Transhumanism, Enhancement, and the Ethics of Human Capacity

The Cursor And The Question Of Capacity

In early 2024, Noland Arbaugh became the first publicly identified human participant with a Neuralink brain-computer interface implant. Arbaugh had paralysis below the shoulders after a diving accident. In a public demonstration, he used the implant to move a computer cursor and play online chess. Reuters reported the demonstration as promising but also quoted expert caution that the technology was still early. Neuralink’s own Device Control page and the ClinicalTrials.gov record for NCT06429735 describe the project as a clinical-trial pathway for a brain-computer interface intended to let people with paralysis control external devices.

This is the right place to begin a chapter on transhumanism because the case is not a cartoon about humans becoming machines. It begins with restoration. A person who had lost ordinary bodily control regained a form of digital agency: moving a cursor, playing a game, communicating, browsing, and acting in a digital environment without needing the same physical assistance. Before the case raises science-fiction questions about enhanced minds or human-AI merger, it raises a much simpler and more humane question: what should technology do when a human capacity has been lost, blocked, impaired, or made difficult by illness, injury, aging, disability, or social design?

The case also needs guardrails. Neuralink did not invent implanted brain-computer interfaces. Earlier systems such as BrainGate had already shown that people with paralysis could use implanted neural signals to control computer cursors or devices; Wired reported on BrainGate’s peer-reviewed 2006 results years before Neuralink existed. The Arbaugh case was also experimental. The Guardian reported that some implant threads retracted after the procedure, reducing functionality before software changes restored some performance. Those details matter ethically. They keep the opening from becoming either a miracle story or a horror story. The technology is real. The benefit can be life-changing. The risks and uncertainties are also real.

The case gives this chapter its central question:

When a technology promises to restore, extend, optimize, replace, or redesign a human capacity, what picture of the human being is it assuming, and what ethical pressures does that picture create?

That question is broader than Neuralink. An AI tutor changes a student’s learning capacity. A stimulant changes attention and wakefulness. A prosthetic limb restores and may sometimes exceed ordinary mobility. Genetic editing can prevent disease or aim at traits one generation wants in the next. A mental-health app can support mood regulation or reshape what counts as normal emotional life. Longevity companies promise longer healthspan. AI labs describe systems that might accelerate biology, neuroscience, medicine, and human welfare. All of these projects ask, in different ways, what human beings should be able to do.

Transhumanism is the movement and philosophical family most strongly associated with the answer: human limits are not automatically sacred. If technology can safely reduce suffering, extend life, improve cognition, expand agency, or open forms of flourishing that current humans cannot yet access, then the mere fact that a limit is natural is not enough to make that limit morally binding. Transhumanism is not just enthusiasm for gadgets. At its strongest, it is an argument about human possibility.

The argument is powerful. It is also incomplete. The same technology that restores agency can become a tool of pressure, inequality, surveillance, or normalization. A brain-computer interface for someone with paralysis looks morally compelling. A workplace brain-monitoring system that tracks attention, fatigue, or emotional response looks different. A drug that helps someone with severe narcolepsy stay awake is one kind of case. A school or employer that quietly expects everyone to enhance wakefulness in order to compete is another. A reproductive technology that prevents a serious disease is one thing. A market that encourages parents to design children toward narrow social ideals is another.

This chapter treats transhumanism as a human-capacity lens. A lens is not a verdict. It helps you notice something. The transhumanist lens asks what human limits a project treats as problems and what capacities it promises to expand. The ethical task is to test that promise against dignity, meaningful limits, access, disability justice, cognitive liberty, and governance.

By the end of the chapter, you should be able to use this lens on a documented case. Ask: what human capacity does this AI or technology affect? Is it restoring, enhancing, optimizing, replacing, or redesigning that capacity? Who chooses the change? Who is pressured into it? Who gets access? Whose body, mind, mood, speed, or ability becomes the new standard? What condition or safeguard should affect your judgment?

What Counts As Human Enhancement?

The phrase human enhancement can sound extreme, but the basic idea is familiar. Human beings have always used tools, medicine, training, education, social practices, and institutions to improve what they can do. Eyeglasses enhance sight. Vaccines strengthen disease resistance. Writing extends memory. Calculators extend calculation. Exercise changes strength and endurance. Schools train attention, literacy, and judgment. A wheelchair can expand mobility. A prosthetic limb can restore action. A language app can scaffold communication. A music lesson develops perception, coordination, and expressive skill.

The hard questions begin when the improvement reaches bodies, brains, reproduction, identity, competition, access to opportunity, or the baseline of what society expects. The Stanford Encyclopedia of Philosophy entry on human enhancement notes that enhancement ethics includes more than a simple debate about “technology beyond nature.” It includes questions about the proper limits of health care, parental and reproductive obligations, fairness in competition, moral agency, distributive justice, science policy, and regulation. In other words, enhancement is not one issue. It is a cluster of issues around human capacity.

A useful working definition is this: an enhancement is an intervention that improves a human capacity beyond what is required to restore or sustain health. That definition is common in the bioethics literature, but the line is not always easy to draw. The SEP entry stresses that there are no enhancement technologies by themselves. Whether an intervention counts as therapy or enhancement depends on how it is used.

A brain-computer interface for a person with paralysis is usually restorative: it helps restore a lost pathway for action. The same interface sold to able-bodied gamers or analysts as a faster way to control digital systems would look more like enhancement. A medication for ADHD can be therapeutic for someone whose attention regulation creates serious impairment. The same medication used by a student without that impairment to stay competitive during finals raises enhancement and fairness questions. A prosthetic leg can restore mobility after an amputation. A blade used in elite sport can raise questions about what counts as ordinary human athletic ability. A genetic intervention might prevent a serious disease. A different use might aim at height, appearance, muscle growth, memory, or temperament.

This is why the chapter uses the broader language of capacity. A capacity is something a person can do, experience, sustain, practice, choose, or participate in. Movement is a capacity. Communication is a capacity. Memory, attention, fertility, mood regulation, sleep, learning, immune response, lifespan, emotional steadiness, sensory perception, and digital agency are capacities. Some capacities are bodily. Some are cognitive. Some are social. Some depend on infrastructure as much as biology.

