Neuralink’s second human brain-computer interface implant, placed in participant Alex Conley in July 2024, has become an important early test of whether the company can turn a high-profile demonstration into a durable medical technology. The surgery at Barrow Neurological Institute in Phoenix followed a mechanical complication in Neuralink’s first participant and gave the company an opportunity to revise its implantation approach.[1]
For Alex, who has quadriplegia following a high cervical spinal-cord injury, the system has progressed from computer-cursor control and gaming to computer-aided design and, by April 2026, control of an assistive robotic arm. The results point to practical potential for people with paralysis, but they remain early feasibility-study findings rather than proof of long-term safety, clinical superiority or commercial readiness.[1][3]
A second implant designed to address a first-patient problem
Neuralink publicly disclosed the second surgery on August 21, 2024, saying it had occurred the prior month. The company did not provide an exact procedure date. Alex was discharged the day after surgery and recovered smoothly, according to Neuralink’s update.[1]
The case mattered partly because the first Neuralink participant, Noland Arbaugh, experienced partial retraction of implant threads after his January 2024 surgery. The movement reduced the number of electrodes producing useful signals and temporarily affected performance. Neuralink said it later restored performance through changes to its software and decoding methods, but the incident highlighted a central question for implanted BCIs: whether electrodes can maintain stable contact with brain tissue over time.[1]
For Alex’s procedure, Neuralink said it reduced brain movement during surgery and narrowed the gap between the implant and the surface of the brain. At the time of the company’s August 2024 report, it said it had seen no thread retraction in the second participant. That was a significant engineering result, though it was a short-term company report rather than independently published, multi-year clinical evidence.[1][5]
Alex was later identified by Barrow as Alex Conley. He fractured his C4 and C5 vertebrae, resulting in quadriplegia. His participation is part of Neuralink’s FDA-authorized PRIME study, short for Precise Robotically Implanted Brain-Computer Interface, an early-feasibility study evaluating the safety of the N1 implant and R1 surgical robot while testing whether people with paralysis can control external devices using neural signals.[1]
How Neuralink’s N1 system works
Neuralink’s N1, also called the Link, is a fully implanted wireless intracortical BCI. It uses 1,024 electrodes spread across 64 flexible threads. Because those threads are exceptionally thin and flexible, Neuralink uses its R1 surgical robot to insert them into motor-related regions of the brain.[2]
The implant records neural activity associated with intended movement. Low-power electronics process those signals and transmit them wirelessly to external software, which decodes learned neural patterns into commands for a cursor, keyboard or other connected device. The process is best understood as a motor-output interface: a user learns to generate reliable control signals, and the system maps those signals to an action.[2]
That distinction is important amid broad claims surrounding brain-computer interfaces. Neuralink has publicly demonstrated device control, not unrestricted reading of thoughts, memories or general internal speech from Alex’s implant. The company’s work may eventually support other applications, but such capabilities had not been publicly demonstrated in this participant as of July 5, 2026.
Neuralink reported that Alex controlled a computer cursor, played games including Counter-Strike, and began using CAD software to create three-dimensional objects. It also said he surpassed a previous non-Neuralink BCI cursor-control record on his first day using the Link. Those reports are encouraging, but they should be treated as company-reported performance claims, not as peer-reviewed clinical findings.[1]
From digital autonomy to physical assistance
The most consequential extension of Alex’s experience came in 2026. Barrow said in April that he became the first Neuralink participant to control an assistive robotic arm under the company’s separate CONVOY feasibility program. Demonstrations included operating a light switch, opening a door and moving objects.[3]
That progression matters because cursor control, while valuable, is only one part of the independence challenge facing people with severe paralysis. A robotic arm can potentially translate digital access into interaction with the physical environment. Barrow President and CEO Michael T. Lawton described the work as expanding “digital autonomy” into physical tasks.[3]
Still, a feasibility demonstration should not be confused with a deployed assistive product. The real-world usefulness of a robotic system will depend on reliability across days and years, setup time, calibration burden, safety safeguards, integration with home environments, caregiver needs and access to technical support. It will also depend on whether users can retain stable neural control as biology and hardware change over time.
Funding momentum, but no public-market verdict
Because Neuralink is privately held, the second implant did not produce a direct public-equity market reaction. Its impact was instead strategic: successful human-use demonstrations strengthened a fundraising narrative centered on an integrated wireless implant, robotic surgery and the prospect of restoring digital and physical control to people with paralysis.
