Neuralink vs. The 1974 Malech Patent

In January 2024, Neuralink implanted a brain-computer interface chip in a human subject for the first time. The announcement was treated by most technology press as a historic milestone. Elon Musk described it as the beginning of a new era in human-machine integration. The coverage was extensive. Almost none of it mentioned US Patent 3951134.

That patent was filed in 1973 and granted in 1976. Its inventor was Robert G. Malech. Its assignee was Dorne and Margolin Inc., a defense electronics contractor based in Bohemia, New York. Its title was straightforward: Apparatus and method for remotely monitoring and altering brain waves. The patent describes a system for detecting brain wave frequencies in a human subject from a distance using radio frequency signals, and for transmitting modified signals back to the subject's brain to alter those frequencies.

Neuralink requires surgery. The Malech patent describes a remote system. That difference matters. What also matters is that the foundational concept — reading brain signals and transmitting signals back to alter them — was filed as a government-assigned patent fifty years before Neuralink's first human trial.

The Malech patent describes a transmitter that broadcasts radio frequency energy toward a subject. The subject's brain waves modulate the reflected signal in ways that a receiver can detect and analyze. The system can identify the dominant brain wave frequencies the subject is producing at any given moment. It can then retransmit a signal at a frequency designed to entrain the subject's brain waves toward a target state.

The patent specifies that the system operates without any physical contact with the subject and without the subject's knowledge or awareness of the process. The range at which the system operates is not specified in the patent claims, which is standard practice when the inventor wants to preserve the broadest possible scope of protection.

The patent was assigned to a defense contractor, not filed by an individual inventor for commercial purposes. Defense contractor patents are filed when the technology has been developed in a research context with potential military application. The assignee retained the rights. The US government, through its relationship with the contractor, had access to the technology and the research behind it.

Neuralink's N1 chip is a surgically implanted device containing 1,024 electrodes that record neural signals from the motor cortex. The signals are processed by onboard chips and transmitted wirelessly to an external receiver. The current application is assistive technology: allowing paralyzed individuals to control computers using neural signals alone. The first human subject, Noland Arbaugh, demonstrated the ability to control a cursor and play chess using the implant within weeks of surgery.

The technical architecture differs significantly from what Malech described. Neuralink requires implantation. Malech's patent describes a contactless system. Neuralink currently reads motor signals. Malech's patent describes reading and altering a wider range of brain wave frequencies. Neuralink's current application is assistive. Malech's patent explicitly describes alteration of brain states as a primary function.

The public record of brain-computer interface research between the Malech patent and Neuralink's founding is incomplete. Academic BCI research produced notable results across the intervening decades, including work at the BrainGate consortium that demonstrated neural control of robotic limbs in paralyzed patients as early as 2004. That research was published. It was peer-reviewed. It established that reading motor cortex signals and translating them into device control was technically feasible well before Neuralink existed.

What the public record does not contain is the full scope of research conducted by defense agencies and their contractors in the fifty years following the Malech patent. DARPA's work in brain-machine interfaces is partially documented through its public program announcements. The Neural Engineering System Design program, announced publicly in 2017, described goals that included implantable neural interfaces capable of reading and writing neural signals simultaneously, at the scale of one million neurons, with wireless transmission. That is a significant capability description for a publicly announced program, implying that the underlying research had already progressed substantially before the announcement.

Neuralink's January 2024 implant was the first time Neuralink implanted a chip in a human. It was not the first time a neural interface had been implanted in a human. BrainGate demonstrated human implants in 2004. Earlier research programs produced implants in the 1990s. The specific claim that Neuralink's coverage made, that this represented a historic first in brain-computer interface technology, was narrower than the headlines suggested.

What Neuralink may genuinely be first at is bringing a specific combination of electrode density, wireless transmission, onboard processing, and commercial development infrastructure to a human trial in a regulatory framework designed for eventual consumer deployment. That is a real achievement with a real distinction. It is different from the claim that the concept of reading and transmitting brain signals was invented in 2016.

The Malech patent sits in the public record. The fifty years of defense research that followed it sit in a combination of published literature, partially declassified program documents, and classified research that has not been released. Neuralink is building in public what has been developed in private for half a century. The question of how far that private development got, and what capabilities it produced, is not answered by anything in the available public record.

A defense contractor held a patent describing remote brain wave monitoring and alteration in 1974. Fifty years of subsequent research, partially public and partially classified, produced the technical foundation that private companies are now commercializing. Neuralink's human trials are real and significant. They are not the beginning of the story. The beginning of the story is in the patent office and in the classified research budgets of the intervening decades. That part of the record has not been fully opened.