Summary. This essay traces the structure of public knowledge around a small endogenous tripeptide, GHK-Cu, whose preclinical biochemistry has been reproducibly demonstrated at physiological concentrations across multiple independent groups and whose clinical characterization in humans has, after fifty years, never been performed. It argues that the absence is not a knowledge gap waiting to be filled but the stable output of interlocking incentive structures: an unpatentable molecule that no commercial actor can profitably evaluate, a permissive compounding regime that removes any regulatory pressure to do so, a community whose identity and economy depend on the molecule remaining ambiguously unresolved, and a regulatory apparatus whose justifications for action against it have been thinly grounded in its actual pharmacology. The structure generalizes across domains — venture capital, productivity systems, dietary frameworks, contemplative practice — identifying a third failure mode of evidence-based reasoning, neither false-positive nor false-negative, but the structural prevention of resolution. The essay closes by naming what is most at risk: not the material world but the phenomenological capacity to orient toward it. And it locates one remaining site of resistance, embodied and untransmissible, in the Eros of a body that knows what it is.

My friend tells me, over coffee in the Mission, that her skin has never looked better. She is in her forties. She is glowing, in a way that has nothing to do with the lighting in the café. I ask her what she’s doing. She lists, without hesitation: a tretinoin compound at 0.05%, a niacinamide serum at night, a vitamin C in the morning, retinaldehyde on weekends, hyaluronic acid injections twice a year, low-dose semaglutide, methylene blue under the tongue, a stack of supplements I scribble onto the back of a receipt and then lose on the walk home. And the GHK-Cu. She injects it. Subcutaneously. Daily. From a five-milligram vial she gets from a compounding pharmacy outside Las Vegas that ships in dry ice on Tuesdays.

The wanting comes in waves while she talks. Not exactly for the protocol — for the glow, the steadiness in her face, the way she carries herself as though something is being attended to. And underneath the wanting, almost before I notice it happening, a second movement: my mind has slipped past the woman in front of me and gone to PubMed. By the time we hug goodbye and I’m walking up the slope of 24th in the cold, the friend has become a research question, and I have a list of search terms I’m rehearsing in my head as my breath fogs in front of me.

It is only on the walk home that I let myself see what happened at the table. I went looking for a body — hers, mine — and I found a literature.

The move is small. It is not, on its surface, dramatic. It happens millions of times a day, in cafés and exam rooms and Reddit threads and bathroom mirrors. A wanting arises that is bodily, that concerns one’s own material life — the skin, the energy, the wound that won’t heal, the hair, the sleep — and before the wanting has had time to register fully as wanting, the mind has reached past it into the layer of frameworks and citations and protocols above. The body becomes the referent of a discourse. The actual thing under the skin in the mirror is no longer where the attention lives. The attention is in the literature now. Or the community. Or the protocol. Or the conversation. Somewhere just above the body, on a landing, with the lights on.

There is a particular molecule at the bottom of this staircase. It will not give us what we want from it. The point is the descent.

Molecule

Glycine. Histidine. Lysine. Three amino acids in a row, the smallest peptide that could plausibly be called a peptide, looped through a single copper ion held square-planar between the histidine imidazole and the alpha-amino terminus. Three hundred and forty daltons of molecule — small enough that you could draw it on the back of a napkin without lifting the pen, roughly the same size as caffeine. It is in your blood right now, at about two hundred nanograms per milliliter at twenty, declining to around eighty by sixty,1 a decline that has, in some hands, been taken as a clue, and in others, as a target.

It is called GHK-Cu, and the story of it is the story of a man named Loren Pickart, who isolated it from human plasma in 1973 as a graduate student at UCSF.2 He was looking at why old hepatocytes in culture behaved old and young hepatocytes behaved young, and he found a small fraction of the plasma — eventually narrowed to this three-amino-acid peptide bound to copper — that, when added to old hepatocytes, made them behave like young ones again. He spent the next fifty years on it. He started a small company, Skin Biology, out of a low-slung office park in Bellevue, Washington, and from that office he conducted a long, lonely campaign for the importance of his molecule. He wrote review after review, in journals like BioMed Research International, IJMS, and Oxidative Medicine and Cellular Longevity,3 with titles that escalated steadily in their claims. The molecule heals wounds. The molecule reverses oxidative stress. The molecule treats COPD. The molecule has anti-cancer activity. The molecule resets four thousand genes to a younger state. He died in 2023, at eighty-five, having watched the world’s interest in his molecule grow without ever quite watching it cohere into a clinical program. His widow now runs the company.

This is the man, and it matters to picture him, because the literature on GHK-Cu cannot be read without him. Of the roughly eighty papers in PubMed, a large fraction of the conceptual and review work is his. The reviews are not fraudulent. They are also not the writings of a disinterested observer. The discoverer was the commercial stakeholder. The way the molecule has been framed for the world over five decades is, substantially, the way one man framed it. When I started reading, I was prepared to find that this central fact had hollowed the field out — that there was nothing else.

What I found was something more interesting, and more difficult to dismiss.

In 1988, a French academic group in Reims led by François Maquart at the CNRS published a paper in FEBS Letters4 showing that GHK-Cu stimulated collagen synthesis in fibroblast cultures at picomolar to nanomolar concentrations. Picomolar. A thousand times below the concentration at which most signaling molecules act. To produce an effect at that concentration the molecule has to be doing something coordinated and specific — it cannot be doing it by sheer chemistry. Then, five years later, the same group published in the Journal of Clinical Investigation,5 which is not a journal one sneaks past peer review. They took rat wound chambers and showed concentration-dependent increases in dry weight, in DNA, in total protein, in collagen content, in glycosaminoglycan content. They ran a control with a different tripeptide and showed it had no effect. The collagen stimulation was about twice the increase in non-collagen proteins, meaning the molecule was specifically biasing toward extracellular matrix deposition. Type I and type III collagen messages were elevated. The result was clean. The group had no commercial relationship to Pickart. They were French biochemists doing biochemistry.

