Ethical and Societal Implications in Advancements
Linas JuozenasShare
Human-Performance Enhancement: Access, Equity & Fair Play
Technology can expand what people are able to do. Its value also depends on who can benefit, who carries the risks, and whether people remain free to choose.
A well-fitted assistive device can support independence. A training tool can make useful feedback easier to obtain. A new treatment can open possibilities that once seemed out of reach. Each deserves careful attention to the person and purpose it serves.
This guide examines enhancement across healthcare, sport, education and work. It connects scientific evidence with questions of dignity, fairness, privacy and public responsibility—and offers a practical framework for making decisions when those values pull in different directions.
Specify what improves, for whom, compared with which alternative, and for how long.
Users, workers, athletes and communities should help define success and acceptable trade-offs.
Access to opportunity should not quietly become conditional on unwanted changes to a person’s body or data.
“Enhancement” covers very different things
Human-performance enhancement is a broad term for interventions intended to improve a capability: movement, strength, endurance, attention, communication or another function. It includes tools used outside the body, treatments acting within it, and changes to the environments in which people live.
The word can conceal important differences. Treating a medical condition, supporting a disabled person’s chosen activity, improving an athlete’s equipment and experimenting with a biological intervention involve different evidence, goals and responsibilities.
The boundary between treatment and enhancement is sometimes contested. Restoring a function, accommodating a difference and extending a capability can overlap. Describing the actual intervention and purpose is more useful than assuming that one label resolves its ethics.
| Approach | A possible purpose | What must be established |
|---|---|---|
| Wearables and AI-based feedback | Measure activity or help adjust a training decision. | Accuracy in the intended setting, usefulness of the recommendation, and the effects of acting on it. |
| Prostheses, orthoses and exoskeletons | Support movement, a task or rehabilitation. | Fit, training, practical benefit and support for the particular user; device-assisted movement is not proof of neurological recovery. |
| Medicines and nutritional interventions | Treat a condition or support a specific performance goal. | Evidence for that use, adverse effects and applicable medical, food and sporting rules. |
| Gene and cell technologies | Treat selected diseases or investigate new biological approaches. | The exact intervention and indication; results in a disease or animal model cannot establish safe enhancement in healthy people. |
| Neurotechnology and immersive systems | Support communication, rehabilitation or task practice. | What the system actually measures or changes, and whether benefits transfer to the real-world activity. |
Some uses have substantial clinical experience; others remain experimental. FDA approval of a CRISPR-based treatment for sickle cell disease in 2023, for example, established a specific medical use. It did not validate genetic enhancement of athletic ability.[1]
This article uses human dignity, autonomy, benefit, harm and justice as starting values, consistent with UNESCO’s bioethics declaration.[2] The practical recommendations below are ethical judgments informed by evidence. They are not claims that every jurisdiction has adopted the same rules.
Evidence is part of the ethical question
A person cannot make an informed choice if a sales claim makes the benefits appear more certain than they are. Overselling a product can waste money, divert care, create unrealistic expectations or shift responsibility for failure onto the user.
Begin with an outcome someone has reason to value: communicating a preference, moving around home, completing a task with less discomfort, or improving a defined sporting performance. A change in a laboratory measurement may help explain a mechanism without establishing that outcome.
| Ask | Why it matters |
|---|---|
| What was measured? | A sensor signal, muscle size, task score and everyday independence are different outcomes. |
| Compared with what? | A useful comparison may be standard care, another device, ordinary training or doing nothing; these answer different questions. |
| Who took part? | Findings from one age group, condition, body size or training level may not transfer to another. |
| How large and durable was the benefit? | Statistical significance alone does not show that a change matters to the user or will persist. |
| What went wrong or remained unknown? | Report adverse events, withdrawals, failures and limits of follow-up alongside favourable results. |
| Who funded and evaluated it? | Commercial involvement should be transparent, and independent scrutiny helps assess the claim. |
Keep the conclusion within the study
A 2025 Cochrane review found that electromechanical-assisted gait training combined with physiotherapy probably improves the chance of independent walking after stroke. That finding concerns a defined rehabilitation context. It cannot establish that every exoskeleton restores walking in every neurological condition.[3]
Likewise, measuring reduced muscle activity during a short work task does not demonstrate fewer injuries over a working lifetime. NIOSH identifies both potential benefits and practical risks of occupational exoskeletons, including fit, balance and changes in physical loading.[4]
Ask for benefits and uncertainty in the same conversation
“Promising” should lead to a clear account of what remains to be learned. An honest explanation can acknowledge a meaningful possibility without presenting the user as guaranteed to benefit.