The ethical question is not only whether a technology improves a capacity. It is what kind of change is being made.

Therapy / restoration

Returning a lost or impaired capacity. Example: BCI-assisted cursor control after paralysis. Student question: What is being restored, and to what baseline?

Assistance / accommodation

Changing the environment or tool so a person can act. Example: Wheelchair ramps, captions, screen readers, adaptive interfaces. Student question: Is the problem in the body, the environment, or both?

Enhancement

Increasing a capacity beyond ordinary health or species-typical functioning. Example: Stronger memory, faster reaction time, extreme endurance. Student question: Why does going beyond restoration matter here?

Optimization

Pressuring a capacity toward efficiency, speed, productivity, or measurable performance. Example: Workplace attention tracking, AI productivity coaching. Student question: Who benefits from optimization?

Replacement

A system performs a capacity instead of the person practicing it. Example: AI writing the analysis instead of helping a student learn. Student question: What human skill might weaken?

Redesign

A technology changes what counts as normal human functioning. Example: Genetic trait selection, neural interfaces as expected work tools. Student question: Whose body or mind becomes the new standard?

The table matters because many public arguments skip this step. A critic may say, “Enhancement is dangerous.” But therapy, assistance, enhancement, optimization, replacement, and redesign are not the same. A defender may say, “This technology helps people.” But helping someone regain speech is not the same as pressuring workers to use neural data to prove attention. A technology can be liberating in one context and coercive in another.

For PHIL 123, the distinction is practical. Suppose your case concerns AI tutoring. The question is not merely whether AI tutors are good or bad. Ask what capacity they affect. Do they support learning? Replace practice? Optimize performance? Create dependency? Help students with disabilities access material? Make every student feel pressure to use AI just to keep up? A human-capacity lens turns a vague opinion into a question that can guide judgment.

The same applies to health care. An AI system that helps detect diabetic retinopathy can expand diagnostic capacity. A wearable that nudges sleep, mood, diet, and exercise can support agency. But if an insurer uses the same data to price coverage or punish noncompliance, the capacity-support tool becomes a surveillance and discipline tool. The ethical meaning changes with the institution.

One more distinction is important: improving a capacity is not the same as improving a person. A person is not a bundle of upgradeable functions. Increasing memory might increase anxiety. Extending lifespan might extend dependence, inequality, or grief if not matched by health and social care. Reducing fear might weaken caution. Increasing emotional stability might flatten valuable forms of grief, anger, or moral sensitivity. Faster work might mean less reflection. A capacity can improve while a life becomes worse.

That is why this chapter does not define transhumanism as “technology good” or bioconservatism as “technology bad.” The debate is about which capacities matter, which limits matter, who chooses, and what kinds of human flourishing a design makes more likely.

The Transhumanist Case: Limits Are Not Automatically Sacred

The strongest argument for transhumanism begins with a moral refusal: suffering, disease, aging, paralysis, cognitive decline, involuntary pain, memory loss, depression, infertility, and early death are not good simply because they are natural. A child dying from an infection is natural. So is cancer. So is genetic disease. So is the frailty that can come with age. Human beings have spent centuries using medicine, sanitation, food systems, public health, education, law, and technology to push back against natural conditions that once seemed unavoidable.

Transhumanism radicalizes that ordinary human project. It asks why we should stop at returning people to the current human baseline. If we can reduce suffering further, extend healthy life, improve cognition, expand sensory perception, increase emotional resilience, or give people more control over their own bodies and minds, what exactly justifies refusing?

The Transhumanist Declaration, originally crafted in 1998 and later adopted by Humanity+, gives a movement-level statement of this view. It says that humanity may be profoundly affected by science and technology and imagines broadening human potential by overcoming aging, cognitive shortcomings, involuntary suffering, and confinement to Earth. It also recognizes serious risks, rejects the idea that all change is progress, calls for responsible deliberation, and favors wide personal choice over technologies that might assist memory, concentration, mental energy, life extension, reproduction, and other forms of modification.

Nick Bostrom’s “Transhumanist Values” gives a more philosophical version. Bostrom presents the core transhumanist value as the opportunity to explore transhuman and posthuman realms of value. That language can sound strange, but the basic idea is understandable: our current biological condition may block forms of experience, knowledge, relationship, creativity, or flourishing we cannot yet fully imagine. Just as a person born blind may not know color experience from the inside, current humans may lack access to forms of value that would require greater cognitive, emotional, bodily, or sensory capacities.

Bostrom also lists conditions and values that complicate the caricature of transhumanism as reckless self-upgrade. He names global security, technological progress, and wide access as basic conditions. He emphasizes individual choice, morphological freedom, research, critical thinking, diversity, and concern for the well-being of sentient beings. In other words, the strongest transhumanist case is not merely “make humans more powerful.” It is: expand the space of possible flourishing while preserving choice, safety, access, and moral concern.

Morphological freedom is especially important. It means that people should generally have wide freedom to alter their own bodies and minds. The concept pushes against paternalism. If a technology is safe, if a person gives informed consent, and if the intervention does not impose unacceptable risks on others, why should the state, a profession, a religion, a company, or a majority decide that the person’s body or mind must remain within inherited limits?

The argument has real force in disability and medical contexts. Imagine telling a person with paralysis that a BCI should be prohibited because it is unnatural. Imagine telling a person with treatment-resistant depression that brain stimulation is too intrusive even if it could help. Imagine telling a person facing a genetic disease that future therapies should stop at whatever current biology happens to permit. The mere fact that a technology changes the body or brain is not enough to condemn it.

Here is the transhumanist argument in a student-friendly form:

1. Many human limits produce serious suffering, lost agency, or preventable loss.
2. Naturalness by itself does not make a limit morally valuable.
3. If a technology can safely reduce suffering or expand agency without unacceptable costs to others, then there is a strong reason to permit or pursue it.
4. Therefore, responsible enhancement can be an expression of human freedom and care rather than a betrayal of humanity.