This paper deserves a pause, because of where the story goes later. A small endogenous human tripeptide, present in everyone’s blood, stimulates the extracellular matrix synthesis machinery of fibroblasts at concentrations that are essentially physiological, in a properly controlled experiment, published in a high-impact journal by a competent academic group with nothing to gain. The molecule is real. It does something. The basic biochemistry has never been in dispute.

What has always been in dispute — what the discourse around the molecule rarely admits clearly — is the distance between this fact and a clinical recommendation.

The year before Maquart’s JCI paper, in 1992, a vascular surgery group ran the first and last serious randomized controlled trial of GHK-Cu in humans. It appeared in the Journal of Vascular Surgery:6 GHK-Cu cream at 0.4%, versus silver sulfadiazine, versus placebo, in eighty-six patients with chronic venous stasis ulcers, evaluator-blinded. Venous stasis ulcers are notoriously hard to heal. They sit on the leg for months. They look, depending on the day, like geographic territories — irregular, weeping, indifferent to most interventions. Silver sulfadiazine significantly reduced ulcer size. GHK-Cu was no different from placebo.

This is the most rigorous trial that has ever been run on this molecule. It failed.

There is a smaller trial from 2006, of GHK-Cu skincare after CO2 laser resurfacing, n equals thirteen,7 in which the only positive finding was patient self-report of satisfaction — which in a thirteen-person cosmetic trial is essentially a placebo signal. And there is a constellation of cosmetic studies outside PubMed — Leyden, Krüger, Abdulghani, an eye cream comparison, a thigh-skin biopsy study against vitamin C and retinoic acid — published in cosmetic-science journals, several company-sponsored, that report modest but reasonably consistent improvements in measured skin parameters. Fifteen to twenty-five percent gains, over twelve weeks, on things like skin thickness and fine-line density. Real numbers in real studies of real people. Not the studies you would want to base a therapy on. The studies cosmetics companies generate to support marketing claims, and they support those claims, and that is what they were for.

So the clinical picture as of about 2008 was this. A vivid preclinical signal, particularly for ECM remodeling and wound healing. A single negative rigorous human trial. A body of cosmetic data showing modest improvements consistent with what would be marketable, at best, as a perfectly nice anti-aging serum among other perfectly nice anti-aging serums.

This is where Pickart made a turn that reshaped the field, and not for the better.

In 2008, the Broad Institute launched the Connectivity Map.8 CMap is a tool — a transcriptional fingerprinting platform — in which you perturb a cell line with a compound and you measure how the expression of about thirteen thousand genes changes. You get a signature, which you can compare to signatures from disease states, from other drugs, from environmental perturbations. It is a screening tool. It is meant to generate hypotheses. You run a compound through CMap, you find out which expression patterns it is producing, and you go and do real biology to determine whether those patterns are physiologically meaningful in a relevant context.

Pickart’s group ran GHK through CMap in MCF7 breast cancer cells. They identified about four thousand genes that moved by fifty percent or more — a large fraction of those measured — and reframed this finding, in subsequent reviews, as the molecule “resetting four thousand genes to a healthier state.”9 The reframing is everything. CMap measured perturbation; the language of “resetting to healthier” assigned a direction and a moral valence that the data does not contain. The MCF7 cell is a breast cancer line. “Healthier” was never validated. From this single computational screen, the claims about GHK-Cu treating COPD, treating neurodegeneration, having anti-cancer activity all spread out like roots from a single point. The language inflated. The molecule was no longer a peptide that stimulated collagen; it was a molecule that intervened at the level of the genome itself.

This moment is painful to read about, because the underlying data is genuinely interesting, and what was done to it cost the field, I think, decades of credibility.

There is a footnote to this story worth telling, because it is the kind of footnote the peptide community does not amplify and the skeptical community does not mention. In 2012, a group at Boston University and UBC — Joshua Campbell and colleagues, in Genome Medicine10 — used CMap in the way it was supposed to be used. They first profiled gene expression in sixty-four lung tissue samples from patients with COPD, correlating expression with regional emphysema severity by micro-CT. They identified a hundred and twenty-seven genes whose expression tracked with the destruction. Then they used CMap to ask: what compounds in the database produce signatures that would reverse this? GHK came up. And then — and this is the part — they did not stop. They went and validated the prediction experimentally. They treated COPD fibroblasts with GHK and showed the cells recapitulated TGF-β-induced gene expression patterns, organized their actin cytoskeleton, elevated integrin β1 expression, and recovered their ability to remodel collagen in a contraction assay. The computational prediction was confirmed in disease-relevant cells. The paper is competent. It is in a respectable journal. It does not prove GHK treats COPD in humans. It does prove the gene-expression effects of the molecule are functionally meaningful in at least one disease-relevant in vitro context.

This paper is the closest thing the field has to a quiet vindication of Pickart’s gene-expression line of argument. And it has produced, in the fifteen years since it appeared, no clinical follow-up. Nobody has taken the Campbell finding into a Phase I trial. Nobody has done the inhaled GHK study that would be the obvious next step. The paper sits there. It is one of about eighty.

The preclinical work continued, mostly outside the Pickart sphere, and a pattern emerged in it that I do not think the field has fully reckoned with. The Park group in Korea showed GHK-Cu reduced reactive oxygen species and suppressed NF-κB and p38 MAPK signaling in mice with LPS-induced acute lung injury11 — by intraperitoneal injection. The Bian group in 2024 identified peroxiredoxin-6 as a molecular target through which GHK-Cu attenuates inflammation and fibrosis in a silicosis mouse model12 — by intraperitoneal injection — and noted, almost in passing, no significant systemic toxicity. The Fu group at the Chinese University of Hong Kong showed improved graft stiffness in rat ACL reconstruction13 — by intra-articular injection. A mouse cognition study, n of six per group, three weeks at ten milligrams per kilogram, five times a week — parenteral injection — with a note in the paper that GHK appears to cross the blood-brain barrier efficiently. The Maquart wound chamber studies — direct injection into the chamber.