The level of evidence needed should reflect the stakes. A reversible preference setting and an implanted device do not justify the same tolerance for uncertainty. Long-term dependence, irreversible effects and consequences for other people strengthen the case for independent review.
Access means being able to use and keep using it
A technology can be available for sale while remaining inaccessible in practice. WHO identifies barriers to assistive technology that include cost, physical access, limited product choice, procurement difficulties, workforce gaps and insufficient funding.[5]
Access has several parts: learning that an option exists, receiving a fair assessment, obtaining a suitable product, learning to use it, and getting help when circumstances or equipment change.
The full cost belongs in the decision
Consider assessment, fitting, training, subscriptions, consumables, transport, repairs, replacement parts and time away from work. A low initial price can conceal expensive dependence on a proprietary service. A higher initial price may or may not be justified by durability and support.
For a clinical intervention, the relevant burden can also include preparation, follow-up, caregiving and travel. A clinical trial’s research budget, a treatment’s market price and what an individual pays are different quantities; quoting one as another misleads readers.
Before access
Can people obtain information, assessment and a realistic explanation in a language and format they can use?
During use
Does the product fit, and are training, maintenance and practical assistance available?
Over time
What happens when a battery fails, a subscription ends, the user’s needs change or the supplier withdraws support?
Affordability and fairness are connected
When an institution makes a tool effectively necessary for participation, the question changes. It should explain how people without the means to obtain it can still take part. Loan schemes, shared facilities or accessible alternatives may help, depending on the activity.
These are policy options, not guarantees that one arrangement will work everywhere. A shared device may improve access to a training tool but be unsuitable where individual fitting or reliable daily availability is essential.
Fair access also involves opportunity cost: what other services could the same resources support? That comparison should consider quality of life and unmet need, without reducing a person’s worth to their productivity.
Disability, healthcare and the meaning of progress
People should have a meaningful role in deciding what assistance or treatment is useful to them. Goals can include comfort, communication, employment, sport, caring for family, moving independently or simply having more choice in daily life.
A technological demonstration may prioritise standing or walking while its intended user prioritises travelling comfortably, managing fatigue or getting through a normal day. Neither appearance nor resemblance to a non-disabled body is a sufficient measure of success.
The Convention on the Rights of Persons with Disabilities places participation at the centre of policy: its Article 4 calls for close consultation with and active involvement of disabled people through their representative organisations.[6]
Support the person and the environment
A device may expand a person’s options, while steps at an entrance, inaccessible transport or an inflexible workplace still restrict participation. Addressing these barriers and providing assistive technology can be complementary choices.
Choosing a wheelchair, a prosthesis, a communication aid or no additional technology should not become a test of ambition. People can value treatment and also resist the idea that their existing life is deficient.
Good provision includes an ongoing relationship
WHO’s wheelchair provision guidelines emphasise individual assessment and selection, fitting, training and follow-up.[7] The practical lesson is to plan for the service around a product, including changes in the user’s needs.
Coverage decisions should explain how benefits, evidence, alternatives and continuing support are assessed. A treatment-versus-enhancement label may be relevant, but it should not substitute for a transparent account of the person’s functional needs and goals.
“What would make this worthwhile in your own life?” Ask it before selecting the outcome measures, and return to it after the technology has been used outside a demonstration.
Global access depends on local capability
Income differences matter, but a simple division between wealthy and poorer countries misses much of the practical picture. People living in the same city can face very different barriers; rural distance, disability, language, migration status and unreliable services may intersect.
A product that depends on continuous internet access, imported consumables or a distant specialist can be difficult to sustain in many settings. The same problem can arise when a supplier stops supporting an older model.
Build a service people can maintain
Before introducing a technology, ask who can fit it, explain it, repair it and supply replacement parts. Establish how the service will continue after a research project, donation or introductory subsidy ends.
Local clinicians, technicians, users and community organisations bring knowledge that a design team may lack: transport realities, household space, work practices, language and the availability of assistance. Their involvement should influence decisions early enough to change the design.
WHO’s assistive-technology approach connects people with policy, products, provision and personnel.[8] A box delivered to a clinic addresses only part of that system.
Avoid transferring hidden burdens
Illustrative example: a donated powered device arrives with no funded maintenance or locally available battery. Its purchase cost has been removed, but users and staff may inherit travel costs, downtime and an unusable product. A simpler serviceable option could offer greater practical benefit in that setting.
Research partnerships should also agree who helps set priorities, who can access findings, how local contributors are recognised and what happens after the study. These questions concern shared decision-making as well as the distribution of equipment.