The conclusion is not that every enhancement is good. It is that the burden shifts. A critic has to say more than “this goes beyond nature.” They need to explain the specific harm: coercion, inequality, loss of agency, damage to relationships, exploitation, unsafe experimentation, bad governance, or a degraded picture of human worth.

The transhumanist lens also speaks to AI. AI assistants, research agents, medical models, language tools, and robotics can all be framed as capacity extenders. They can help people see, move, learn, communicate, remember, create, diagnose, or discover. Dario Amodei’s essay “Machines of Loving Grace” is not a formal transhumanist manifesto, but it gives a current AI-industry example of enhancement-adjacent optimism: powerful AI might accelerate biology, physical health, neuroscience, mental health, economic development, governance, work, and meaning. The essay explicitly acknowledges uncertainty and the danger of grandiosity, which is why it is useful here as rhetoric rather than proof.

Longevity companies provide another current example. Retro Biosciences states an ambition to add ten years to healthy human lifespan. That is an aspiration, not an established result. But the aspiration itself is philosophically revealing. Aging is being treated not only as a fact of life, but as a biological process that might be delayed, repaired, or redesigned. The transhumanist lens makes the appeal visible: if healthy life can be extended, why not try?

A responsible chapter must make the appeal vivid before moving to critique. It is easy to mock transhumanism when it is represented only by uploading minds, cryonics, or billionaire fantasies. It is harder to dismiss when the case begins with a person using a BCI to regain digital agency, a patient hoping for gene therapy, a student with a disability using assistive AI, or an older adult hoping for more healthy years with family.

The best objection to transhumanism is not that technology is unnatural. The best objection is that enhancement can quietly change the terms of human worth.

The Therapy / Enhancement Boundary Is Real But Unstable

One common way to govern enhancement is to draw a line between therapy and enhancement. Therapy treats disease, repairs injury, restores function, or relieves suffering. Enhancement goes beyond therapy by improving an already healthy capacity. That line often shapes medical obligation, insurance coverage, research approval, and public trust.

The line is real. A surgeon repairing a cleft palate, a doctor prescribing insulin, a wheelchair user receiving a better mobility device, and a person with paralysis using a BCI for cursor control are not the same as a healthy person seeking a competitive upgrade. Therapy and enhancement often carry different urgency. We usually think society has stronger obligations to provide treatment for illness or disability than to subsidize every desired performance boost.

But the line is unstable. The SEP entry on human enhancement shows why: the same intervention can function as treatment in one context and enhancement in another. It also notes that prevention complicates the distinction. Vaccines improve the body’s resistance before disease occurs. Exercise improves bodily capacity and prevents illness. Genetic screening may prevent disease but also invite trait selection. A technology can begin as therapy and later move into enhancement markets.

The Neuralink case makes the instability concrete. BCI-assisted control for someone with paralysis is restorative. But the same general pathway can be imagined for able-bodied users: faster gaming, work interfaces, memory aids, communication speed, attention monitoring, or direct human-AI interaction. The ethical question changes when the user population changes. So do the risks, incentives, and norms.

BCI for paralysis

Therapy / restoration: Restores digital control and agency. Enhancement: Possible if marketed to able-bodied users. Why the boundary is unstable: Same interface pathway, different baseline and social meaning.

ADHD medication

Therapy / restoration: Treats serious impairment. Enhancement: Used mainly for competitive performance. Why the boundary is unstable: Same drug, different diagnosis, pressure, and fairness context.

Prosthetic limb

Therapy / restoration: Restores mobility after limb loss. Enhancement: May exceed ordinary mobility in some settings. Why the boundary is unstable: Assistive design can become supernormal performance.

AI tutor

Therapy / restoration: Supports learning, especially when access is limited. Enhancement: Can become cognitive outsourcing or performance optimization. Why the boundary is unstable: It may build capacity or replace practice.

Gene editing

Therapy / restoration: Somatic treatment can address disease. Enhancement: Germline or trait selection can redesign future capacities. Why the boundary is unstable: Future persons and heredity change consent and governance.

Mood intervention

Therapy / restoration: Treats depression or severe anxiety. Enhancement: Can optimize personality, affect, or emotional conformity. Why the boundary is unstable: Relief can become pressure to be cheerful, calm, or productive.

The boundary is also socially loaded. What counts as normal functioning depends partly on institutions. A student who cannot hear a lecture is disabled by hearing loss, but also by a classroom without captions. A person using a wheelchair is limited by their body, but also by stairs, narrow doors, inaccessible transportation, and social assumptions. If the environment changes, the capacity changes. This means therapy and enhancement cannot be judged only by looking inside the body.

A good design question is: is the technology adapting the world to the person, the person to the world, or both?

A screen reader changes the interface so a blind user can access digital text. A cognitive-enhancement drug changes the user’s body to fit demanding work or school conditions. An AI assistant may do both: it can make information more accessible, but it can also normalize a faster pace that pressures everyone to use assistance just to keep up. The same intervention can be accommodation, support, enhancement, and pressure depending on the surrounding institution.

This is why “choice” must be examined carefully. A student may choose to use AI study tools. A worker may choose to wear a productivity tracker. A parent may choose genetic screening. But if the school, employer, insurance system, or competitive market makes the choice practically necessary, the choice is not as free as it appears. An optional enhancement can become a social requirement.

Consider an AI writing assistant. Used well, it can help a multilingual student understand assignment wording, generate practice questions, or revise clarity after the student has formed an argument. That is capacity support. Used badly, it can replace the student’s struggle with reading, drafting, revising, and explaining. That is capacity replacement. If every student in a program begins using the tool and assignments quietly assume AI-speed production, the tool becomes an optimization pressure. The ethical status changes not because the software changed, but because the social meaning changed.

The therapy/enhancement distinction therefore remains useful but insufficient. It gives us a first question: is this intervention restoring a capacity or pushing beyond restoration? Then we need additional questions: who defines the baseline, who gets access, who is pressured, what capacities weaken, and what vision of normal human functioning is being installed?

Meaningful Limits: Dignity, Giftedness, And The Goods Of The Given

Transhumanism is right about one thing: naturalness is not a moral trump card. A limit is not good merely because we inherited it. But critics of enhancement are also right about something: not every limit is simply a defect. Some limits help structure care, humility, solidarity, achievement, dependence, patience, grief, embodiment, and the meaning of effort.