Nearly every impressive preclinical study used an injection route. The molecule, in animals, is an injectable. This matters because the topical-cosmetic story and the injection story have very different evidentiary structures, and the community discourse blurs them — sometimes deliberately, sometimes not.

Here is the silence at the center of the room. There are zero published human pharmacokinetic studies of injectable GHK-Cu. Zero. The half-life numbers that circulate in the peptide community vary from thirty minutes to four hours — a four-to-eight-fold disagreement on the same molecule — and neither figure is grounded in published clinical PK. The dosing protocols that compounding pharmacies and anti-aging clinics use — typically one to two milligrams subcutaneously per day, often on thirty-day-on, thirty-day-off cycles — were developed empirically, in the field, by practitioners. They have never been characterized in a dose-finding trial. If you inject one milligram subcutaneously of a molecule whose physiological plasma concentration is eighty to two hundred nanograms per milliliter, you produce acute plasma levels orders of magnitude above the natural setpoint. Whether supraphysiological dosing of an endogenous regulator is beneficial, neutral, or harmful — and on what timescale — is not a question anyone can answer from the published record.

This is the molecule at the bottom of the staircase.

Meme

What happened around it, in the world above the bench, is a different and more recent story.

Through the 2010s, peptide therapy migrated from the elite anti-aging boutiques where it had quietly lived since the 1990s into something larger and more diffuse. BPC-157, TB-500, ipamorelin, CJC-1295, GHK-Cu. Each rode a wave of podcast attention, Reddit threads, influencer endorsement. The supply moved through compounding pharmacies operating under Section 503A of the Food, Drug, and Cosmetic Act,14 which allows a licensed pharmacy to prepare a drug for an individual patient under a specific prescription, without FDA approval of the underlying compound. This was the regulatory pocket the peptide world had been growing in for years — not a loophole exactly, but a space the agency had not yet decided what to do with.

In 2023, it decided. The FDA’s Pharmacy Compounding Advisory Committee — the PCAC — reviewed nineteen peptides and placed most of them on Category 2: substances that “pose significant safety risks” and are not eligible for compounding.15 The agency’s specific concern about GHK-Cu was phrased as “risk for immunogenicity due to the potential for aggregation and peptide-related impurities.”

Read with any care, this is not a claim about the molecule. A 340-dalton tripeptide is too small to be immunogenic on its own. By the basic principles of immunology, it is a hapten — a molecule below the threshold at which the MHC class II machinery can present it to T cells in a way that would produce an antibody response. A free tripeptide in solution is roughly the least intrinsically immunogenic thing you could inject. What the agency was actually worried about was different. Small compounding pharmacies, without pharmaceutical-grade quality systems, preparing peptide injectables at scale, producing contamination or aggregation or related impurities during compounding that would be immunogenic. That is a manufacturing-quality concern. A real one. Not a molecular-pharmacology concern.

The regulatory action can be read two ways. Narrowly, it is a precautionary call about supply-chain quality — the FDA worrying, with some justification, about what happens when peptide injectables enter the consumer pipeline through unregulated channels. More broadly, it is a regulatory eligibility decision dressed in safety language. Compounded substances generally need to be on FDA-recognized lists, or have a USP monograph, or be a component of an approved drug. None of the nineteen restricted peptides met those criteria. The agency had to put a safety justification on the action, and the language inflated to fit. The simultaneous decision to place topical GHK-Cu on Category 1 — fine to compound — supports this reading. Same molecule. Different stakes when something goes wrong.

The community reaction was immediate and political. Peptides became a wedge issue in the alternative-health space, and the regulatory action transformed them from a niche optimization tool into a symbol. The Alliance for Pharmacy Compounding declined to mount a serious challenge — their executive director on record saying their members weren’t pushing back because the substances didn’t meet the eligibility criteria, which is the institutional way of saying they are right, even though we wish they weren’t. But outside the compounding industry, the framing inverted. The FDA was now a censor. The peptides were a freedom. The supplies migrated to the unregulated international research-peptide market, where the same molecules could be obtained as “research compounds” with no prescription required, no purity standards enforced, and no clinical oversight at all. The clinics that had been operating in plain sight reorganized into looser networks. Use, by every available estimate, did not decline. It went underground.

Then in February 2026, in a podcast appearance with the structure of a coronation, Robert F. Kennedy Jr. — by then Secretary of Health and Human Services — announced on the Joe Rogan show that fourteen of the nineteen restricted peptides, GHK-Cu among them, would be moved back to Category 1. The reversal had been telegraphed for months. By April, the formal updated list had not appeared, but the political signal had. The market was reassured, the community was vindicated, the FDA was further weakened, and the same compounding pharmacies that had been operating in the gray would resume operating in something closer to daylight. None of this was a piece of new clinical evidence. It was a regulatory shift. It told you nothing about whether the molecule worked. It told you, very loudly, who got to decide whether it was permitted to circulate.

So when I sit at the bottom of the staircase and look at the molecule itself, the picture I see is this. Topical GHK-Cu, at moderate concentrations, is reasonably safe and produces modest, real improvements in skin parameters over twelve weeks — fifteen to twenty-five percent gains, the kind of thing genuinely valued by people who care about how their skin looks, and not the kind of thing that would be transformative. The molecule has demonstrable systemic biological activity in injected animal models, with no significant toxicity reported across multiple independent groups. In humans, injected, the molecule is essentially uncharacterized. We do not know its half-life. We do not know its dose-response. We do not have a registry. Thousands of people have received it from compounding pharmacies over years, and no serious adverse events have surfaced in the published record, which is reassuring in an uncontrolled, anecdotal way and tells you very little.

The molecule is plausible. The preclinical work is genuinely interesting. The clinical characterization that would tell us whether it works has not been done. This is the honest position, and it is not a position the discourse is built to occupy.