Technology transfer is more useful when it strengthens the ability of people and institutions to make their own choices. That may involve training, repair documentation, adaptable designs or locally negotiated support.
Inclusive evidence requires more than an average result
A device can perform well on average while fitting poorly or producing less reliable measurements for some intended users. Evaluation should examine relevant variation in body dimensions, age, sex, disability, skin pigmentation, movement patterns and conditions of use.
These characteristics are not interchangeable. Skin pigmentation is not a substitute for race or ethnicity; sex and gender can raise different questions; two people with the same diagnosis may need different support.
Distinguish measurement, prediction and decisions
A sensor can make an error when recording a signal. An algorithm can make an error when interpreting that signal. An institution can then use the result in a way that disadvantages someone. These are connected but distinct problems.
For example, Bent and colleagues’ 2020 study of optical heart-rate wearables found no statistically significant accuracy differences across skin tones in its sample, while finding differences associated with device and activity. That result neither proves every device is equally accurate for everyone nor supports a blanket claim that all wearables perform poorly on darker skin.[9]
The practical demand is appropriate validation and honest reporting of uncertainty for the specific product and use. Avoid transferring findings from facial recognition or pulse oximetry directly to a different sensor without evidence.
Participation should influence what gets measured
The 2024 Declaration of Helsinki calls for appropriate access to research participation for underrepresented groups.[10] Inclusion still requires scientifically justified eligibility criteria and protection of participants; demographic variety alone does not make a study valid.
Recruitment materials, accessible visits, travel support and usable instructions can affect who is able to participate. Outcomes should reflect what matters to users, including fatigue, comfort, independence and the burden of maintaining a device.
For AI, the NIST risk-management framework treats validity, reliability, transparency and harmful bias as connected concerns.[11] It is a voluntary framework, not a worldwide legal requirement. An institution still needs to explain how errors will be detected, challenged and corrected in its own setting.
Fairness has several meanings
In healthcare, a central question is whether people can obtain appropriate care. In education, it may be whether an assessment gives a fair opportunity to demonstrate the intended learning. In sport, it includes what abilities the contest is organised to compare. At work, it includes safety, reasonable opportunity and freedom from unjustified intrusion.
The same device can have different ethical significance in these settings. A communication aid may make participation possible. Automated assistance in a competition may alter what the event measures. Context and purpose need to be explicit.
Equal treatment and equitable opportunity
Giving everyone the same equipment can appear equal while excluding people who cannot use it. An accommodation can improve a person’s opportunity to demonstrate the relevant skill without making every participant’s experience identical.
Conversely, equal access to a risky intervention would not by itself establish that requiring it is acceptable. Affordability, safety and freedom of choice are separate considerations.
The “natural” label cannot do all the work
Training methods, coaching, equipment and medical care already shape performance. A technology’s artificial origin does not alone explain why it should be permitted or prohibited.
A more useful discussion asks which skills and challenges an activity should preserve, how large an advantage is relevant, what risks are acceptable, and whether a rule can be applied consistently. People may disagree about those values even after agreeing on the scientific evidence.
Make disagreement reviewable
A fair decision process explains its purpose, criteria and evidence; hears affected people; handles conflicts of interest; and provides a proportionate route to review or appeal. Rules should be communicated before people make major investments in equipment or training.
Where evidence is uncertain, decision-makers should state what is unknown and when the decision will be reconsidered. Neither permanent exclusion nor unrestricted permission follows automatically from uncertainty.
Sport, anti-doping and assistive equipment
Competitive sport uses several rule systems. Anti-doping rules address prohibited substances and methods. Technical rules govern equipment and event conditions. Eligibility and classification rules address who competes and in which category. Permission under one system does not settle the others.
What the anti-doping framework says
Under the 2021 World Anti-Doping Code, inclusion on the Prohibited List can follow when a substance or method meets at least two of three criteria: potential performance enhancement, actual or potential health risk, and conflict with the spirit of sport. Masking potential is an additional basis. The Code also places responsibility on athletes for prohibited substances found in their samples, with consequences assessed under its rules.[12]
As of 5 September 2026: the 2026 WADA Prohibited List is in force. Its M3 category prohibits specified uses of nucleic acids, cells and cell components with the potential to enhance sporting performance, including gene-editing, gene-silencing and gene-transfer approaches. Gene and cell doping are therefore already addressed; this is not merely a future proposal.[13]
A therapeutic use exemption protects legitimate treatment
A therapeutic use exemption (TUE) can authorise a prohibited substance or method for a medical condition when the required criteria are met. In summary: the condition must be supported by clinical evidence; treatment should not add enhancement beyond the expected return to normal health; there must be no reasonable permitted therapeutic alternative; and the need must not result from prior prohibited use without a TUE.