Neuralink announced a $650 million Series E financing round in June 2025, naming investors including Founders Fund, ARK Invest, Sequoia Capital, Qatar Investment Authority, G42 and Thrive Capital. Semafor reported, citing people familiar with the deal, that the company was valued at roughly $9 billion before the new investment.[7][8]
The company said in January 2026 that it had reached 21 participants across its programs, with devices being used for computer control, communication and assistive robotics. That figure is company-reported and extends beyond the original two-person PRIME milestone, but it signals that Neuralink is expanding clinical activity beyond its initial public cases.[6]
Neuralink has also broadened its stated product direction. Its Blindsight vision program received FDA Breakthrough Device designation in September 2024, and its speech-related program received the designation in May 2025. Breakthrough designation can accelerate regulatory interaction for technologies addressing serious conditions; it is not marketing authorization and does not establish clinical effectiveness.[6]
A competitive field with unresolved questions
The second implant did not establish Neuralink as the clear technical leader in BCIs. Synchron is pursuing a less-invasive endovascular interface, while Blackrock Neurotech has extensive implanted-BCI research experience. Onward Medical is working on neural-interface and stimulation approaches for movement restoration, and companies including Paradromics and Precision Neuroscience are developing alternative high-bandwidth and less-invasive systems.[4]
Neuralink’s potential advantages are substantial: a high electrode count, fully implanted wireless architecture, robotic thread insertion and unusual access to capital and public attention. University of Washington researcher Rajesh Rao has noted that Neuralink’s robot and flexible threads could enable recordings from more neurons than some alternatives. But he also cautioned that claimed advantages have not yet been conclusively demonstrated, and that other developers may lead in less-invasive or more versatile approaches.[4]
The remaining questions are fundamentally clinical and operational. They include implant durability, signal stability, surgical complications, device replacement or removal, cybersecurity, patient-specific software maintenance, reimbursement and the level of long-term support required after implantation. Marco Baptista of the Christopher & Dana Reeve Foundation characterized BCIs as high-risk, high-reward technology: potentially transformative, but still too early to determine which technical approach will prove safest and most feasible.[4]
A Nature report similarly framed the second patient as a test of whether Neuralink could correct the mechanical issue encountered in the first implant, rather than a validation of the platform as a whole.[5] A separate Nature Reviews Bioengineering editorial has warned that the field must address hype, ethics, equitable access, user-centered design and long-term support alongside technical performance.[9]
What the second implant establishes—and what it does not
Alex Conley’s experience establishes a meaningful early milestone. Neuralink has shown, at least in one second participant, that its revised surgical approach was associated with no reported thread retraction in the initial follow-up period, and that its system could support both digital control and robotic-arm tasks. Those outcomes are relevant to people with paralysis because they suggest a route toward greater independence.
They do not yet establish that Neuralink’s implant is durable over years, broadly reproducible across patients, safer or more capable than competing systems, or ready for routine clinical use. The decisive evidence will come from transparent, peer-reviewed results across larger groups of participants and from the long follow-up required to assess implanted-device reliability. For now, Neuralink’s second human implant is best viewed as an engineering and clinical-recruitment breakthrough with substantial promise—and substantial work still ahead.
Editor’s Take
I see the second implant less as a victory lap than as an important engineering retest. Neuralink encountered a real mechanical failure mode in its first participant—thread retraction—then changed the surgical approach and reported a better early outcome in the second. That is how serious medical hardware develops: identify the failure, alter the design or procedure, and prove the fix holds over much longer periods than a launch update.
The robotic-arm demonstration is the more commercially meaningful milestone. Cursor control can restore communication and computer access; reliable physical manipulation could affect feeding, doors, lights, work tasks and independence at home. But the market will ultimately care about uptime, calibration time, revision procedures, support costs and reimbursement—not a polished demonstration. The next evidence to watch is independently reported, multi-year data on signal stability and safety across a larger participant group, plus evidence that users can operate practical assistive systems repeatedly outside a tightly supported lab setting.
References
- Neuralink – https://neuralink.com/updates/prime-study-progress-update-second-participant/
- Neuralink – https://neuralink.com/updates/prime-study-progress-update/
- Barrow Neurological Institute – https://www.barrowneuro.org/about/news-and-articles/patient-stories/barrow-patient-begins-to-reclaim-independence-with-neuralink-robotic-arm-after-spinal-cord-injury/
- Associated Press – https://apnews.com/article/elon-musk-neuralink-brain-computer-interface-9dbc92206389f27fd032825cf1597ee5
- Nature – https://www.nature.com/articles/d41586-024-02368-8
- Neuralink – https://neuralink.com/updates/
- Neuralink – https://neuralink.com/updates/neuralink-raises-650m-series-e/
- Semafor – https://www.semafor.com/article/05/27/2025/elon-musks-neuralink-raises-fresh-cash-at-9b-valuation
- Nature Reviews Bioengineering – https://www.nature.com/articles/s44222-024-00230-0