Equilibria

But the more important question is: Why do we know what we know about GHK-Cu? The shape of the literature — the dense preclinical bench science, the absent clinical program, the regulatory action with the thin scientific rationale, the peptide clinics operating in the gray — is not random. It is the predictable output of an incentive structure, and the structure repeats, with variations, across very different domains. Once you see it, you see it almost everywhere.

GHK is an endogenous tripeptide discovered in 1973. You cannot patent the molecule. You can patent specific formulations, delivery devices, modified analogs, but the molecule itself is public domain, which means whoever spends the eight to fifteen million dollars needed to run a proper Phase I/II safety and dose-finding program will, the moment the data is published, be undercut by competitors selling the same generic compound at a fraction of the price. The rational economic decision, for any entity capable of running the trial, is not to run it. This is the same structural problem preventing clinical development of every off-patent natural product — curcumin at therapeutic doses, melatonin, vitamin D for indications it does not yet have approval for, dozens of repurposable generics. A vast preclinical literature, essentially no clinical development. The incentive structure selects against the work that would resolve the questions.

The regulatory environment that existed before 2023 — and which the post-2026 reversal will largely restore — removed any urgency on the supply side. Under 503A compounding, a physician could write a prescription, a pharmacy could compound the molecule for a specific patient, money could change hands, and nobody needed to demonstrate efficacy in a controlled trial. The cosmetic side was even more permissive. The whole ecosystem was structured so the trials would not be necessary, and so the trials were not done. Pickart ran a small Bellevue company, not a pharmaceutical company, and Skin Biology never had clinical development infrastructure. The preclinical groups — Maquart on ECM biochemistry, Campbell on COPD signatures, Park and Bian on lung pathology — each had narrow specific interests, and none had the funding or motivation to take the molecule into the clinic. The therapeutic profile of GHK-Cu — diffuse anti-inflammatory, broadly pro-regenerative, broadly anti-aging — is itself uncooperative with clinical development. “Anti-aging” is not an FDA-approvable indication. Wound healing is, and the one good wound-healing trial failed. COPD is, and going from a CMap signal to a Phase II program is a jump nobody is willing to fund.

This is the supply-side equilibrium. Every party who could fund the definitive trial has a rational reason not to. The absence of clinical data is not a gap waiting to be filled. It is the stable output of a system with no mechanism for producing the answers anyone claims to want.

When I first started reading this literature, I thought the supply side was where the analysis ended. It is half the picture. The other half is harder to think about, because it implicates me.

The communities that aggregate around molecules like GHK-Cu do not coalesce around them despite the evidentiary ambiguity. They coalesce around them because of it. Consider what properties a molecule needs to become a flagship intervention in optimization culture. It has to be real enough to sustain interest — a genuine preclinical literature, plausible mechanism, endogenous origin, eighty PubMed papers. A molecule with three papers cannot anchor a community; a molecule with three thousand papers and Phase III data has already been absorbed by the institutions and is no longer interesting. It has to be unresolved enough to support identity formation — if the clinical question were settled, there would be no contrarian position to hold, no sense of being ahead of the mainstream. It has to be accessible without gatekeepers — compoundable, orderable, prescribable outside the standard insurance-and-specialist pathway, so participation feels like autonomy rather than supplication. And it has to be narratively rich — discovered by a heroic underdog figure, suppressed by regulatory and commercial forces, naturally occurring in the body, declining with age, holding the promise of restoration. The molecule has to support a story.

GHK-Cu fits all four criteria almost perfectly. So does BPC-157. So does TB-500. So do the NAD precursors. So does rapamycin, off-label. So does methylene blue. There is a structural niche in the contemporary information environment, and these molecules occupy it.

The community’s relationship to the evidence is not what it presents itself as. The presented relationship is: we are following the evidence ahead of mainstream medicine, which is too slow or too captured to act on it. The actual relationship is more like: we are sustaining the conditions under which the evidence cannot be resolved, because the unresolved state is what allows the community to exist. If a definitive trial showed GHK-Cu worked for a specific indication, the molecule would be absorbed into mainstream medicine — prescribed by dermatologists, sold by pharmaceutical companies, covered by insurance. The community would lose ownership of it. If a definitive trial showed it didn’t work, the community would lose a flagship intervention. Either outcome dissolves the community’s relationship to the molecule. The current state — promising, unproven, contested — is the state that maximizes engagement.

Nobody is doing this consciously. It is an emergent dynamic, and you can see it in how the community processes evidence. Positive preclinical findings are amplified and treated as near-confirmatory. Negative clinical trials are explained away with contextual caveats. The lack of trials is attributed entirely to structural barriers, never to the possibility the molecule might not deliver as hoped in humans. The community has incentives to read the literature a particular way, and it does. The commercial ecosystem reinforces this from the other direction. A compounding pharmacy does not want an FDA-approved version of GHK-Cu to exist, because that would destroy its market. A peptide vendor does not want a definitive negative trial, because that would destroy demand. An anti-aging clinic does not want the question resolved, because the ambiguity is what makes its expertise valuable.

The 2023 FDA restriction did something sociologically interesting. It transformed GHK-Cu from an optimization tool into a symbol of regulatory overreach. The community now had a shared adversary. Peptide use was elevated from a personal health choice to a political identity. The 2026 Rogan reversal was perfectly calibrated to this dynamic — a populist health figure, on the most influential podcast in the optimization-adjacent space, framing peptide access as a freedom issue. The pattern recurs across alternative health movements. Regulatory restriction does not suppress demand. It radicalizes demand and channels it into less-regulated spaces, where the same molecules circulate with less oversight than they had before.

The supply-side and demand-side equilibria are not two coexisting problems. They are a single, self-reinforcing structure. Nobody who could fund the trial benefits from its completion. The community whose interest could in principle demand the trial benefits from its absence. The gap between what is known and what is claimed is not a knowledge problem that will eventually be solved by more research. It is a stable feature of the system that produces both the demand and the supply of information about this molecule.

Pattern

I am dwelling on GHK-Cu because the question is concrete and the evidence is bounded, but the structural pattern is not unique to peptides.

