Elon Musk's new brain chip: A major tech-science milestone

From the flatworm to the computer

This process of evolution reached its peak with the capabilities of human beings. The human brain became a factory of ideas, allowing people to learn. Soon after, people discovered fire and have been creating and innovating ever since.

In the mid-20th century, the human brain began working on one of its most ambitious and transformative inventions yet: the computer.

In the 1940s, the first computers appeared. Like the human brain, they were able to store information. From the early days, innovators also began to use external memory sources.

The capabilities of the machines developed rapidly. Then, from the early 1990s, the world's computers were linked together via the Internet.

The arrival of the internet allowed communication along the lines of the rudimentary nervous system developed by jellyfish.

But unlike organic nervous systems – including the human one –  which can be damaged by disease or neurotoxins, these links between computers are much more durable and potentially indestructible.

Transhumanism

Neuralink is seeking to exploit this development. If electronic nerves are immune to damage, why not integrate them into damaged biological nerve locations to perform their function?

The concept of uniting organic intelligence with technology in this way is called "transhumanism".

The term gained prominence in the 20th century, popularised by biologist Julian Huxley. He envisioned a future where humanity could transcend its limitations by embracing technological advances.

Transhumanism wants to improve the human condition by developing technologies capable of significantly enhancing longevity, cognition, and well-being and making them widely accessible. It is for this purpose that Neuralink was founded.

Based in California in the United States, the company specialises in brain-computer interfaces or BCIs.

Since its early days, Neuralink has attracted top talent, including renowned neuroscientists from leading universities, enhancing its position as a pioneer in the field.

It is supported by substantial funding, including a significant $100mn contribution from Musk himself. The company expanded in size, employing 90 people by July 2019.

One of Neuralink's many innovations is a revolutionary device resembling a sewing machine. It is designed to implant incredibly thin thread chips directly into the brain. The threads are between 4 and 6 micrometres wide and can communicate with living neurons with unprecedented precision.

It also has a system capable of reading information from the brains of laboratory mice via 1,500 electrodes, paving the way for transformative applications in human experiments.

While Neuralink was initially slated to begin human trials in 2020, the company pushed that back to 2023. But the pace of progress picked up, and in May of that year, regulators approved the trials. It was a major milestone toward the company realising its vision.

Then came Musk's announcement on 29 January 2024.

A Neuralink chip was successfully implanted into the brain of a human patient who is currently recovering. But what exactly is the chip, and what does it do?

Brain-machine interfaces

Neuralink' 's chips are based on a long-established concept called brain-machine interfaces, or BMIs. This is the term for the technology that allows direct communication between the brain and an external device – such as a computer or a prosthetic – without traditional input methods, like keyboards or mice.

BMIs typically detect and interpret brain signals, such as nerve activity or electrical pulses, and translate them into commands that can control devices or trigger reactions.

There are different types of these devices. The most well-known are non-invasive BMIs, which do not require surgery but rely on external sensors to detect brain activity.

Examples include electroencephalograph (EEG), which measures electrical activity on the scalp, and functional magnetic resonance imaging (fMRI), which detects changes in blood flow in the brain.

Neuralink's chips belong to a different type of device: the invasive BMI. They require electrodes or sensors to be directly implanted into brain tissue or affixed to the brain's surface. They allow high-resolution neural recordings and precise control but need surgical intervention and pose certain risks.

These interfaces face two fundamental questions: How can accurate information be extracted from the brain? And how can accurate information be sent to the brain?

The first question involves capturing brain outputs from neural signals and recording what the neurons say. The second involves introducing information into the brain's natural neural flow, altering that flow, or stimulating neurons.

In a healthy, functioning brain, both processes occur naturally. When a person sees a fast car approaching, neurons will extract information and send it directly to the brain.

This, in turn, stimulates neurons in the occipital lobe –  the part of the brain area associated with vision – to analyse the scene and process information, sending commands to the limbs to move quickly away.

In the human brain, billions of neurons interact with each other naturally as part of the central nervous system. However, this system may malfunction due to illness or sudden injury.

Neuralink's brain threads

In 2019, during a live presentation at the California Academy of Sciences, the Neuralink team revealed the first prototype technology they were working on.

They debuted their thin, flexible thread chips and the neurosurgical robot to perform the procedures alongside an electronic system capable of processing information from neurons.

