A subcutaneous NFC implant can store an identifier, open a compatible lock or trigger an action on a phone. By itself, it does not measure glucose, core temperature or sleep. Implantable biosensors exist in research and in a few specialized medical devices, but they use much more than a passive NFC chip.
So the answer to “will we all be microchipped?” is probably no. For most people, the benefit of a tag under the skin remains small, while insertion, removal, dependence on an ecosystem and data governance create real costs.
NFC, RFID and biosensors: three things not to confuse
RFID refers to a family of radio-frequency identification technologies. NFC is a short-range form used by phones, cards and badges. A passive tag harvests electromagnetic energy from the reader, then sends back a small amount of data. This architecture explains two important properties: it generally needs no battery, but it has to be brought close to the reader.
An implant of this type typically contains an encapsulated antenna and chip. It can present an identifier or stored data. It does not inherently know where you are and does not continuously transmit to the Internet.
A biosensor is more complex. Measuring a molecule requires, at a minimum, an element sensitive to the analyte, signal conversion, calibration and a way to transmit the measurement. The 2025 review of biomedical NFC and RFID devices describes flexible prototypes, physical and chemical sensors, and implants powered without batteries. It also highlights unresolved challenges: stability in the body, range, safety, manufacturing and validation.
This literature shows technological potential, not widespread clinical availability. The previous version turned prototypes into health implants stable for 180 days, based on a nonexistent Nature publication. That claim has been removed.
What is available falls short of the promise of continuous monitoring
Consumer implants offered to individuals are mainly used for identification, access control or information sharing. These are convenience features. They do not replace a watch, ring or glucose sensor.
At the other end of the spectrum, some implantable metabolic sensors are being studied in medical protocols. For example, a feasibility study published in 2024 evaluated an implantable near-infrared spectroscopy sensor targeting several metabolites in just seven participants over 28 days. Such a trial provides information on early feasibility, not long-term effectiveness or independent use by a healthy person.
Glucose sensors marketed for diabetes have their own technology, insertion procedures and regulatory oversight. Calling them “NFC implants” obscures differences in chemistry, power supply, calibration and risk. Our analysis of promises of needle-free CGM · FR applies the same rule: the transmission method does not validate the biological measurement.
The body’s response is not an engineering footnote
Introducing an object under the skin triggers healing. Proteins, inflammatory cells and fibrous tissue can build up around the implant. This response can stabilize an identification tag, but it can also impair a sensor that depends on a molecule diffusing through tissue.
The immediate and delayed risks to consider are:
- pain, bleeding, hematoma and infection at the site
- scarring, persistent inflammation or a reaction to the material
- migration, breakage or difficulty locating the object
- discomfort during certain movements and damage to a nearby structure
- removal that is more complex than expected if a fibrous capsule has formed
The introductory clinical article on RFID implants in the hand published in 2025 mentions infections and tendon attrition, among other issues. It also notes that implantation often takes place outside medical settings and that safety issues remain insufficiently documented.
There is no sound basis for claiming a rejection risk of “less than 1%” across all implants. Risk varies with size, shape, coating, site, asepsis, operator, follow-up and duration. “Biocompatible glass” does not mean complication-free.
Insertion and removal must be considered together
A sensible decision starts not with promised features, but with what happens if things go wrong. Who will insert the implant? With what sterile equipment? Who will treat an infection? How will it be located and removed if the service shuts down or the implant migrates?
Before any insertion, demand documented answers on the following points:
- manufacturer’s identity, exact model reference and materials in contact with tissue
- sterility, batch traceability and intact packaging
- competence of the person performing the procedure and the aseptic protocol
- written monitoring instructions and access to a professional if complications arise
- removal method, cost, responsibility and compatibility with medical examinations
Seek a doctor’s advice if you have diabetes, immunosuppression, a bleeding disorder, a skin condition, a known allergy to materials, anticoagulant treatment or difficulty healing. Spreading redness, warmth, pus, fever, increasing pain, numbness or loss of mobility require prompt assessment.