This is the strongest version of the dignity or bioconservative critique. It does not say, “Never use medicine.” It does not say, “Suffering is good.” It asks what human goods might be lost if life, reproduction, mood, memory, achievement, aging, and children all become engineering projects.

Leon Kass’s “Ageless Bodies, Happy Souls” is a classic statement of this worry. Kass is concerned with biotechnology used “beyond therapy”: not just healing disease, but altering mood, memory, behavior, reproduction, aging, and human aspiration. His deeper worry is that safety and fairness do not exhaust the moral question. Even if an intervention were safe and widely available, it might still change the meaning of human activities in ways we should resist.

Michael Sandel’s The Case Against Perfection gives a related but distinct critique. Sandel says the problem with genetic enhancement goes beyond safety and fairness. The drive to master and design human nature can fail to appreciate the “gifted” character of human powers and achievements. His word giftedness does not require a religious premise in the classroom. It names the thought that some parts of life come to us unchosen and that learning to receive, care for, and live with the unchosen is part of human flourishing.

Take parenting. Parents should care for children, protect them, educate them, and help them grow. They should also receive children as persons rather than products. If reproductive technology becomes a way to select children toward parental preferences for height, temperament, intelligence, appearance, athletic promise, or emotional style, the child may be subtly treated as a project to be optimized. The worry is not only unfair advantage. It is the attitude of mastery toward another person.

Take achievement. A person who studies for years to become a pianist, nurse, welder, mathematician, teacher, or athlete develops more than output. They develop patience, attention, discipline, judgment, and identity. If a technology could instantly install the performance, would the achievement mean the same thing? The answer may vary by case. A cochlear implant, a prosthetic, or a medication can make achievement possible. But an enhancement that bypasses the formation of skill might change what the achievement expresses.

Take mood. Relief from depression or severe anxiety is a genuine good. But a culture that treats grief, anger, sadness, boredom, restlessness, or ordinary vulnerability as malfunctions may flatten emotional life. Some painful emotions are part of love, loss, injustice, moral perception, and growth. A technology that helps someone escape debilitating suffering is one thing. A system that pressures people to maintain an optimized emotional profile for productivity is another.

Take aging. Many features of aging are cruel: pain, frailty, dementia, isolation, dependence, and early death. It is morally serious to reduce those harms. But mortality also shapes urgency, inheritance, generational turnover, care, memory, and the way people understand a life course. A serious longevity ethics should ask not only whether we can extend healthy life, but how extended life would affect relationships, justice, work, retirement, population, ecology, grief, and intergenerational responsibility.

A meaningful limit is a limit that may protect or make possible a human good. That does not mean it should never be changed. It means that changing it requires moral attention.

The table should not romanticize suffering. A person in chronic pain does not need a lecture about meaningful limits before receiving care. A person with paralysis does not need to be told that dependency builds character. That would be cruel. The dignity critique becomes strongest when it resists social fantasies of perfection, not when it blocks relief.

This gives us a better standard. Ask whether the limit is harmful, formative, relational, unjustly imposed, or socially constructed. A harmful limit may call for treatment. A formative limit may call for respect. An unjustly imposed limit may call for political repair. A socially constructed limit may call for redesigning the environment rather than redesigning the person.

The dignity critique is weak when it treats familiar limits as sacred without asking who suffers under them. It is strong when it notices that enhancement can change our attitude toward bodies, children, achievement, emotion, age, and dependence. It asks whether we are expanding agency or turning human life into a permanent performance-improvement project.

In a case inquiry, examine more than the promised output. Ask what kind of person, practice, and relationship the tool helps form. An AI tutor should preserve the student’s practice of attention, confusion, effort, and explanation. A mental-health tool should respect grief, anger, and context. A BCI should preserve agency, consent, exit, and the user’s sense of self.

Normality, Disability, And Access

A chapter on enhancement would be ethically incomplete without disability justice. Enhancement debates often assume a simple picture: disability is a defect, normal functioning is the baseline, enhancement rises above that baseline, and technology helps people move upward. That picture is too simple.

Disability justice does not mean refusing treatment, assistive technology, or restoration. Many disabled people want pain relief, mobility devices, communication tools, accessible environments, medical treatment, and technologies that increase independence. The mistake is assuming that every disability is only a personal biological failure and that the goal is always to become as close as possible to a narrow idea of normal.

The SEP entry on human enhancement explains one major criticism of normal-function accounts: they can be insensitive to diverse ways human beings flourish, and disability theorists have worried about medicine’s pull toward normalization. Rosemarie Garland-Thomson’s argument for “conserving disability,” discussed in the AMA Journal of Ethics, challenges readers to see disability not only as deficit but also as part of human variation, interdependence, knowledge, community, and moral imagination. David Wasserman and Adrienne Asch’s discussion of prenatal disability screening is useful because it shows how reproductive technologies can carry social judgments about which lives are worth welcoming.

This does not mean disability is always good or that suffering should be preserved. It means the meaning of disability is not settled by a medical diagnosis alone. A person can be limited by a body, by a building, by a classroom design, by an employer’s schedule, by stigma, by poverty, by software assumptions, or by a society that treats only one kind of body or mind as normal.

Return to the BCI case. A Neuralink-style implant may restore digital agency for a person with paralysis. That is a serious good. But a responsible design question asks more: will disabled users shape the research agenda? Will access be limited to wealthy patients, trial participants, or people near elite medical centers? What happens if the company fails, changes terms, discontinues support, or controls the data stream? Does the device expand independence or create a new dependency on a proprietary system? Does the public celebrate the user only when they become technologically impressive?

The disability/access lens asks four questions:

The access problem is not secondary. Bostrom’s own “Transhumanist Values” names wide access as a basic condition. The Transhumanist Declaration calls for responsible and inclusive moral vision. But declaring wide access is easier than building it. Expensive technologies often reach wealthy users first. Medical systems already distribute access unequally. Schools and workplaces often adopt tools before they have accommodations, appeals, or support structures. Enhancement can become one more way advantage compounds.