Neuralink's sensors are primarily made of polyimide, a biocompatible material, with thin gold or platinum connectors.  Each probe contains multiple electrodes capable of detecting electrical signals in the brain, with up to 3,072 electrodes in each configuration.

The means of insertion used by the robot uses a needle made of tungsten and rhenium, capable of inserting up to six wires (192 electrodes) per minute. In contrast, the imaging set in the insertion tool head allows real-time viewing and verification of the process.

Furthermore, the company has developed an application-specific integrated circuit (ASIC) to create a recording system comprising 1,536 channels. This system includes signal amplifiers, analog-to-digital converters, and peripheral circuit control devices to digitise and process neural information.

In July 2020, another milestone was achieved when Neuralink received a breakthrough device designation from the US Food and Drug Administration (FDA), allowing limited human testing under FDA guidelines for medical devices.

But controversy was to follow.

Monkey business

Neuralink conducted experiments on monkeys in collaboration with the University of California.

Claims emerged in 2022 from the Physicians Committee for Responsible Medicine accusing Neuralink and the University of California, Davis, of mistreating many monkeys, resulting in psychological distress, severe suffering, and chronic infections due to surgical procedures.

The committee alleged that at least 15 out of the 23 monkeys participating in the experiments died or were humanely euthanized as a result of chip implantation. It accused the University of California of covering up evidence of abuse.

In February 2022, Neuralink denied allegations of animal abuse following reports of macaque monkeys being humanely killed and euthanised following an experiment.

Musk himself proposed a potential alternative implantation method via the jugular vein instead of opening the skull.

Several months later, the company became the subject of a federal investigation by the US Department of Agriculture over alleged animal welfare violations. It found no evidence of animal rights violations other than a self-reported incident from 2019.

Wireless brain links

The Neuralink chip contains a set of electrodes meticulously designed to interact with neural tissues. These electrodes detect the electrical signals neurons produce and can also provide electrical stimulation to regulate neural activity.

Once implanted, the chip's electrodes detect the electrical signals neurons produce in the surrounding brain tissue. These signals represent the stimulation of individual neurons and provide valuable information about brain activity.

The chip includes integrated electronics to process the neural signals detected by the electrodes, including signal amplification, noise filtering, and converting analogue signals into digital data that can be analysed and transmitted.

The processed neural data is then transmitted wirelessly from the chip to external devices for further analysis and interpretation. This data transmission allows real-time monitoring of brain activity and enables two-way communication between the brain and external technology.

In addition to recording neural activity, the Musk chip can provide electrical stimulation to specific regions of the brain. This capability allows researchers to study how stimulating certain neural circuits affects brain function and behaviour.

By connecting the chip to these devices, users can interact with technology using only their thoughts, bypassing traditional input methods such as keyboards or touchscreens.

Brain-computer interfaces can be used to control prosthetics or external structures. This use allows people who are paralysed or have lost a limb to regain some mobility and independence.

However, Neuralink's primary focus now is to help individuals unable to speak or write communicate with others by allowing them to control a virtual mouse or keyboard or send messages through brain signals.

Obstacles

However, despite the immense potential of the chip to revolutionise various aspects of healthcare, communications, and human-computer interaction, there are still many significant obstacles and challenges ahead.

The brain is highly complex, with billions of neurons communicating through intricate networks. Deciphering the codes of these signals and interpreting them accurately poses significant challenges, not least because brain activity varies across individuals and contexts.

In addition, capturing and processing electrical signals from the brain can be susceptible to noise, interference, and signal attenuation.

Therefore, it is necessary to achieve high signal quality and reliability for accurate interpretation and effective communication between the brain and external devices.

Ensuring the stability and long-term durability of the implanted chip is essential for its operation and effectiveness. Issues such as electrode drift, tissue response, and device degradation over time must be addressed to maintain reliable performance and minimise the need for device replacement or overhaul.

Brain-computer interfaces also raise many ethical and privacy concerns, especially regarding the confidentiality and security of neural data. Therefore, protecting user privacy, obtaining informed consent, and ensuring fair access to BCI technology are essential considerations during development and deployment.

And so, there is much to think about at this important moment on a new frontier for the technical capabilities of human thought itself, with transhumanism likely to be part of the next chapter in the story of evolution.