Privacy, neuroethics, and the abandoned implant
Neural data privacy laws, agency and identity questions, the neurorights movement, and what actually happened to patients when their implant company went out of business. The questions companies now hire people to think about.
You are skimming: the title, the first figure, and the short version. Switch to Read in the header for the full page, or Deep to open every deep dive.
The neural coding page argued that mind-reading is not near. That does not make the ethics easy; it makes them specific. Who owns a recording of your motor cortex? Can a company sell your EEG to an insurer? What does a person owe, and what is owed to them, when a device is part of their body? And the one with a documented answer: what happens to a patient when the company that made their implant goes bankrupt. Companies in this field now hire people who can think clearly about these, and clear thinking starts with the cases.
Neural data privacy
EEG and intracortical recordings are health data when a clinician collects them and often are not when a consumer headband does; the gap is where the concern lies. Consumer neurotech companies’ privacy policies have, in reviews, allowed broad sharing of recorded brain data with third parties. What can be inferred from it today is limited (attention, drowsiness, some emotional states, medical conditions such as epilepsy) but the recordings persist and inference improves. Colorado in 2024 and California soon after amended their privacy laws to treat neural data as sensitive personal information, with consent and deletion rights; other states and countries have followed or are debating. Chile wrote mental privacy into its constitution in 2021.
For a designer: collect the minimum, store locally where possible (this site’s own choice), make deletion real, and write the privacy policy in a sentence a person can understand. If you would not want your own recordings handled that way, do not build it that way.
A person uses an implanted BCI for years. The company goes bankrupt. What happened, in the real cases?
Patients were left with unsupported implants. Second Sight’s Argus II retinal implant: the company stopped supporting the device in 2019 and later merged away; hundreds of users had implants that could no longer be repaired, with software and parts unavailable, and some faced explantation surgery or interference with MRI. Similar situations occurred with an early cochlear implant company and with a deep brain stimulation trial for depression whose funding ended. The question “what is the plan for these patients if the company fails” is now asked by ethicists, regulators, and, increasingly, by the patients before they consent.
The abandoned implant
The Second Sight case made explicit what an implant is: a relationship, not a product. The patient depends on the company for software, batteries, external components, and surgical support, indefinitely. Proposed fixes include escrow of designs and source code that is released if the company fails, requirements for continuity plans in trial protocols, open standards so another company’s hardware could take over, and funds set aside for explantation. None is standard yet. A student who can explain this case and the proposed remedies is having a conversation the field is having right now.
Agency, identity, and responsibility
Deep brain stimulation for Parkinson’s disease has, in a minority of patients, produced personality changes: impulsivity, apathy, mania. Patients and families have asked whether the person is still themselves and whether the device or the person is responsible for what they do. Closed-loop systems that act without the user’s moment-to-moment awareness sharpen the question. For BCIs that decode intent: if the decoder errs and the wheelchair hits someone, whose action was it? These questions do not have settled answers; the responsible position is to design so that the user’s control is maximal, the device’s actions are transparent and logged, and the person can always turn it off.
Consent for the unknowable
A person consenting to an implanted BCI trial is consenting to something whose long-term effects, including the effect of eventually losing it, nobody can fully describe. Ethicists argue for consent as an ongoing process, revisited, with explicit discussion of the end of the study and the end of the company. The participants in current BCI trials are, by every account, extraordinarily well informed and brave. The field owes them the continuity plans it does not yet have.
Neurorights
A movement, led by Rafael Yuste and others, to establish rights to mental privacy, personal identity, free will, fair access to cognitive enhancement, and protection from algorithmic bias, as a framework for the coming technologies. Chile adopted a version constitutionally. Critics say existing human rights law already covers most of it and that new categories may confuse; supporters say the specificity is needed before the technology arrives. Know both arguments.
Enhancement and fairness
If a BCI or a stimulation device improves memory or attention in healthy people (the evidence that current devices do is weak), who gets it, and what happens to those who do not? The military and the wellness industry are both interested. This site’s safety rule against self-stimulation is a small local answer; the large one is a policy question a technically literate person should be able to discuss.
Deep dive What a company's ethics work actually looks like 2 min
Data governance policies that say who can access neural recordings and for what. Privacy-by-design reviews of every feature. Patient advisory boards. Continuity-of-care planning in trial protocols. Publications in neuroethics journals by company scientists. An undergraduate with technical skill who has read the Second Sight case, the Colorado law, and the FDA BCI guidance is qualified to sit in those meetings and add something, which is a rare combination.
Explain what this page was about to your roommate in three sentences. No jargon they would not know.