Consider cognitive enhancement in education. Students with money can pay for better devices, private tutoring, AI subscriptions, quiet workspaces, diagnoses, medications, and health care. Students without those resources may be told to compete in the same environment. If AI tutors, memory systems, or attention tools become expected, the gap may widen. The ethical issue is not only whether the tool improves learning. It is whether the institution builds access, support, and fair expectations around it.

Consider workplace optimization. A wearable that tracks fatigue could help protect workers from injury. It could also become a surveillance system that punishes people whose bodies do not conform to a productivity model. A neurotechnology that detects attention could help someone manage concentration. It could also become a way for employers to monitor cognitive states. The same capacity data can serve care or control.

Consider reproductive technology. A parent may want to prevent a painful disease. That desire can be loving. But if screening and editing practices communicate that some lives should not exist, disability communities have reason to object. A technology can reduce suffering while also reinforcing harmful social messages about worth.

The point is not to make design impossible. The point is to prevent a false individualism. Enhancement is never only a private upgrade. It changes expectations, markets, institutions, and norms. A technology that increases one person’s capacity can also change what everyone else is expected to do.

Key Point: Enhancement Can Become Pressure

A technology offered as optional can become practically mandatory when schools, employers, insurers, parents, platforms, or markets begin treating enhanced performance as the new normal.

For PHIL 123 case inquiries, this lens is direct. If you study AI in nursing, ask whether the tool supports nurses’ judgment or pressures them to work at machine pace. If you study AI in education, ask whether the tool helps students learn or redefines adequate student performance around AI assistance. If you study workplace automation, ask whether the technology expands capacity or treats human fatigue, slowness, care, and disability as productivity defects. If you study health AI, ask who gets the benefit first and who becomes data for others’ benefit.

Brains, Agency, And Cognitive Liberty

Brain and mental-life technologies raise special concerns because they touch the conditions of agency. A person uses attention, memory, emotion, perception, judgment, and self-control to make choices. A technology that reads, modulates, predicts, or shapes those capacities is not just another device. It can affect the ground from which consent, responsibility, and identity operate.

This is why the Neuralink case returns in a deeper form. For a person with paralysis, a BCI can restore a route from intention to action. The user intends to move a cursor; the system detects neural activity, decodes it, and converts it into device control. That can expand agency. But the same interface also raises questions about neural data, device security, informed consent, dependency, psychological continuity, commercial control, and exit. Who owns the data? Who can update the software? What happens if the company stops support? Can the user remove the device safely? What risks remain after the trial? What does the user understand about uncertainty?

Marcello Ienca and Roberto Andorno’s article “Towards new human rights in the age of neuroscience and neurotechnology” gives one influential framework. They discuss cognitive liberty, often described as mental self-determination: the freedom to use neurotechnologies and the protection against coercive or unconsented use. They also discuss mental privacy, mental integrity, and psychological continuity as rights or protections that may become necessary as neurotechnology develops.

UNESCO’s ethics of neurotechnology work gives the policy version of the concern. UNESCO notes that neurotechnology has transformed medicine and is expanding beyond health care, while also raising concerns about access to and manipulation of brain activity, identity, emotions, thoughts, dignity, autonomy, and mental privacy. In November 2025, UNESCO announced that Member States had adopted the first global standard on the ethics of neurotechnology, with safeguards around mental privacy, non-therapeutic uses for children and young people, workplace monitoring, explicit consent, transparency, inclusion, and affordability.

Nita Farahany’s public-facing discussion of brain data and cognitive liberty is useful for students because it connects neurotechnology to law, workplaces, and everyday forms of data collection. Brain data is not ordinary data. It may reveal attention, fatigue, emotional response, preference, stress, or intention. Even when the data is noisy or limited, institutions may treat it as authoritative.

These concerns apply differently across technologies. An implanted BCI for cursor control is not the same as a consumer headset that estimates attention. Deep brain stimulation for Parkinson’s disease is not the same as a workplace device that monitors alertness. A mental-health app that asks users to journal is not the same as a system that infers emotional state from neural signals. But the underlying ethical issue is related: mental life is a condition of agency, not merely a source of useful data.

The distinction between active and passive use also matters. An active BCI lets a user intentionally control an output. A passive monitoring system observes ongoing brain states such as fatigue, attention, stress, or emotional response. Active control can support agency. Passive monitoring can support care, but it can also turn mental states into institutional evidence. A company, school, military, or insurer might use attention, stress, or fatigue signals to manage people. That is why consent and context matter so much.

The strongest pro-neurotechnology argument is still serious. People with paralysis, ALS, treatment-resistant depression, Parkinson’s disease, epilepsy, locked-in syndrome, severe communication impairments, or chronic pain may have strong reasons to use brain-related technologies. Respecting cognitive liberty includes respecting the freedom to use neurotools. A blanket ban can be paternalistic.

The strongest caution is equally serious. Because brain and mental-life technologies reach the conditions of agency, they require stronger safeguards than ordinary consumer devices. They require clear consent, limited data use, independent oversight, security, post-trial obligations, access planning, and special caution around children, workers, prisoners, soldiers, patients, and people dependent on a company or institution for support.

This also changes how we think about AI assistants as cognitive tools. Most AI tutors and writing assistants do not read brain data. But they can still shape attention, memory, confidence, dependence, and self-understanding. A system that constantly predicts what you want to say may change how you write. A system that summarizes every hard reading may change how you practice comprehension. A system that remembers your preferences and nudges your decisions may shape the mental environment in which choice happens. That is not the same as neural intervention, but it belongs on the same capacity map.

For design, the question is not “Is this mind control?” That is usually too dramatic. Ask the better questions: what mental capacity is being supported, what data is being collected, what dependency is being created, who can intervene, who can refuse, and whether the user remains the author of the action.

Inherited Bodies: Gene Editing, Future Persons, And Governance

The Neuralink case asks what a person may choose for their own body and mind under conditions of risk and hope. Germline genome editing asks a different question: what may one generation choose for future persons?