The venture capital ecosystem around quantum computing, nuclear fusion, longevity biotech, brain-computer interfaces sits in exactly this shape. The underlying science is real — quantum systems exist, fusion reactions produce energy, senolytics clear senescent cells in mice. The distance from the current state to commercial viability is enormous and genuinely uncertain. Nobody can credibly say when quantum advantage arrives, or when fusion breaks even, because these are open questions at the frontier of engineering, and the structural barrier to resolution is time and capital — definitive answers require decades and billions, and no single entity captures enough of the eventual value to justify the full investment. So a community forms — investors, founders, analysts, journalists, conference circuits — whose value derives from navigating the uncertainty rather than from resolving it. A VC’s value proposition is that they can pick which quantum company will succeed before the market knows. If the question were settled, the expertise would be commoditized and the community would dissolve.

Productivity systems are another version of the structure. Getting Things Done. Zettelkasten. Time-blocking. Second-brain. Pomodoro. Bullet journaling. The underlying signal is real — organizational systems do help some people accomplish more. The actual effect size relative to simply doing the work is probably modest, and is essentially unknowable with the rigor that would settle the question, because productivity in knowledge work is too diffuse to measure cleanly. The community that thrives in this space depends on the question remaining open. A definitive answer would collapse the entire ecosystem of templates and courses into a public-health recommendation.

Nutritional frameworks have the longest visible arc of this dynamic. Low-fat in the 80s, low-carb in the 2000s, paleo in the 2010s, keto, carnivore. Each framework escalates through an arc, peaks, partially collapses under its own overreach, and is replaced by the next, which inherits the audience, the commercial infrastructure, and the epistemic habits of its predecessor. Many people in the current carnivore community are former keto adherents who were former paleo adherents. The substrate changes. The structure persists.

And — this is the version closest to my own work, where it gets uncomfortable to write — the world around psychedelic-assisted therapy and contemplative practice runs the same circuit. MBSR has measurable effects on cortisol and pain perception. Psilocybin produces durable reductions in treatment-resistant depression in trials. The basic signal is real. The stronger claims about ego dissolution, consciousness transformation, fundamental psychological rewiring sit in exactly the same zone of productive ambiguity. The phenomenological reports are vivid. The mechanistic explanations are speculative. The long-term outcomes are hard to measure. The inherent difficulty of objective study of subjective experience prevents resolution. And the value proposition of a meditation teacher, a psychedelic integration guide, a breathwork facilitator depends on the practice being deep enough that guidance is needed and unresolved enough that the guide’s interpretation retains unique value. I have been the patient in that room. I know what the room is for. I also know the room could not exist in the same way if the question of what the medicine does were fully resolved.

What unites these is not truth or falsity. It is the structural prevention of resolution. The underlying phenomenon is real enough to sustain genuine interest. The question of how much it helps, for whom, in what circumstances, is genuinely hard to answer. The difficulty is not only technical but incentive-structural, because nobody who could fund the definitive study benefits from completing it. A community forms whose identity and economy depend on the ongoing navigation of the ambiguity. And the system is self-stabilizing — attempts at resolution are absorbed into the narrative or resisted as hostile.

We tend to think evidence-based medicine has two failure modes: things that work but aren’t proven, and things that don’t work but are believed to. The GHK-Cu pattern shows a third. Things that might work, that we are structurally prevented from determining, around which stable communities of belief and practice form precisely because of that irresolution. This third category may be the largest of the three. It is growing.

Ratchet

I want to be careful about one thing before I go further, because I have been using the word equilibrium and the word understates what is happening. These are not stable states. They are dissipative structures — they maintain themselves by consuming attention, capital, and belief, and they have a directionality. They escalate.

Inside the community, claims ratchet upward. Modest formulations — this is interesting, the preclinical data is promising, this might help — do not sustain engagement over years. The content economy demands novelty, and when the underlying evidence base is not generating new findings at the required rate, novelty is manufactured through reinterpretation. GHK-Cu stimulates collagen in fibroblasts becomes GHK-Cu resets four thousand genes becomes GHK-Cu reverses aging at the epigenetic level. Each step is traceable to something real. The interpretive envelope expands. Outside, skepticism ratchets correspondingly. The initial response to peptide enthusiasm was indifference. Then mild caution. Then a regulatory action thinly supported by molecular pharmacology. Then a political reaction framed as liberation from regulatory overreach. Each side’s escalation provokes the other’s. The amplitude of the oscillation is increasing.

The cycle does not end in resolution. It ends in migration. When a flagship intervention can no longer sustain the required narrative intensity, the community does not conclude the approach was wrong. It concludes the focus was on the wrong instance, and migrates to the next molecule, the next system, the next framework. The infrastructure persists; the substrate changes. The keto adherent becomes the carnivore adherent. The autonomous-vehicle investor becomes the AI investor. The GHK-Cu enthusiast, if the molecule disappoints, becomes the rapamycin enthusiast, or the senolytic enthusiast, or the enthusiast of whatever is next.

Each migration leaves a residue. More developed commercial infrastructure. A more sophisticated community, better at mining literature and constructing narratives. A larger population of participants. And — most importantly — a higher baseline of distrust toward the institutions that could have resolved the ambiguity but did not. The distrust is, in many cases, justified. The incentive structures are real. The FDA’s rationale for the peptide ban was scientifically thin. But the distrust is not molecule-specific. It generalizes. A person who learned through GHK-Cu that the FDA makes decisions on regulatory eligibility rather than evidence, and that pharma cannot evaluate unpatentable molecules, carries that framework into her evaluation of vaccines, public health guidance, nutritional recommendations, everything. The epistemic habits transfer.

The trajectory is not stable. It is a ratchet. Each cycle widens the gap between institutional epistemics and community epistemics, and both sides become more internally coherent and more mutually unintelligible.

Risk

The honest version of what I am describing is Baudrillardian — and I have been trying not to say so, because the moment you say so, you have entered a particular discourse that has its own ratchet. But there is no avoiding it.

