paid EEG and neural rights

Scope: BCI companies, clinical work, research, funding, policy and neural data activity first reported or materially changed from 1 to 10 August 2026.

Conduit advertises paid EEG data collection. Its current page offers 50 dollars for two hours and 100 dollars for four hours. It allows up to ten two hour sessions. An August 2026 news report stated a thirty two session limit. The company page is the operative source for the current terms. The company's public terms and privacy policy do not name an institutional review board, an ethics board, neural data or a research protocol.¹ This does not establish what appears in a private consent form. It establishes what the public documents reviewed do not disclose.

Neural rights are here. While everybody is talking about AI, AGI and new models, your neural data is in danger.

Neural data is information generated by measuring activity in a person's central or peripheral nervous system. Neuro rights is the argument that this data, and the mental activity it can reveal, requires protection beyond ordinary privacy law. The four US provisions below do not create a standalone right of that kind. Three of them add neural data to the sensitive data category of an existing consumer privacy statute, which applies that statute's existing consent and disclosure duties to it. Montana used a different vehicle and extended its genetic privacy act to neurotechnology data with an express consent requirement. Chile is the contrasting model and amended its constitution.²

Four US states now have enacted provisions that cover neural data. Colorado amended its privacy law in 2024. California followed in 2024. Montana enacted its provision in 2025. Connecticut's provision took effect on 1 July 2026.²

Connecticut's law has been effective for five weeks. No enforcement action there is a short duration result. Colorado has about two years of operating history. No neural data enforcement action was found in the Colorado Attorney General material reviewed.

Chile added neuro rights language to its constitution in 2021. Its Supreme Court decided the Girardi v Emotiv case in August 2023.

The court did not order a sales ban. It did not suspend sales. The operative text granted the appeal only so that the Instituto de Salud Pública and customs would evaluate the Emotiv Insight device. The deletion direction concerned the plaintiff's data.³

The principal referral ended in March 2024. The Instituto de Salud Pública said the Emotiv Insight was not a medical device or pharmaceutical and was outside its jurisdiction. The reviewed record contains no resulting prohibition, licence, fine or further regulatory order.⁴

The case produced a judgment, one person's data deletion and two agency referrals. The medicines regulator said it lacked jurisdiction.

The US Common Rule applies through federal department adoption and related institutional coverage. It does not impose a general federal institutional review board requirement on every privately funded commercial neural data collection programme.⁵

The Food and Drug Administration governs clinical investigations within its device remit. Data protection regulators in the United Kingdom and Europe have published guidance or analysis on neural data and neurotechnology. UNESCO adopted a Recommendation on the Ethics of Neurotechnology in November 2025. That recommendation is non binding.⁶

These instruments do not amount to an enforcement action against the paid collection of consumer EEG data.

1. Paid EEG data collection is a current commercial offer.

Conduit is a San Francisco company building what it describes as a non-invasive thought-to-text system, a headset meant to complete sentences from thought. In the paid sessions participants speak and type responses to on-screen prompts while wearing the headset. Conduit lists the prices and session limit on its public page. Its terms and privacy policy do not publicly identify research oversight. The company stated in December 2025 that it had collected about 10,000 hours. That figure is a company claim and has not been independently audited.¹

2. Two California bills entered the suspense process.

AB 1883 and AB 1542 were placed on the California Senate Appropriations suspense file on 3 August. AB 1883 is a workplace surveillance bill. It would bar employers from using monitoring tools to collect neural data or to infer a worker's emotional state. AB 1542 would bar businesses from selling or sharing sensitive personal information. That is the category California extended to neural data in 2024. The suspense file is where bills with a fiscal cost wait for a single later vote that decides whether they advance. They are proposals. They are not current law.⁷

3. Two Chinese BCI standards took effect.

GB/T 47023-2026 and GB/T 47127-2026 took effect on 1 August. Per the national standards registry the first sets out a reference architecture for BCI systems and the second defines a multi-modal data format for them. They are recommended national standards rather than mandatory standards, so adoption is voluntary. Together they give Chinese BCI development a common technical baseline.⁸

4. Active Tech disclosed an RMB 330 million round.

Shanghai based Active Tech disclosed an RMB 330 million angel round on 3 August. The company was founded in February 2026 and works on invasive interfaces. Its roadmap spans flexible electrodes, dedicated chips and decoding algorithms. TechNode reported the round as a record for angel financing in China's BCI sector. Zhongke Chuangxing led it with Lenovo Star among the participants. The money is earmarked for facilities, clinical trials and hiring. The amount was repeated by news outlets and traces to the company. It is a company claim.⁹

5. EPFL registered an exoskeleton study.

The VALSE study was first posted on 4 August. It is an EPFL safety study for people with chronic traumatic spinal cord injury. It pairs the TWIICE Rise exoskeleton with epidural electrical stimulation, and where a participant has an implanted brain-spine interface the robot can take commands from brain signals. The registry lists safety as the primary outcome over one year of follow-up, twelve planned participants, a not yet recruiting status and an estimated start date of 15 September 2026. No participant result exists.¹⁰

6. A stroke meta analysis reported a 2.38 point difference.

The analysis included ten randomised controlled trials and 366 participants. The trials tested BCI training for lower limb recovery after stroke, where the system reads movement intent from the brain and uses it to drive feedback or assistance in a closed loop meant to promote neuroplasticity. The Fugl Meyer scale is a standard clinical measure of motor recovery. The analysis reported a 2.38 point between group difference on its lower extremity score, with a 95 percent confidence interval of 1.72 to 3.04. The paper cited a six point minimal clinically important difference, so the gain is statistically solid and below the paper's own bar for clinical meaning.¹¹

7. A neurorights paper set out eighteen recommendations.

Sirbu and colleagues published eighteen recommendations for responsible neurotechnology on 3 August. The authors write from Yale and the University of Bologna, Luciano Floridi among them. Their argument is that device regulation built for pacemakers and deep brain stimulators does not cover the risks of next generation networked BCIs. Nine recommendations go to developers and nine to policymakers. They cover autonomy, identity, mental privacy, cybersecurity and data governance. The paper is a normative analysis. It is not a law or regulatory action.¹²

This issue combined three weekly research passes and eight STORM evidence files, each produced by a separate AI model working from a written brief, then merged and verified against primary sources before publication. STORM is a multi-perspective research method from Stanford. Three of three independent support was treated as established for the merge. Two of three support could be published with the missing confirmation stated. One of three support required primary verification. Absence findings received less weight than affirmative records. Structured APIs were used for ClinicalTrials.gov, PubMed, Crossref and preprint discovery, and public pages and primary instruments were then opened for verification.

One of the three STORM runs was not blind, because it read the other two first. Its agreements received reduced weight. Its corrections received full weight when the operative court text, statute or regulator document supported them.

A note on scope. Synapse Dispatch reports on brain-computer interface developments already in the public domain. It is journalism and analysis, not laboratory or biological research, and it publishes nothing confidential. The value we add is verification and context, not new experiments.