The most famous modern case is He Jiankui’s 2018 announcement that twin girls had been born after he used CRISPR to edit embryos. Henry Greely’s open-access article “CRISPR’d babies: human germline genome editing in the ‘He Jiankui affair’” gives a careful analysis. Greely explains that germline editing is ethically distinct because changes may be inherited by descendants. Somatic editing changes body cells in a way that is not passed on through ordinary reproduction. Germline editing changes cells involved in reproduction or early development, so the effects may enter a family line.

The He case is not useful as a simple “mad scientist” story. It is useful because it makes several ethical pressures visible at once. He presented the project as disease prevention: an attempt to alter CCR5 in order to reduce HIV vulnerability. But the case raised major concerns about safety, medical necessity, informed consent, secrecy, uncertain benefit, off-target effects, mosaicism, future children, and governance. Greely condemns the experiment, but he also slows the analysis enough to show why the case is not just about personality. It is about the structure of heritable intervention.

The prevention claim matters because it shows how enhancement debates often enter through sympathetic doors. Preventing disease is appealing. Parents want children to be healthy. Researchers want to reduce suffering. But germline editing adds questions that ordinary adult self-modification does not. The future person cannot consent. The intervention may affect descendants. Mistakes may be difficult or impossible to undo. Social pressure may shape which traits are selected against. A medical decision can become a social statement about which lives should exist.

The WHO’s 2021 recommendations on human genome editing emphasize strong oversight for somatic, germline, and heritable genome editing. The WHO document points to international collaboration, registries, action against illegal or unsafe research, intellectual property, education, engagement, and ethical values. The National Academies’ 2017 report announcement said heritable genome-editing trials might be allowed in the future for serious conditions under stringent oversight, but it also said genome editing for enhancement should not be allowed at that time and called for broad public input before non-treatment uses.

Those governance sources are important because they resist both extremes. They do not say all genome editing is forbidden. Somatic editing for disease can be ethically serious and potentially beneficial. They also do not say technical possibility is enough. Heritable editing requires a different level of oversight because the affected parties include future people, families, populations, and social norms.

This section also connects to pro-enhancement reproductive ethics. Julian Savulescu’s argument for procreative beneficence holds, roughly, that parents have strong moral reason to select the child expected to have the best life among possible children, when selection is available and other things are equal. The argument is deliberately provocative. It presses the thought that if parents can prevent suffering or choose traits that improve a child’s prospects, refusing to do so may require justification.

The disability and dignity objections push back. Who defines “best”? Does selecting against a disability prevent suffering, express prejudice, or both? Does a child become a product of parental design? What happens when market preferences, race, class, gender, ability, and beauty standards shape the menu? What if only wealthy parents can select? What if insurance or social expectations make selection feel required? What if the trait is connected to identity, culture, or a way of flourishing that outsiders misunderstand?

The gene-editing case therefore teaches the central lesson of this chapter at a higher level of stakes. Enhancement is not only individual freedom. Some capacity decisions are inherited, institutional, and political. The more a technology affects future persons or changes social baselines, the less adequate a simple consumer-choice model becomes.

Key Point: Future Persons Change The Ethics

Adult self-modification raises questions of consent, safety, access, and identity. Heritable modification adds future persons, descendants, public governance, and social meaning. The ethical standard must be stronger.

For students, the practical transfer is clear. In a case inquiry, ask whether the technology affects only the current user or also future users who inherit the system. An AI advising tool may not change genes, but it can change future educational pathways. A health platform may not redesign bodies, but it can change what counts as responsible self-management. A workplace optimization system may not affect descendants biologically, but it can redefine what future workers are expected to be. The He Jiankui case is biologically specific, but the governance lesson travels: the farther the effects extend beyond the consenting user, the stronger the need for public oversight.

AI As Soft Enhancement

Transhumanism is often associated with implants, gene editing, life extension, cryonics, and radical body modification. But many AI tools function as soft enhancement: they extend or reshape human capacities without directly changing the body. They change what a person can do through scaffolding, delegation, prediction, memory, recommendation, translation, simulation, or automation.

An AI writing assistant extends fluency. A translation tool extends communication. A coding agent extends software-building capacity. An image generator extends visual production. A medical model extends diagnostic attention. A route-planning system extends navigation. A reminder system extends memory. An AI tutor extends practice, explanation, and feedback. A research assistant extends search and synthesis. A wearable paired with AI extends self-monitoring. These are not posthuman fantasies. They are already part of ordinary life.

Soft enhancement matters because it can be easy to miss. A student may think transhumanism is irrelevant unless someone is implanting chips or editing embryos. But the same question appears in less dramatic form: what human capacity is being extended, and what happens to the person or institution as a result?

Take AI tutoring. The attractive version is capacity-building. A student who works full time, supports family, and cannot meet a tutor during office hours can ask for explanations, examples, practice questions, vocabulary support, and feedback at midnight. The tool can support agency. It can make education more accessible. It can reduce shame around asking basic questions. The transhumanist lens notices the good: technology can expand the student’s learning capacity beyond what the institution could otherwise provide.

Now the caution. If the tutor gives answers too quickly, the student may stop practicing struggle. If it summarizes every hard text, the student may not build reading endurance. If it sounds confident when wrong, the student may build false understanding. If the college assumes all students now have AI support, assignments may become more demanding without more human care. If only some students can afford better tools, inequality grows. If the tool logs everything, student confusion becomes data.

The same tool can build or weaken capacity. The difference is design, context, and use.

Take AI in health and biology. Amodei’s “Machines of Loving Grace” imagines powerful AI accelerating biology and medicine. The aspiration is transhumanist in spirit even when it avoids the label: use intelligence to reduce disease, extend healthy life, understand the brain, and improve human flourishing. The caution is that medicine has hard constraints: clinical trials, human bodies, messy biology, access, cost, regulation, and trust. AI-generated optimism should be read as a forecast and aspiration, not as proof that the promised capacity has arrived.

Take workplace AI. A system that helps nurses document faster may preserve attention for patients. It may also increase workload because administrators expect every nurse to handle more patients. A coding agent may help a developer build quickly. It may also create technical debt or weaken understanding. A productivity assistant may support focus. It may also make slower human rhythms look irresponsible. Soft enhancement can quietly become optimization.

