicollais Labs are trying to manipulate a robotic arm through a brain-computer interface. Photo courtesy of the interviewee

The era of commercialization is coming

China Science News: Can you talk about your current work in the field of brain-computer interface?

Nicol Lallis: We are working to achieve non-invasive, efficient, and affordable brain-computer interface therapy for various diseases such as Parkinson's disease, chronic epilepsy, stroke, and more, and to translate the results of nearly 40 years of animal testing into clinical therapies to benefit more patients.

China Science News: Do you think this technology can be put into practice and commercialized?

Nicol Lallis: I think so. If Elon Musk can commercialize existing implantable brain chips, we can commercialize safer and more efficient technologies. Non-invasive brain-computer interfaces do not require surgery and are much easier to get approved by regulatory agencies. In fact, we may have some surprises for Mr. Musk in the near future.

China Science News: It has been 25 years since you proposed the concept of brain-computer interface, how do you see the application prospects of related technologies?

Nicol Laelis: In 1999, I was doing brain-computer interface research with my postdoctoral mentor, then a professor at Hahnemann University Hospital (HUH). We worked together for the next 20 years until John's death a few years ago. It will take a long time to fully translate these findings into a commercial product, but I think we're pretty close to that now. That's why there's so much buzz in the market right now, and why businessmen like Musk can mislead so many people because the field is already so exciting.

Recovering patients at Nicol Lallis Labs. Photo courtesy of the interviewee

China Science News: At present, what challenges are still faced by the large-scale commercial application of brain-computer interface technology?

Nicol Lailis: There will definitely be challenges, but the hardest times are behind us.

China Science News: When was the hardest time?

Nicol Rillis: After 25 years of intensive basic and clinical research, we've removed the solutions that don't work in the brain-computer interface space and focused on what works, and we've made substantial progress. So, now it's closer to that goal than it was in previous years.

The difficulty in the current industry is to distinguish between what is real and what is fictional, who really wants to treat patients and who just wants to make a quick buck by misleading people. From a scientific point of view, we are ready to do what scientists promise.

Profile

Miguel Nicolellis, M.D., Ph.D., born on March 7, 1961, is a proposer and founder of the brain-computer interface (BMI) concept, a member of the French Academy of Sciences, the Brazilian Academy of Sciences and the Vatican Pontifical Academy of Sciences, an emeritus professor at Duke University in the United States, the founder of the Santos Dumont Institute in Brazil, the co-founder of the International Natal Institute for Neuroscience, and the founder of the world-famous "Re-Walking Project". He was named one of the top 50 technology leaders in the United States, was named one of the 100 Global Thinkers by Foreign Policy magazine in 2015, received the Outstanding Person of the Year Award from the World Alliance for Information Technology and Services in 2016, and received the IEEE Emerging Technology Award in 2017.

This article is reprinted with permission from the WeChat public account "Science Net".

Special Reminder

1. Enter the "Boutique Column" at the bottom menu of the "Huipu" WeChat official account to view a series of popular science articles on different themes.

2. "Back to Park" provides the function of searching for articles by month. Follow the official account and reply to the four-digit year + month, such as "1903", to get the article index in March 2019, and so on.