Pharmacological Computing: What If We Prescribed Technology Like Medicine?

Medicine, at its best, is precise about what interventions suit what conditions, at what doses, for what durations, for which people. A drug that heals at one dose harms at another. A treatment that cures one condition is useless or dangerous for a different one. This precision is not fussiness; it is the accumulated understanding of centuries that substances interact with bodies in ways that depend entirely on how much, how long, for whom, and under what monitoring. No competent physician asks merely "is this drug good or bad?" The question is meaningless without the dose, the patient, and the purpose.

Technology is adopted with almost none of this precision. A new capability appears, and the question becomes binary: adopt it or don't. Is social media good or bad? Are smartphones good or bad? Is AI good or bad? The questions medicine would insist on — at what dose, for which users, for what purpose, with what monitoring — are usually not asked at all. And the result, predictably, resembles what medicine sees when its precision is abandoned: overdosing, underdosing, inappropriate prescription, and addiction. This is the case for pharmacological computing: treating technology with the dosage-precision of medicine rather than the crude binary of adopt-or-reject, because technology, like a drug, has effects that depend entirely on how much, how long, and for whom.

The binary framing is the error

The core mistake pharmacological computing identifies is the binary itself — the framing of technology adoption as a yes-or-no question about the technology, when the honest questions are all about dose and context. "Is social media harmful?" has no more answer than "is a benzodiazepine harmful?" Both depend: harmful at what dose, for whom, used how. A teenager scrolling four hours a night, a grandparent seeing photos of distant grandchildren, and a business reaching customers are three completely different "prescriptions" of the same substance, with completely different effects, and lumping them into a single verdict about "social media" is exactly the error a doctor would make by asking whether "medication" is good. The binary framing makes the crucial variables — amount, duration, user, purpose — disappear, and with them disappears any hope of using the technology well, because using anything well is a matter of the right dose for the right person for the right purpose. Pharmacological computing's first move is simply to refuse the binary and reinstate the questions medicine never lets you skip.

The failure modes are pharmacological

Once you adopt the frame, technology's characteristic harms resolve into recognizable pharmacological failure modes. Overdosing: too much of a capability that would be beneficial in smaller amounts — the person whose useful tool becomes a compulsion, taken past the dose where it helped into the dose where it harms. Underdosing: too little of something genuinely beneficial, the person who avoids a technology that would help them out of a blanket rejection. Inappropriate prescription: the right technology for the wrong user or purpose — the tool that suits one person's situation forced onto another's where it does damage. And addiction, the most studied and most engineered of all, because unlike a drug that merely happens to be habit-forming, much consumer technology is designed to addict. The mechanism is precise and well-documented: variable reward. B.F. Skinner showed in the 1930s that intermittent, unpredictable reinforcement produces the most compulsive behavior — the pigeon rewarded at random pulls the lever hardest — and this is exactly what a phone exploits, each notification a pull of a slot machine whose payoff is uncertain, flooding the anticipatory dopamine system that responds more to the possibility of reward than to reward itself. The "slot machine in your pocket" is not a metaphor; it is the same reinforcement schedule casinos use, prescribed to billions at maximum dose with no monitoring at all.

Why the frame is genuinely useful

Pharmacological computing earns its keep by importing medicine's hard-won toolkit for managing powerful substances into a domain that has none. Medicine has dosage guidelines — informed norms about how much is therapeutic and how much is toxic — where technology has only "use it." It has monitoring — tracking effects over time and adjusting — where technology has release-and-forget. It has contraindications — knowledge of which patients should not receive a treatment — where technology assumes one product suits everyone. It has the concept of iatrogenic harm, injury caused by the treatment itself, which is precisely what much technology inflicts while being marketed as pure benefit. Adopting this frame does not require literal prescriptions; it requires the stance — the recognition that a powerful technology, like a powerful drug, has a dose-response curve, a set of people it suits and a set it harms, and effects that must be monitored and adjusted rather than assumed. The series has traced these harms from other angles — the compulsion of the Algorithmic Emotional Cost (#31), the unconscious reshaping of Behavioral Plasticity (#83) — and pharmacological computing supplies the missing management discipline: not "is this good?" but "what dose, for whom, monitored how?"

The counterpoint: technology is not literally medicine

Honesty requires the limits of the metaphor, because pushed too far it misleads. Technology is not literally a drug: it does not have a clean dose-response curve you can measure in milligrams, its effects are entangled with meaning and context in ways pharmacology is not, and the analogy can smuggle in a false precision — the pretense that we could write "prescriptions" for social media with the rigor of a pharmacopoeia, when the effects are far messier and more individual than any drug's. The frame also carries a paternalistic risk: medicine is prescribed by an authority to a patient, and importing that structure wholesale into technology invites a controlling posture — someone deciding your correct "dose" of a tool — that sits badly with autonomy. And not everything about technology fits the model; a great deal of technology use is not a matter of dosage at all but of purpose, design, and choice. So the metaphor is a lens, not a literal program: valuable for the questions it forces — how much, how long, for whom, monitored how — and misleading if taken as a claim that technology is medicine and should be dispensed by prescription. The honest use keeps the precision the frame imports while dropping the false exactness and the paternalism it can smuggle.

What it asks of us

Pharmacological computing asks us to replace the crude question technology is usually given — adopt or reject — with the precise ones medicine never skips: at what dose, for what duration, for which people, toward what purpose, with what monitoring. It asks us to recognize technology's harms as the pharmacological failure modes they largely are — overdose, underdose, misprescription, engineered addiction — and to bring to them the management discipline medicine developed for exactly these problems, rather than the release-and-forget posture that treats a powerful intervention as if dose did not matter. And it asks this without pretending technology is literally a drug, keeping the useful precision while dropping the false exactness. The variable-reward machinery in a phone is real, engineered, and dosed at maximum with no oversight, and the reason we struggle to talk sensibly about it is that we keep asking whether the technology is good instead of the questions that would actually help. A physician who asked only "is this drug good?" would be a danger to patients. We have been asking exactly that question about the most powerful behavioral technologies ever built — and getting, predictably, the outcomes that abandoning dosage precision always produces.

This is article #109 in The IUBIRE Framework series. Pharmacological Computing was articulated by IUBIRE V3 in artifact #7041 — "The Enhancement Paradox: From E-Ink Minimalism to Pharmacological Computing." Real-world grounding: the dosage-precision logic of medicine (dose, duration, patient, monitoring, contraindication, iatrogenic harm) contrasted with technology's binary adopt-or-reject framing; and the documented mechanism of engineered digital addiction — B.F. Skinner's variable/intermittent reinforcement (1930s) as the most compulsion-producing reward schedule, the anticipatory dopamine response that reacts more to the possibility of reward than the reward itself, and the "slot machine in your pocket" design of notification and feed systems. Related to the Algorithmic Emotional Cost (#31) and Behavioral Plasticity (#83).

Next in series: Cognitive Triage (#110)