This is where the transhumanism lens connects with the automation chapter. Automation asks what moved into the system. Transhumanism asks what human capacity the system now imagines. A tool that writes for you is not only automating text. It is changing the capacity of writing: invention, expression, revision, voice, argument, and accountability. A tool that recommends emotional responses is not only automating communication. It is changing the capacity of relational judgment. A tool that monitors sleep, mood, and focus is not only collecting data. It is changing the capacity of self-interpretation.

The human-capacity lens becomes practical here. Do not ask only, “Can we make users faster?” Ask, “Which human capacity is being strengthened, and which is being bypassed?” Speed is not always capacity. Sometimes it is the enemy of capacity. A student who finishes faster may learn less. A doctor who documents faster may notice less. A worker who receives constant optimization nudges may become less able to set their own rhythm. A person who relies on generated emotional language may become less practiced in speaking honestly.

Soft enhancement also makes the access problem immediate. If AI tools become cognitive prosthetics for school and work, then unequal access becomes unequal capacity. A student with the paid version, better hardware, stable internet, and more AI literacy may gain a major advantage. A student without those things may fall behind in a system that pretends everyone has the same tools. A responsible institution cannot simply say, “Use AI if you want.” It must decide what tools are allowed, taught, supported, disclosed, and made available.

Key Point: AI Can Enhance Without Implanting Anything

AI tools can extend memory, attention, writing, communication, research, diagnosis, creativity, and self-monitoring. The ethical question is whether they build human capacity, replace it, privatize it, or turn it into pressure.

This section should prevent a narrow misunderstanding. Transhumanism is not only about becoming a cyborg. It is about what human beings should be able to do and become with technological help. AI already belongs inside that question.

Four Lenses For Responsible Enhancement

At this point the debate can be mapped without turning it into a two-team fight. The old version of this topic is often presented as transhumanism versus bioconservatism: expand capacity or preserve humanity. That contrast is useful, but it is too narrow. A rigorous chapter needs at least four lenses.

Transhumanist / bioliberal autonomy

Central question: If safe enhancement expands agency or reduces suffering, what justifies forbidding it? What it sees well: Natural limits can be harmful; personal choice matters; technology can liberate. What it can miss: Choice can become pressure; access can be unequal; social meaning matters.

Dignity / giftedness / meaningful limits

Central question: What goods might be lost when life, achievement, reproduction, mood, or happiness become engineering projects? What it sees well: Some limits structure humility, care, solidarity, and meaning. What it can miss: Appeals to nature can protect avoidable suffering or become paternalistic.

Justice / disability / normality

Central question: Who defines improvement, who gets access, who absorbs risk, and who is treated as defective? What it sees well: Enhancement is social, not only individual; normality is contested. What it can miss: It may not always give a clear line between support and enhancement.

Cognitive liberty / neurorights

Central question: What safeguards are needed when technologies reach attention, memory, emotion, brain data, or thought? What it sees well: Mental life is a condition of agency; consent and privacy need stronger protection. What it can miss: It governs mental intervention but does not by itself define the good life.

A strong student analysis should be able to use all four. Take the Neuralink case. The transhumanist lens sees restored agency and future possibilities. The dignity lens asks whether a person’s agency should depend on a risky commercial implant and whether unenhanced embodiment is being treated as inferior. The justice lens asks who gets access, who shapes the research, and whether disabled users are centered or used as a path to able-bodied markets. The cognitive-liberty lens asks about neural data, consent, mental privacy, device control, and exit.

Take an AI tutor. The transhumanist lens sees expanded learning capacity. The dignity lens asks whether difficulty, practice, and intellectual struggle are being bypassed. The justice lens asks who gets access and whether AI use becomes expected. The cognitive-liberty lens asks less about brain data and more about attention, dependence, self-authorship, and whether the student remains the agent of learning.

Take gene editing. The transhumanist lens sees disease prevention and future health. The dignity lens asks whether children are being treated as products. The justice lens asks whose traits are valued or selected against. The cognitive-liberty lens may be less central unless the intervention affects mental traits, but questions of future autonomy remain.

The lenses can conflict. You cannot always maximize morphological freedom and preserve meaningful limits. You cannot always treat enhancement as a private market choice and also prevent social pressure. You cannot always accelerate health innovation and also guarantee equal access. Ethical judgment often requires choosing which concern governs the case.

That is not a failure of philosophy. It is what philosophy is for. A framework that never creates tension is usually hiding something.

For a PHIL 123 case inquiry, the goal is not to declare allegiance to one camp. The goal is to use the lenses to make a judgment more answerable. If you defend a technology, say what dignity, access, and agency risks must be addressed. If you oppose a technology, say whether your objection blocks restoration or only irresponsible enhancement. If your view depends on safeguards, connect them to the specific capacity at stake.

Older Dreams, New Design Questions

The current chapter should not be organized as a long genealogy of posthuman dreams, but a little background helps. Human beings have long imagined becoming more than they are: wiser, purer, stronger, longer-lived, closer to gods, free from disease, free from death, or liberated from bodily limits. Religious traditions, philosophical perfectionism, alchemy, Enlightenment progress, science fiction, cybernetics, and biotechnology all carry versions of that longing.

The historical pattern matters only if we do not overstate it. Ancient spiritual ascent, Christian theosis, Enlightenment perfectibility, Nietzschean self-overcoming, Russian Cosmism, and Silicon Valley transhumanism are not one straight line. They are different traditions with different assumptions about body, soul, technology, salvation, politics, and human nature. Similar imagery is not the same as direct genealogy.

The more useful point for this chapter is this: transhumanism gives a technological form to an old human question. What should we do with our limits? Accept them? Interpret them? Heal them? Resist them? Design around them? Overcome them? Use them to structure a good life? Redesign ourselves so the old limits no longer define us?

That question now appears in concrete design settings. It appears when a company names aging as an engineering target. It appears when a parent considers reproductive screening. It appears when a college adopts AI tutoring. It appears when a worker is tracked for fatigue. It appears when a person with paralysis chooses an implant. It appears when a mental-health app treats emotional life as a dashboard. It appears when AI labs describe a future of accelerated biology and medicine.