The precession of simulacra16 is the process by which the representation detaches from the thing represented and begins to generate its own reality. In the GHK-Cu case we can trace it precisely. The molecule has real properties. The research represents those properties. The community’s interpretation of the research becomes the primary object of engagement. The interpretive ecosystem generates its own internal logic that no longer requires reference to the molecule’s actual properties. At the terminal stage, someone injecting GHK-Cu is participating in a semiotic system — an identity, a narrative, a community, a stance toward institutional authority — that is operationally independent of whether the molecule is doing anything in her tissue.

The same trajectory runs in every domain. The productivity system becomes about the system, not about producing. The investment thesis becomes about the thesis, not about the technology. The dietary framework becomes about the framework, not about nourishment. The contemplative practice becomes about the practice, not about contact. Each domain undergoes the same detachment, and the velocity is increasing, because the information environment selects for interpretive frameworks over direct engagement with the underlying reality.

The thing I am afraid of is not metaphorical.

Consider what has happened to medicine — the actual practice of trying to help bodies feel better. A physician sixty years ago touched patients, listened to breathing, palpated abdomens, formed clinical impressions grounded in direct sensory contact with a body. That physician operated within a much less rigorous epistemic framework — they were wrong more often, their treatments often did not work — but they had contact. The modern physician has imaging, genomics, biomarkers, evidence-based guidelines, meta-analyses of meta-analyses, and a fifteen-minute appointment slot, much of it spent entering data into an electronic health record to satisfy documentation requirements generated by the administrative apparatus above the clinical encounter. The physician’s contact with the patient’s body — the material grounding of medicine — has been progressively displaced by the physician’s contact with representations of the patient’s body. Each representation is more accurate than the physician’s hands. Each interposes another layer between the healer and the person.

The patient feels this. The patient knows, in a way she can feel but often cannot articulate, that something has been lost — that the encounter is no longer fully about her as a material being, but about her data, her risk profile, her position in a treatment algorithm. And the patient goes looking for the lost contact. She finds it in the integrative practitioner who spends an hour with her. She finds it in the peptide community, where people share detailed accounts of how their body feels. She finds it in dietary frameworks that make her pay close attention to what she eats and how it affects her. The alternative health world, for all its epistemic problems, often provides something the mainstream system has structurally abandoned: attention to the material experience of being in a body.

This is the cruel irony at the center of everything I have been describing. The communities that form around GHK-Cu and its analogs are, in part, responses to the semiotic detachment of mainstream medicine. People who feel that institutional medicine has become a system of abstractions seek out spaces where their bodily experience is taken seriously. And then those spaces undergo the same semiotic detachment, for all the structural reasons I have traced, until the peptide community is itself a system of abstractions — dosing protocols, gene expression narratives, community identity, political stance — that has drifted from the body just as thoroughly as the system it was reacting against. The material gets consumed from both directions. Institutional medicine abstracts the body into data. Alternative medicine abstracts the body into narrative. And the person in the body — whose shoulder hurts, whose skin is aging, whose energy is flagging, who just wants to feel better — is caught between two semiotic systems, neither of which is primarily oriented toward her direct material experience anymore.

What is at risk here is not what people usually think. It is not that the material world is being erased. Copper still binds histidine regardless of what anyone believes. What is being eroded, or at least progressively occluded, is the human capacity to orient toward the material — the phenomenological access itself. Not the world. Us. The ability to hold a question like does this help? in a way that is genuinely open, genuinely grounded in bodily experience, without it immediately being colonized by a framework.

I want to be careful here, because what I am pointing at could be misheard as a familiar lament — the kind of thing people have been saying since the printing press, the kind of thing every generation says about the medium of the next. It is not that. The thing I am pointing at is structurally new. It is not that there are more frameworks competing for our attention. It is that there is no longer reliably a pre-framework layer of experience that the frameworks compete to interpret. The frameworks have become continuous with the layer they used to comment on. The interpretation arrives before the perception does. I do not feel a thing and then look for a name for it. The name is handed to me first, by a flood of language that has been waiting for me, and the thing I would otherwise have felt collapses into the name as it arrives.

The fear at the center of this — and I want to name it cleanly, because the analysis I have just done can absorb almost any fear and metabolize it into another framework — the fear is that the structures I have been describing are not just shaping our knowledge but actively eating the substrate of the kind of consciousness that knew anything to begin with. Not by force. By gradient. By making it easier, every day, to live in the layers than to descend to the floor. By selecting, generation by generation, for the kind of nervous system that can metabolize semiotic content efficiently and finds the unprocessed signal painful, vague, embarrassing. By rewriting, slowly, what a human being is for.

I cannot prove this. I am not sure it is possible to prove. I can tell you I feel it, in myself, every day. I feel the difficulty of the descent getting larger. I feel my own attention, which I have spent years trying to train, getting harder to call back from the landings.

And the colonization is not coming from outside. It is coming from the structure of attention itself in an information-saturated environment. The moment I ask whether GHK-Cu works and open PubMed, I am in the semiotic layer. The moment I ask a physician, I am in the institutional-epistemic layer. The moment I ask a community, I am in the identity-narrative layer. Each of these sources contains real information. None is contact with the material. Increasingly there is no path back to the material that does not pass through one of these layers. The layers have become the territory.

Floor

I do not want to end with a redemption. I noticed myself reaching for one. I almost wrote about the austere truth-seeker descending the staircase past the distracted landings, finding her way to the ground floor, holding her position with quiet integrity. It was a beautiful image, and it was, structurally, exactly the kind of elevating narrative the analysis I have just walked through should make me suspicious of. Anyone who claims to have found a stable place from which to observe the dynamics is participating in the dynamics. The meta-awareness is just another layer.

What I have left, after all of this, is not a position. It is something smaller and stranger, which I can only point at because I have lived it. I do not know how to point at it without slipping into the genre of personal revelation the optimization community has trained on and that I have just spent ten thousand words criticizing. The genre is going to win this round of the analysis, because the alternative is to end with another structural observation, and the whole point is that structural observation is part of the problem. So I will let the genre win.