A prescription alone does not establish a TUE. Application, recognition and timing depend on the athlete’s circumstances and responsible anti-doping organisation. A TUE is also distinct from permission to use particular equipment.[14]
Assistive equipment deserves specific assessment
Using a prosthesis is not evidence of cheating. The relevant questions concern the sport’s purpose, its technical rules and evidence about the particular equipment and event. Appearance or assumptions about disability cannot substitute for that assessment.
Paralympic classification is sport-specific because impairments can affect different sporting tasks differently. It is not a universal scale of how much technological assistance an athlete receives.[15]
Athletes should obtain clarification from the relevant governing body before relying on a treatment or equipment choice for competition. General legality, medical authorisation and sporting permission answer different questions.
Clear rules, accessible medical-exemption procedures, privacy and a meaningful opportunity to challenge a decision matter alongside the competitive outcome.
Consent needs understanding and a real alternative
A signature records a decision; it does not establish that the decision was informed or free from pressure. Someone may agree because a coach controls selection, an employer controls income, or a service appears to offer the only route to an important opportunity.
Ask what would happen if the person declined. Would they lose ordinary access, be treated as less committed, face worse shifts or feel obliged to reveal private information? These are questions to investigate in the actual setting, not assumptions that every programme is coercive.
Explain the choice in terms the person can use
A useful consent conversation covers the purpose, realistic benefits, known risks, uncertainty, alternatives, costs, data use and continuing responsibilities. It should allow questions and time to consider the choice, with independent advice where the stakes justify it.
Understanding can be checked by inviting the person to explain the decision in their own words. This should help the explanation improve, rather than become an examination the person has to pass.
Separate decisions that do not need to be bundled
Agreeing to use a device, take part in research, share information with a coach and allow commercial reuse of data are different decisions. Optional purposes should be presented clearly rather than hidden inside access to a necessary service.
For data processing under the GDPR, the European Data Protection Board emphasises that employment power imbalances often prevent consent from being freely given. Exceptional situations can allow genuine consent without adverse consequences for refusal. Consent is one possible legal basis, not a universal solution for workplace data collection.[16]
Be honest about withdrawal
Stopping future participation may be possible even when an intervention’s effects cannot be undone. An implant may require clinical management; a biological alteration may persist; some research records may need to be retained under applicable rules. These limits should be explained before the decision.
The ethical aim is continuing agency: people should know how to raise concerns, obtain care, change an optional preference or leave a programme where possible. A person’s willingness to try a technology should never be treated as unlimited permission for later uses.
Body data can affect opportunities beyond training
Activity, location, heart-rate, genetic and neural data can be sensitive because of what they reveal—or what someone claims to infer from them. A mistaken inference about health, attention or reliability can still influence decisions about a person.
Useful feedback does not automatically justify sharing the underlying data with an employer, insurer, school, advertiser or selection committee. The purpose and recipients deserve their own assessment.
Understand the legal categories
Where the GDPR applies, personal-data processing needs a lawful basis and must follow principles including purpose limitation, data minimisation and appropriate security. Health and genetic data have additional protection under Article 9; biometric data fall within that article when processed to uniquely identify a person. Not every fitness reading is automatically biometric identification data.
Consent is not the only possible basis, and special-category processing also needs an applicable Article 9 condition. Rights concerning access, erasure or portability have conditions and exceptions; an organisation should explain what applies to its service.[17]
| Decision | A clear explanation should cover |
|---|---|
| Collection | Which measurements and identifiers are necessary for the stated service? Can optional collection be switched off? |
| Interpretation | Which conclusions are estimates? How are errors or missing data handled before they affect a person? |
| Access and reuse | Who receives raw data, summaries or predictions? Are research, advertising and model-training uses separate? |
| Retention and security | How long are records kept, how is access controlled, and what happens after an account closes or a breach occurs? |
| Challenge and exit | Who can correct a record or review a decision? What happens to data and essential functions if the person leaves? |
Minimise dependence as well as disclosure
A device’s usefulness can depend on cloud services, updates or an account. Before making it central to daily life, ask which functions remain if a connection fails or the provider changes its terms.
Removing a name from a dataset does not by itself make re-identification impossible. The information available, the way records are combined and the safeguards in place matter. Claims such as “anonymous” or “secure” should have a concrete explanation.