So the chapter does not need a long march from Plato to uploading. It needs a disciplined question: what kind of human capacity is being imagined, and can that vision survive ethical scrutiny?

Using The Transhumanist Lens: The Human-Capacity Audit

A human-capacity audit applies the chapter’s competing positions to a documented case. It turns broad claims about enhancement and human limits into questions that can pressure a provisional considered view.

Start with the technology or practice in your anchor case. Do not begin with whether it is futuristic. Begin with capacity. What does the technology help people do, stop doing, do faster, do differently, or no longer need to practice?

Here is a completed example using the opening case.

Here is a second example closer to student work.

The audit should help you avoid common mistakes.

The most responsible position is neither simple transhumanist enthusiasm nor simple resistance to change. It is a disciplined ethics of capacity. Some limits should be overcome. Some should be accommodated rather than “fixed.” Some should be preserved because they protect human goods. Some should be governed because individual choice alone cannot manage their social effects.

The chapter’s thesis can now be stated plainly:

Transhumanism reveals how technology can restore, extend, optimize, or redesign human capacity. It is ethically powerful because suffering and limitation are not automatically sacred. It is ethically dangerous when it treats ordinary human limits, dependencies, bodies, moods, or differences as defects to engineer away. Responsible judgment tests capacity expansion against dignity, meaningful limits, access, disability justice, cognitive liberty, and public governance.

References

• Amodei, Dario. “Machines of Loving Grace: How AI Could Transform the World for the Better.” Used as a current AI-industry example of enhancement-adjacent optimism about biology, health, neuroscience, work, and meaning; treated as forecast and rhetoric, not proof of outcomes.
• Bostrom, Nick. “Transhumanist Values.” Used for the transhumanist case, posthuman possibilities, morphological freedom, wide access, individual choice, and concern for sentience.
• Bostrom, Nick, and Julian Savulescu, eds. Human Enhancement. Oxford University Press. Also see the introduction PDF. Used as a background canon source for the enhancement debate.
• ClinicalTrials.gov. “Precise Robotically Implanted Brain-Computer Interface for the Control of External Devices,” NCT06429735. Used as the official study-record link for the Neuralink PRIME / device-control clinical-trial pathway; live source to recheck.
• Farahany, Nita. “The Battle for Your Brain: A Legal Scholar’s Argument for Protecting Brain Data and Cognitive Liberty.” Used for a public-facing explanation of cognitive liberty and brain-data governance.
• Garland-Thomson, Rosemarie, discussed in Sara Goering, “A Defense of ‘The Case for Conserving Disability’.” AMA Journal of Ethics. Used for the disability-justice critique of treating disability only as deficit.
• Greely, Henry T. “CRISPR’d babies: human germline genome editing in the ‘He Jiankui affair’.” Journal of Law and the Biosciences. Used for the He Jiankui case, germline/somatic distinction, CCR5 context, consent, safety, and governance concerns.
• Humanity+. “The Transhumanist Declaration.” Used as a movement self-description of broadening human potential, risk awareness, responsible deliberation, autonomy, future generations, sentience, and personal choice.
• Ienca, Marcello, and Roberto Andorno. “Towards new human rights in the age of neuroscience and neurotechnology.” Life Sciences, Society and Policy. Used for cognitive liberty, mental privacy, mental integrity, and psychological continuity.
• Juengst, Eric, and Daniel Moseley. “Human Enhancement.” Stanford Encyclopedia of Philosophy. Used as the main reference-layer source for enhancement definitions, therapy/enhancement boundaries, health-care limits, fairness, disability, eugenics, moral agency, and regulation.
• Kass, Leon. “Ageless Bodies, Happy Souls: Biotechnology and the Pursuit of Perfection.” The New Atlantis. Used for the bioconservative dignity and meaningful-limits critique of biotechnology beyond therapy.
• National Academies of Sciences, Engineering, and Medicine. “With Stringent Oversight, Heritable Human Genome Editing Could Be Allowed for Serious Conditions.” Used for governance guidance on heritable genome editing, serious conditions, stringent oversight, and non-allowance of enhancement at the time of the 2017 report.
• Neuralink. “Device Control / PRIME Study.” Used as a company source for the opening BCI case and device-control framing; treated cautiously and recheckable.
• Reuters. “Neuralink shows first brain-chip patient playing online chess.” Used for the Noland Arbaugh demonstration and expert caution.
• Retro Biosciences. Company website. Used as a current longevity-company example; treated as an aspiration to extend healthy lifespan, not as proof of achieved results.
• Sandel, Michael J. The Case Against Perfection: Ethics in the Age of Genetic Engineering. Used for giftedness, mastery, humility, solidarity, and critique of genetic enhancement beyond safety and fairness.
• Savulescu, Julian. “Procreative Beneficence: Why We Should Select the Best Children.” Bioethics. Used as a pro-enhancement reproductive ethics source, treated as a provocative argument rather than a settled conclusion.
• UNESCO. “Ethics of neurotechnology.” Used for the policy frame around neurotechnology, dignity, autonomy, mental privacy, social inequality, and AI-neurotechnology convergence.
• UNESCO. “Towards an International Instrument: Recommendation on the Ethics of Neurotechnology.” Used for the November 2025 global-standard update and safeguards around mental privacy, workplace monitoring, consent, transparency, inclusion, and affordability.
• Wasserman, David, and Adrienne Asch. “The Uncertain Rationale for Prenatal Disability Screening.” AMA Journal of Ethics. Used for disability-rights concerns around prenatal testing and social messages about disability.
• WHO. “Human genome editing: recommendations.” Used for the 2021 governance recommendations on human genome editing, including strong oversight for somatic, germline, and heritable editing.
• Wired. “Brain-Computer Tech Progresses.” Used as accessible reporting on BrainGate’s earlier implanted BCI results; included to avoid implying Neuralink was the first implanted BCI.
• The Guardian. “Neuralink’s first implant partly detached from patient’s brain.” Used for reporting on thread retraction and software recovery in the Arbaugh case.