Note: The ethical judgments on this page refer exclusively to the action — never to the person who performs it or who came into existence through it. Cf. Note on Ethical Judgments.
A governance framework is a set of rules for the containment of a technically mediated practice. Its typical stock is remarkably stable across the fields of application: human oversight, validation of the systems employed, transparency about how they work, assignment of responsibility to named offices, data protection, information of those affected, and equity of access. Whoever surveys the relevant specialist literature on algorithmic decision-making in medicine finds these seven points, in varying order, almost everywhere.
The concept is not meant pejoratively, and this observation does not stand here out of politeness. Governance frameworks achieve something real. They prevent harms that would occur without them. They create attributability where otherwise there would be none. They give those affected claims that would otherwise not exist. As against an unregulated practice they are an advance, and whoever thinks little of them because they do not achieve everything confuses range with worth. The objection of this page concerns the range exclusively.
Ontological classification
- is a subclass of: normative set of rules for a technically mediated practice
- answers exclusively: consequence-based objections
- leaves untouched: act-based objections
- presupposes: the fundamental admissibility of the regulated practice
- demands among other things: human oversight — whose failure is nominal oversight
- stands in connection with: responsibility gap, responsibility
The third line is the most important. A framework regulates how something happens. Thereby it is tacitly settled that it happens. This presupposition is seldom stated, because it follows from the genre of the text: a set of rules for a practice one held to be inadmissible would be a pointless document.
Why the range is limited
Objections to a practice fall into two kinds. A consequence-based objection is directed against effects, risks or attendant circumstances and lapses when these lapse; it is remediable by procedure. An act-based objection is directed against the action itself, independently of circumstances, diligence and consequences; it is remediable by no procedure and can only be contested or acknowledged.
If one goes through the seven points one by one, it shows itself that each of them answers a consequence-based objection. Oversight answers the risk of uncontrolled outputs. Validation answers the risk of unreliable results. Transparency answers a lack of retraceability. Assignment of responsibility answers diffusion of attribution. Data protection answers abusive use of data. Information of those affected answers a deficit of consent. Equity of access answers unequal distribution. One and all these are objections that disappear along with their object — a fully validated, transparent, supervised, data-protection-compliant and equitably accessible procedure gives occasion to none of them any more.
An act-based objection survives this completeness. The objection of Donum vitae against in vitro fertilization (IVF), for instance, holds expressly also for the case in which no embryo perishes; it hangs on the separation of procreation from the personal act of the spouses and not on a harm. The objection to the discarding of a person by characteristics likewise does not hang on a harm — it obtains independently of how reliably, transparently and supervisedly the discarding takes place. No point of the framework addresses these objections, because no point of the framework lies on their level.
From this follows the practically consequential sentence: where an act-based objection obtains and only a governance framework is offered, the question of the admissibility of the practice is not answered but passed over. For this passing over no formal revocation is needed and no express rejection. It suffices that the debate be given a different object. Whoever discusses the quality of the oversight is no longer discussing whether what is being supervised there ought to be supervised at all — and the more carefully the first discussion is conducted, the less the absence of the second is noticed.
The special case of equity of access
Among the seven points one deserves attention of its own, because it goes furthest beyond a merely procedural statement. The demand for equity of access — that a procedure must not remain reserved to the wealthy — not only presupposes the legitimacy of the practice but treats its wider availability as a good.
The difference from the remaining points is considerable. Whoever demands transparency says: if at all, then retraceably. Whoever demands oversight says: if at all, then under control. Both are compatible with an open attitude to the fundamental question, indeed even with rejection — one can demand that something undesirable at least happen under control. But whoever demands equitable access says: to withhold this is an injustice. That is no longer a procedural statement but a substantive value judgment about the matter itself.
The point is thereby not refuted. On the assumption that the practice is admissible, the demand is right and even urgent; unequal access to a legitimate good is a serious problem. But the demand cannot itself justify this assumption, and it does not do so either — it stands in the framework as though it were of the same kind as the other six, and it is not.
The strongest objection to this assessment
The most serious counter-objection runs: a set of rules can also limit a practice in fact or render it impossible. Whoever reads governance merely as evasion underestimates its real blocking effect. Strict ties to indications, binding maximum numbers of embryos produced, requirements of approval for individual cases, prohibitions of certain evaluations of characteristics — such provisions demonstrably prevent more than any moral appeal. German embryo protection law is a set of rules and works in several places like a prohibition. A framework that ties an application to conditions that are practically unfulfillable has ended it, no matter what heading it bears.
The objection holds, and it strikes a real weakness in the thesis if one takes it too coarsely. Two refinements nevertheless sustain it. First: factual blocking effect and conceptual range are two different things. A tie to indications that reduces a practice to zero has not answered the act-based objection — it has coincidentally corresponded to it. The difference becomes visible as soon as the condition becomes fulfillable: if the price falls, if reliability rises, if the situation regarding indications changes, then the block falls away while the act-based objection stands unchanged. A block that depends on empirical circumstances is a consequence-based answer, even if its effect at the moment amounts to a prohibition.
Second, and more importantly: precisely for this reason the distinction is practical and not merely conceptual. Whoever rejects a practice on act-based grounds has occasion not to be reassured by regulatory blocks — they are a truce, not an answer. Whoever rejects it on consequence-based grounds does well, conversely, to rely on sets of rules, because they are the only means that works on his level. The distinction does not say which means are effective; it says which question a means answers.
What remains open: in practice, frameworks are seldom of one kind throughout. A provision such as “no selection by characteristics without disease value” is in form a procedural rule and in content the acknowledgment of an act-based objection. Such cases exist, and they show that a framework can take up an act-based limit. The thesis of this page is not that it never does so, but that it does not do so by virtue of being a framework — and that the examination whether it does so is to be carried out at each individual point.
See also
- Act-Based Objection — what a framework does not reach
- Consequence-Based Objection — what it answers
- Nominal Oversight — the failure of the first point of the framework
- Responsibility Gap, Responsibility
- Algorithmic Embryo Selection, Algorithmic Embryo Assessment
- In Vitro Fertilization, Preimplantation Genetic Diagnosis (PGD)
- AI-Algorithmic Arrangement
- Intrinsically Evil Act
Sources: Generated by querying the ontology of personhood. Research as of 29 July 2026.
Further sources:
- Koplin, J. J., Johnston, M., Webb, A. N. S., Whittaker, A., Mills, C. (2025): Ethics of artificial intelligence in embryo assessment: mapping the terrain. Human Reproduction 40(2): 179–185.
- Congregation for the Doctrine of the Faith (1987): Donum vitae. Instruction on Respect for Human Life in Its Origin, II,B,4–5.
- Congregation for the Doctrine of the Faith (2008): Dignitas personae. Instruction on Certain Bioethical Questions, nn. 14, 17, 22.
- Dicasteries for the Doctrine of the Faith and for Culture and Education (2025): Antiqua et nova. Note on the Relationship Between Artificial Intelligence and Human Intelligence, n. 74.