A dashboard should support a discussion about a person’s situation. Treating a single readiness or attention score as a complete judgment of character, effort or capability gives the measurement authority it has not earned.
Work and education should not require endless optimisation
Enhancement becomes an institutional issue when organisations set the conditions under which people work or learn. A voluntary tool can become effectively compulsory if those who decline are penalised, excluded or expected to meet targets achievable only with it.
Improve the task as well as the equipment
NIOSH’s hierarchy of controls gives priority to removing hazards, replacing them or controlling them through engineering before relying on measures further down the hierarchy.[18] This supports asking whether a heavy or repetitive task can be redesigned before expecting workers to compensate with a wearable device.
Illustrative example: a warehouse proposes an exoskeleton trial. A useful evaluation would involve workers, consider the full shift and different tasks, assess practical difficulties, and compare alternatives such as lifting aids or changes in workflow. A short demonstration of lower effort would not justify an automatic increase in workload.
Workers should have an accessible route to report problems without being blamed for poor performance. Procurement and safety decisions should identify who is responsible for acting on those reports.
Protect learning and legitimate support
In education, distinguish learning support, disability accommodation and changes that undermine what an assessment intends to measure. The school should explain the intended skill, permitted tools and accommodation process in advance.
Claims that a drug, stimulation device or attention tracker will produce better grades require evidence for that specific use. Pressure to adopt such products can be an ethical problem even when their advertised benefit is uncertain.
Some boundaries are already law
In the EU, Article 5(1)(f) of the AI Act prohibits the specified placing on the market, putting into service or use of AI to infer emotions in workplaces and education institutions, with exceptions for medical or safety reasons. This prohibition has applied since February 2025. Its scope should not be confused with a blanket ban on every physiological or fatigue measurement.[19]
Hierarchical settings, including military organisations, raise further questions about refusal, confidentiality and long-term care. A claimed institutional benefit does not settle whether an intervention is acceptable for the people expected to undergo it.
Genetic and neural technologies require careful distinctions
Somatic and heritable editing are different
Somatic editing targets cells in an existing person’s body, outside the reproductive line. Such changes are generally not passed to their children. Editing reproductive cells or embryos in a way that leads to an inherited change raises distinct questions about descendants and future generations.[20]
Gene therapy, gene editing, cell therapy and mRNA-based approaches are not interchangeable terms. Nor does using a nucleic-acid technology automatically mean that the person’s genome, or that of their descendants, has been permanently edited.
Animal findings about muscle growth or a successful treatment for a defined disease do not establish safe, predictable enhancement of strength, endurance or coordination in healthy people. Complex performance depends on many biological and environmental factors.
WHO’s genome-editing recommendations address oversight, research registries, international cooperation and public engagement across somatic, germline and heritable applications.[21] They are a governance framework, not permission for unregulated enhancement.
Neurotechnology is not a single capability
A brain-computer interface uses neural signals to help interact with another system. Brain stimulation attempts to influence neural activity. A virtual training environment changes the experience of practising a task. Evidence from one approach cannot validate the others.
FDA guidance for implanted brain-computer interfaces addresses nonclinical testing and clinical studies for people with paralysis or amputation.[22] The existence of that guidance is not evidence that a marketed product can read anyone’s thoughts or safely increase a healthy person’s intelligence.
UNESCO’s 2025 Recommendation on the Ethics of Neurotechnology addresses autonomy, mental privacy, inclusion and responsible use across the technology’s life cycle.[23] It is an international ethical recommendation, not a directly enforceable worldwide product law. Questions about ongoing support, access to neural data and the ability to stop using a system deserve attention before a person becomes dependent on it.
Protect children’s future options
Children’s developing ability to understand should shape how they participate in decisions, alongside the applicable protections and parental responsibilities. A desire for future success does not establish a present medical need.
A genetic “talent” label can narrow expectations long before a child has explored an activity. A 2015 international consensus statement rejected the use of then-available direct-to-consumer genetic tests for sports talent identification or individualised training prescription.[24] New products still need their own evidence; a label or updated algorithm does not supply it.
Across these technologies, people’s value should remain independent of their willingness or ability to optimise. Choosing rest, ordinary training, support or no intervention can be a considered decision.
Governance should distribute benefits and responsibility
Public policy can influence which technologies are developed, what evidence is required, how services are funded and who can challenge a harmful decision. Commercial incentives also shape design: a subscription model, proprietary part or data-dependent service can influence the user’s choices long after purchase.