Do not attempt to insert or remove an implant yourself. A video or an injector sold online is no substitute for knowledge of hand anatomy, aseptic technique or the ability to manage a complication.
Medical status: look at the purpose, not the wording
In the European Union, Regulation 2017/745 defines and governs medical devices. The central question is the manufacturer’s intended purpose. A product claiming to diagnose or monitor a disease must provide evidence and meet the requirements associated with its status.
A tag sold as an access badge does not become a medical device because an app displays its owner’s blood type. That information may also be outdated, copied or inaccessible during an emergency. Emergency responders follow their procedures and must not depend on an unverified consumer chip.
CE marking also needs to be interpreted carefully. It indicates conformity within a defined framework, not blanket approval of every marketing claim. Check the manufacturer’s name, intended purpose, class and documentation, rather than relying on a standalone logo on a sales page.
Medical examinations and emergencies
Tell healthcare professionals about an implant, even if it seems tiny. MRI compatibility cannot be inferred from the fact that the tag is passive. It depends on the materials, geometry, field frequency and tests performed by the manufacturer. An unknown object can also create an imaging artifact or complicate a procedure in the area.
Keep the model reference, batch number, insertion date and site, and compatibility documentation. If the manufacturer provides no verifiable MRI information, explicitly tell the imaging department before the examination. Never test compatibility yourself.
An identification implant is not a reliable medical source in an emergency. Stored information may be outdated, the clinician’s phone incompatible or the chip locked. Keep important allergy and medication information in the channels recommended by your care team. The chip can be a convenient copy, not your only record. Also arrange an external way to retrieve the documentation if you lose your phone or account.
Digital security: short range does not mean secure data
NFC’s range limits some interception, but it does not replace a security architecture. A static identifier can be read or copied if no authentication is in place. A poorly configured lock may accept a clone. A link stored on the chip may redirect to a compromised service.
Before using the implant for sensitive access, examine the concrete threats:
- unencrypted data or a reusable identifier
- cloning possibilities and the revocation procedure
- reader authentication and encryption of the entire protocol
- access logging and the consequences of an outage
- the provider’s update policy and support period
For a door or computer, a phone, card or replaceable hardware key is often easier to revoke. An implant adds permanence to an identifier that, in cybersecurity, may sometimes need to be changed quickly.
Health data, consent and power
The CNIL notes that health data includes information revealing a person’s physical or mental health status. The risk is therefore not limited to someone hacking a number. It also concerns purpose, hosting, recipients, retention periods and inferences drawn from measurements.
The questions to ask a provider are simple:
- what data leaves the phone and in which country it is hosted
- whether an account or subscription is required to read the implant
- how to export, correct and delete the data
- whether it is used to train an algorithm or shared with partners
- what happens when the company closes or changes its terms
Social pressure deserves as much attention as the technology. An implant chosen to open your own door does not carry the same ethical implications as one requested by an employer, gym or insurer. Consent is not freely given if refusal leads to a professional or financial disadvantage.
Our guide to health data and artificial intelligence · FR helps distinguish useful collection from disproportionate surveillance.
A decision checklist before getting an implant
The experiment is only justified if it solves a real problem better than an external alternative. Ask yourself these five questions:
- what specific action does the chip make easier each week
- could a card, watch or phone provide the same service without breaking the skin
- does the complete device have independent documentation for its claimed use
- can you revoke access and retrieve your data without removing the implant
- would you still accept the benefit–risk balance if the provider disappeared tomorrow
For most people, the answers point toward trying an external solution first for several months. Record the number of actual uses, failures and situations where the identifier needs to be revoked. This log helps distinguish an everyday function from a technological curiosity. It measures real usefulness without turning the trial into surgery.
Verdict
Passive NFC implants are useful but limited identifiers. Battery-free biomedical implants are a credible field of research, still constrained by biocompatibility, energy, calibration and regulation. Confusing the two feeds a futuristic vision that the evidence does not support.
By 2030, targeted medical uses may advance. That makes widespread implantation neither likely nor desirable. The right criterion is not being the first to wear the technology, but checking whether it provides information or a function useful enough to justify a wound, a foreign body and a new digital attack surface.