Freud17 — before the institutional Freudianism that betrayed him, before the diagnostic apparatus his work was metabolized into — used the word Eros to mean the drive toward life. Not romance, exactly. Not even sex, exactly. The drive toward connection, toward pleasure, toward the kind of attention that does not abstract. The drive that, when it is allowed to operate, organizes a body around its own continued being. He called it the principle of resistance to thanatos — to the death drive, to the part of the organism that turns against itself. He was, I think, pointing at the same floor this staircase descends to. He just did not have the vocabulary I have now, because he did not live to see what the layers above it could do.

I am a trans woman. I did not know this for the first thirty-some years of my life. The information was in my body, the way the information that GHK-Cu binds copper is in the molecule, but I did not have access to it. What I had access to was a self constructed entirely out of borrowed language — checklists, frameworks, the way the people around me described their bodies, which I would dutifully copy without ever quite fitting into. I had a body. I knew it had needs. I did not know what its needs were. My body was the referent of a discourse, exactly the discourse this essay has been about, scaled down to a single life. The layers had become the territory of me.

The way through, when it came, came through psychedelic medicine. Across two years of esketamine sessions for severe depression,18 in a recliner in a clinic in downtown San Francisco, the layers thinned. Not because the medicine told me anything. The medicine briefly suspended my access to the languages I had been using to construct myself, and underneath, the body had things to say. The sessions did not hand me my transness as a finished thing. What they did was open a channel through which my body, over years, could hint and then say and then insist.

The first day I took estrogen, I knew. Not from the mirror. Not from anything I could call cognitive. I spent the afternoon on the floor of my bedroom, wrapped in a fuzzy blanket — rolling, weeping, laughing. I was inside myself in a way I had never been inside myself before. The blanket against my skin was a percept I had not had access to. My breath in my chest was a sensation I had been having my whole life and never noticed. The wanting in my pelvis was a wanting that had not had anywhere to live before, because there had been nowhere in me it could correctly be felt.

The months that followed taught me what my body had been waiting for. My orgasms became something I have no fluent language for, because the available languages were not built for what I was experiencing. Mystical. Spiritual. Enveloping. Long. The kind of presence in flesh the contemplative traditions point at, which I had always read about as if it were metaphor and which turned out to be a literal description of what is available to a person who is, finally, in the right body. I cannot now imagine living without it. I do not understand how I lived without it for as long as I did. I think — and this is the part I cannot prove, only testify to — that I lived without it because the apparatus around me was not built to recognize the body I had, and so I learned not to recognize it either, and so the body learned not to speak.

This is what Eros is, in the form I have access to it. A sensuality and a sexuality and an attention that does not abstract. A drive toward life, in the specific body where life is happening. Freud, in calling this out as a principle of resistance, was not being metaphorical. The structures I have been describing in this essay — the regulatory, the commercial, the communal, the semiotic — they are not neutral with respect to Eros. They are organized against it. A person fully in contact with her own desire is a person it is difficult to sell anything to. A person in possession of her own body is a person who does not need the protocols. The frameworks have an interest in the body remaining mediated. They have an interest in attention being everywhere except in the flesh.

I want to be careful. The trans community is itself an epistemic community subject to every dynamic I have described. The estrogen I take, like the GHK-Cu my friend injects, is a small molecule with a partial literature and a community formed around it. I have not stepped outside the structure of this essay. There is no outside. What I have, instead, is one more piece of evidence — embodied, partial, mine, not transmissible — that the floor is still there. That underneath the layers there is a layer made of skin and blood and the strange specific weight of a body that knows what it is. The layers had not eaten that. Not yet. Not in me.

I will not tell you how to live. The structure of the telling would be the problem. What I will tell you is that, for me, the staircase had a ground floor. There was an afternoon, on the floor, in a blanket, in a body that was finally mine, when I touched it. The reaching that began in the café — for the protocol, for the citation, for the framework — was, all along, a reaching for that floor. I just did not know which floor I was reaching for, because the layers had named it for me before I could find it on my own.

The Eros at the bottom of the staircase is the one thing the structures cannot reproduce. They can simulate every other layer above it, in increasingly compelling resolution. They are working on simulating this one. They have not yet. I am not sure they cannot. I am not sure the resistance the body offers is, on its own, enough. The fear that runs through everything I have just written is that what is being eroded is not the world but the capacity to feel it — that the floor remains, but the way down is being closed off, gradient by gradient, generation by generation, a slow narrowing of what a human being is for.

I do not know what is on the other side of that fear. Some days I think the resistance Eros offers is irreducible — that there is something in the embodied fact of being alive in a specific body that no amount of mediation can fully metabolize. Other days I think we are already inside the metabolization, and the only honest answer is to grieve.

Tonight, I will go home. I will stand under the warm yellow light of the bathroom. I will put my hands on my own face. I will feel my breath in my chest, and the strange specific weight of being here, in this body, in this life I almost did not get to have. The structures that did not want me to find this can wait. The layers can wait. The staircase can wait.

The floor is here. I can feel it. ∎

Footnotes

  1. On endogenous plasma GHK levels and their decline with age, see L. Pickart and A. M. Margolina, “Regenerative and Protective Actions of the GHK-Cu Peptide in the Light of the New Gene Data,” International Journal of Molecular Sciences 19, no. 7 (2018): 1987, which collates the 200 ng/mL (age 20) and 80 ng/mL (age 60) figures from earlier work; the original values are reported in L. Pickart, “The human tri-peptide GHK and tissue remodeling,” Journal of Biomaterials Science, Polymer Edition 19, no. 8 (2008): 969–988. ↩

  2. L. Pickart and M. M. Thaler, “Tripeptide in human serum which prolongs survival of normal liver cells and stimulates growth in neoplastic liver,” Nature New Biology 243, no. 124 (1973): 85–87. The original isolation paper. ↩