There is no single arrangement that resolves access, innovation, safety and cost in every setting. The following are policy and procurement options, not a list of universal legal duties.
| Option | Potential contribution | What still needs attention |
|---|---|---|
| Public funding with access conditions | Can link investment to affordability, service continuity or sharing of findings. | Conditions need clear measures, workable enforcement and attention to development and maintenance costs. |
| Loan schemes and shared provision | Can reduce initial costs for suitable products. | Availability, individual fitting, maintenance and reliable access must match users’ needs. |
| Interoperability and repair support | Can reduce dependence on a single supplier and extend useful life. | Safety, responsibility for modifications, cybersecurity and sustainable support still matter. |
| Transparent evaluation and reporting | Can expose limits, adverse outcomes and conflicts of interest. | Reporting should protect participants’ privacy and be understandable enough to inform decisions. |
| User and community participation | Can improve priorities, practical fit and accountability. | Participation needs resources and actual influence; one invited representative cannot speak for everyone. |
Research responsibilities continue after enrolment
The 2024 Declaration of Helsinki calls for advance arrangements for participants who still need an intervention found beneficial and reasonably safe in a trial, with exceptions subject to research-ethics review.[10] This makes continuing access part of research planning rather than an afterthought.
More broadly, organisations should clarify responsibility when a product fails, an algorithm changes or a supplier exits. Open-source software and published designs can help some forms of scrutiny and repair; they do not remove the need for testing, maintenance or accountable clinical decisions.
What about separate technology categories or “open” leagues?
Equipment categories may sometimes help a sport define a coherent contest. They can also divide participation, increase administrative burdens or disadvantage athletes whose devices do not fit a simple category. Any proposal needs sport-specific evidence and involvement from affected athletes.
A league allowing additional enhancements would not automatically make participation safe or consent free from pressure. Commercial incentives, access inequalities and responsibility for long-term harm would remain. Medical and other applicable laws would still need to be considered.
Governance should include a way to revise decisions as evidence changes. A review date, published reasons and a route to report problems are more useful than a promise to be “ethical” without assigned responsibilities.
A practical framework for people and organisations
Use these five steps when considering a device, treatment, programme or policy. The depth of review should match the consequences. The framework is a way to organise judgment, not a numerical score that can cancel out a serious problem.
-
Define the purpose and whose goal it is.
Describe the task or outcome in plain language. Ask the intended users whether that outcome matters to them. Identify affected people beyond the direct user.
-
Match the evidence to the claim.
Specify the intervention, comparison, intended population and follow-up. Separate demonstrated benefits from plausible mechanisms and unanswered questions. Consider available alternatives.
-
Examine choice, access and distribution.
Ask who benefits, who pays, who carries the risks and what happens to those who decline. Include the full service costs and any effect on existing inequalities or essential support.
-
Establish responsibilities and applicable rules.
Clarify consent, data handling, medical or equipment requirements and any sporting conditions. Name who will provide support, act on problems and review contested decisions.
-
Decide, document and revisit.
Record the reasons for proceeding, changing the proposal, limiting it to research or declining it. Agree what outcomes will be reviewed, when, and what would prompt a change or stop.
Worked example: a team proposes mandatory readiness tracking
This is an illustrative scenario. A sports team wants every athlete to upload sleep and heart-rate data to help plan training. The aim may be reasonable, but the proposed system still needs assessment.
The team should establish whether the device and algorithm support the intended decision, who can see the data, and what happens when readings are missing or an athlete declines. It should consider whether voluntary summaries or a direct conversation could meet the need with less intrusion.
A trial could then evaluate actual usefulness and burden, with athletes involved and a defined review point. A coach should be able to reconsider an algorithm’s suggestion in light of the athlete’s experience. This process may support adoption, a narrower use or rejection of the proposal; the answer depends on the evidence and arrangements.
What each role can contribute
Individuals and families can ask for realistic benefits, alternatives and continuing costs. Clinicians and coaches can keep the person’s goals central and explain uncertainty. Developers can test intended uses and plan for accessibility and ongoing support.
Employers, schools and sports bodies can set transparent boundaries and protect a meaningful way to raise concerns. Funders and policymakers can make evidence, access and accountability part of the decisions they control.
Responsibility should follow power. An individual cannot resolve every problem created by a supplier, institution or market; organisations that set the conditions of use must carry their share.
Answers to recurring dilemmas
Is human-performance enhancement inherently unethical?
No. Its ethical significance depends on its purpose, evidence, risks, effects on others and conditions of use. A technology can expand valuable choices while still requiring careful decisions about access, privacy or competition.
Where is the boundary between treatment and enhancement?
It is sometimes disputed and context-dependent. State the medical condition or functional goal, the proposed intervention and the expected benefit. A label can help organise a discussion, but it does not replace the evidence or the person’s priorities.
Does a prosthesis automatically give an unfair sporting advantage?
No. Any competitive assessment needs the relevant sport’s rules and evidence about the device, athlete and event. Disability accommodation, technical equipment rules and classification are distinct matters.
Does a prescription make a medicine permitted in sport?
A prescription establishes neither anti-doping permission nor a TUE. Athletes subject to anti-doping rules should check the applicable requirements with their responsible organisation and treating clinician.
Are all gene therapies inherited by future children?
No. Somatic interventions generally affect the treated person rather than the reproductive line. Heritable editing is a distinct application with additional implications for descendants. The exact technology matters.
Is an opt-in button enough for workplace consent?
The surrounding conditions matter. Refusal must be genuinely possible without inappropriate disadvantage. For GDPR data processing, consent may be unsuitable where a power imbalance prevents a free choice; another lawful basis cannot simply be assumed.
Would making a technology cheaper solve the equity problem?
It could help, but information, fitting, training, physical access, maintenance and continuing support also matter. A product someone can buy may still be difficult or impossible for them to use.
Can an AI score decide whether someone is fit, motivated or reliable?
A score represents particular measurements and assumptions. Its validity must match the decision being made, and uncertainty and context matter. A consequential decision should have an accountable process for review and correction.
Would an open enhancement league remove the ethical concerns?
It could change the competitive rules, but would not remove questions about safety, pressure, affordability, long-term care or applicable law. Formal permission alone does not establish informed, freely chosen participation.
What is the most useful first question?
Ask: “What meaningful improvement is this expected to make in this person’s life or activity?” Then ask what evidence supports that expectation and what the person would give up, pay or risk to obtain it.
Sources, scope and keeping this guide useful
The sources below include scientific research, clinical and technical guidance, legal texts and ethical frameworks. Their roles differ: a study can inform an expected outcome; a law can set a requirement within its scope; an ethical framework can help explain a judgment. None alone settles every question in this article.
Sources and selected rules checked: 5 September 2026. The WADA discussion refers to the 2026 Prohibited List and the 2021 Code. EU examples describe selected GDPR and AI Act provisions; they are not a global legal summary. Check the rules in force for the relevant place, product and activity when making a decision.
The examples and decision framework are educational illustrations. They do not recommend an individual medical intervention or determine an athlete’s eligibility. Performance claims should be reconsidered when stronger evidence becomes available.
- US Food and Drug Administration. FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease. 8 December 2023.Historical approval announcement used to distinguish a defined medical indication from athletic enhancement, not to describe every current indication. Return to text ↑
- UNESCO. Universal Declaration on Bioethics and Human Rights. Adopted 19 October 2005.Ethical framework concerning dignity, autonomy, benefit, harm, justice and discrimination. The article's examples apply these ideas; they are not quotations or legal determinations. Return to text ↑
- Mehrholz J, Kugler J, Pohl M, Elsner B. Electromechanical-assisted training for walking after stroke. Cochrane Database of Systematic Reviews. 2025;CD006185. DOI: 10.1002/14651858.CD006185.pub6.Updated synthesis of gait training in adults after stroke. Independent walking is a particular outcome; benefits do not transfer automatically to every robotic task. Return to text ↑
- National Institute for Occupational Safety and Health. Exoskeletons: Potential for Preventing Work-related Musculoskeletal Injuries and Disorders in Construction Workplaces. NIOSH Science Bulletin. 2022.Discusses potential benefits, implementation questions and risks. Short-term workload measures alone cannot establish long-term injury prevention. Return to text ↑
- World Health Organization. Assistive technology. Fact sheet, 2 January 2024.Identifies barriers to access and the importance of services around assistive products. Return to text ↑
- United Nations. Convention on the Rights of Persons with Disabilities — Article 4: General obligations. Convention adopted 13 December 2006.Article 4(3) concerns consultation and active involvement of disabled people. Treaty obligations apply to States Parties within the relevant scope. Return to text ↑
- World Health Organization. Wheelchair provision guidelines. Guideline. 2023.Emphasises assessment, fitting, training and follow-up through services responsive to the user’s needs. Return to text ↑