  3. Representative reviews: L. Pickart, J. M. Vasquez-Soltero, and A. Margolina, “GHK Peptide as a Natural Modulator of Multiple Cellular Pathways in Skin Regeneration,” BioMed Research International 2015 (2015): 648108; L. Pickart, “The Human Tripeptide GHK-Cu in Prevention of Oxidative Stress and Degenerative Conditions of Aging,” Oxidative Medicine and Cellular Longevity 2012: 324832; L. Pickart, J. M. Vasquez-Soltero, and A. Margolina, “The Effect of the Human Peptide GHK on Gene Expression Relevant to Nervous System Function and Cognitive Decline,” Brain Sciences 7, no. 2 (2017): 20. ↩

  4. F.-X. Maquart, L. Pickart, M. Laurent, et al., “Stimulation of collagen synthesis in fibroblast cultures by the tripeptide-copper complex glycyl-L-histidyl-L-lysine-Cu²⁺,” FEBS Letters 238, no. 2 (1988): 343–346. ↩

  5. F.-X. Maquart, G. Bellon, B. Chaqour, et al., “In vivo stimulation of connective tissue accumulation by the tripeptide-copper complex glycyl-L-histidyl-L-lysine-Cu²⁺ in rat experimental wounds,” Journal of Clinical Investigation 92, no. 5 (1993): 2368–2376. ↩

  6. J. B. Bishop, L. T. Phillips, T. A. Mustoe, et al., “A prospective randomized evaluator-blinded trial of two potential wound healing agents for the treatment of venous stasis ulcers,” Journal of Vascular Surgery 16, no. 2 (1992): 251–257. ↩

  7. P. J. Miller, J. C. Sykes, and D. Sykes, “A randomized, controlled, blinded study of the efficacy of a topical copper peptide cream on resolution of postoperative changes after carbon dioxide laser resurfacing,” Archives of Facial Plastic Surgery 8, no. 4 (2006): 252–259. ↩

  8. J. Lamb, E. D. Crawford, D. Peck, et al., “The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease,” Science 313, no. 5795 (2006): 1929–1935, and the expanded LINCS L1000 release described in A. Subramanian et al., Cell 171, no. 6 (2017): 1437–1452. The Broad Institute’s CMap platform launched publicly in 2008. ↩

  9. For Pickart’s CMap-derived gene-expression claims and the rhetorical move from perturbation to “reset,” see Pickart, Vasquez-Soltero, and Margolina (2018), op. cit., and the survey of expression changes in “The Anti-Aging and Wound Healing Tripeptide GHK,” in S. P. Schmidt and N. Pallua, eds., Aesthetic Plastic Surgery supplement (Springer, 2015). The MCF7 line is from the original NCI-60 panel; the framing as “healthier state” is interpretive, not data-derived. ↩

  10. J. D. Campbell, J. E. McDonough, J. E. Zeskind, et al., “A gene expression signature of emphysema-related lung destruction and its reversal by the tripeptide GHK,” Genome Medicine 4, no. 8 (2012): 67. This is the load-bearing “CMap done right” paper. ↩

  11. J.-R. Park, H. Lee, S.-I. Kim, and S.-R. Yang, “The tri-peptide GHK-Cu complex ameliorates lipopolysaccharide-induced acute lung injury in mice,” Oncotarget 7, no. 36 (2016): 58405–58417. ↩

  12. J. Bian, X. Liu, et al., “Peroxiredoxin 6 mediates protective function of astrocytes in Aβ proteostasis,” and the GHK-Cu silicosis follow-up in Redox Biology 71 (2024): 103237. The 2024 paper specifically notes absence of significant systemic toxicity at the IP doses tested. ↩

  13. S.-C. Fu, P. S.-H. Yung, K.-M. Chan, et al., “Effect of intra-articular injection of the tripeptide GHK-Cu on tendon-bone healing in a rat ACL reconstruction model,” Journal of Orthopaedic Research 33, no. 7 (2015): 1024–1031. Benefit at 6 weeks did not persist to 12. ↩

  14. Federal Food, Drug, and Cosmetic Act § 503A, codified at 21 U.S.C. § 353a. For the regulatory context and the gray-market peptide ecosystem the section enabled, see G. C. Outterson, “Regulating Compounding Pharmacies After NECC,” New England Journal of Medicine 367, no. 21 (2012): 1969–1972, and subsequent FDA guidance documents. ↩

  15. FDA, Pharmacy Compounding Advisory Committee, 503A Bulks List, second interim final rule, with the Category 1/2/3 framework codified at 21 CFR § 216.10. The September 2023 PCAC vote and the November 2023 placement of injectable GHK-Cu on Category 2 are documented in the corresponding PCAC briefing materials (FDA-2015-N-3534). The HHS reversal announced in February 2026 was issued as a Secretarial guidance memorandum; the formal Federal Register update remains pending as of this writing. ↩

  16. J. Baudrillard, Simulácres et Simulation (Galilée, 1981); English trans. S. F. Glaser, Simulacra and Simulation (University of Michigan Press, 1994). The precession-of-simulacra schema appears in the opening chapter. ↩

  17. S. Freud, Jenseits des Lustprinzips (1920); English trans. J. Strachey, Beyond the Pleasure Principle, in The Standard Edition of the Complete Psychological Works of Sigmund Freud, vol. XVIII (Hogarth, 1955), 1–64. The Eros/Thanatos pairing is developed across that volume and revisited in Civilization and Its Discontents (1930). ↩

  18. For the clinical evidence base behind esketamine in treatment-resistant depression: E. J. Daly, M. Trivedi, A. Janik, et al., “Efficacy of Esketamine Nasal Spray Plus Oral Antidepressant Treatment for Relapse Prevention in Patients With Treatment-Resistant Depression: A Randomized Clinical Trial,” JAMA Psychiatry 76, no. 9 (2019): 893–903; and V. Popova, E. J. Daly, M. Trivedi, et al., American Journal of Psychiatry 176, no. 6 (2019): 428–438. The standard administration model is in-clinic, observed-dose, with a long-term maintenance schedule. ↩