- World Health Organization. Assistive Technology: policy, products, provision and personnel. Official programme overview.Describes a system centred on people and the services needed for effective access. Return to text ↑
- Bent B, Goldstein BA, Kibbe WA, Dunn JP. Investigating sources of inaccuracy in wearable optical heart rate sensors. npj Digital Medicine (2020), 3:18.Original study of specific wearable optical heart-rate devices. No statistically significant accuracy differences across skin tones were found in its sample; this does not establish equal performance for every device or population. Return to text ↑
- World Medical Association. Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Participants. Revised 19 October 2024.Professional research-ethics declaration. Relevant provisions address independent review, consent, reporting, responsibility and post-trial arrangements; it is not itself a worldwide statute. Return to text ↑
- National Institute of Standards and Technology. Artificial Intelligence Risk Management Framework (AI RMF 1.0). NIST AI 100-1. 2023.Voluntary cross-sector framework covering trustworthy AI and context of use. It is not evidence that a particular fitness algorithm is effective. Return to text ↑
- World Anti-Doping Agency. World Anti-Doping Code 2021. Code effective 1 January 2021.Selected Articles 1–2, 4 and 10: violations, List criteria, TUEs and consequences. The Code differs from the annual List. Return to text ↑
- World Anti-Doping Agency. The 2026 Prohibited List. International Standard; effective 1 January 2026.2026 International Standard, including M3 gene and cell doping. Applicable sporting requirements must be checked when a decision is made. Return to text ↑
- World Anti-Doping Agency. Therapeutic Use Exemptions (TUEs). Athlete guidance and questions and answers.Explains criteria, application, recognition and retroactive exceptions. Approval depends on the individual case. Return to text ↑
- International Paralympic Committee. Classification. Official overview; checked 5 September 2026.Explains sport-specific classification. Individual eligibility and equipment decisions require the relevant federation rules. Return to text ↑
- European Data Protection Board. Guidelines 05/2020 on consent under Regulation 2016/679. Adopted 4 May 2020; version 1.1.Official GDPR interpretation, including employment power imbalances and genuine choice. Data-processing consent and consent to an intervention are distinct. Return to text ↑
- European Parliament and Council of the European Union. General Data Protection Regulation — Regulation (EU) 2016/679. 27 April 2016; selected provisions checked.Articles 4–9 and related provisions inform the discussion of personal data, processing principles, lawful bases and special categories. Dataset-specific assessment remains necessary. Return to text ↑
- National Institute for Occupational Safety and Health. Hierarchy of Controls. Official occupational-health guidance. 2024.Prioritises controls that remove or reduce hazards through the work system. Specific interventions require task-appropriate assessment. Return to text ↑
- European Parliament and Council of the European Union. Artificial Intelligence Act — Regulation (EU) 2024/1689. Consolidated text dated 27 July 2026.Article 5(1)(f) addresses emotion inference in work and education, with medical or safety exceptions. This prohibition has applied since 2 February 2025. Return to text ↑
- World Health Organization. Human genome editing. Official topic overview.Distinguishes somatic, non-reproductive germline and reproductive germline applications. Return to text ↑
- World Health Organization. WHO issues new recommendations on human genome editing for the advancement of public health. 12 July 2021.Official account of WHO's governance framework and recommendations. Distinguishes somatic, germline and heritable editing; international recommendations do not create uniform national law. Return to text ↑
- US Food and Drug Administration. Implanted Brain-Computer Interface (BCI) Devices for Patients with Paralysis or Amputation — Non-clinical Testing and Clinical Considerations. Guidance, 20 May 2021.Guidance for device development and clinical investigations, not approval of a general cognitive-enhancement claim. Return to text ↑
- UNESCO. Recommendation on the Ethics of Neurotechnology. Adopted 11 November 2025.International ethical recommendation addressing the technology's whole life cycle. A normative framework, not a directly enforceable worldwide product statute. Return to text ↑
- Webborn N, Williams A, McNamee M, et al. Direct-to-consumer genetic testing for predicting sports performance and talent identification: Consensus statement. British Journal of Sports Medicine, 2015;49:1486–1491; doi:10.1136/bjsports-2015-095343.Historical consensus on the lack of a supported role for commercial genetic talent testing. New products still require appropriate validation. Return to text ↑
Let greater capability widen human choice.
Human achievement includes independence, connection, play, meaningful work and the freedom to shape a life. A faster time or a higher score can matter, but it cannot capture all of those goods.
Enhancement deserves support when there is a sound reason to expect benefit and when the conditions of use respect the people affected. That requires evidence, accessible services, honest consent and institutions willing to take responsibility for their decisions.
The enduring task is to make new capabilities serve lives people have reason to value—and to preserve dignity and opportunity for those who choose a different path.