Artificial Intelligence, Humanistic Ethics (Daedalus 151(2):232-243)

John Tasioulas · 2022 · Daedalus 151(2), Spring 2022, AI & Society issue anchor high priority coded

Main argument

Thesis: the 'optimizing mindset' of computer science and economics has installed preference-based utilitarianism as AI ethics' default, but it is open to serious objections; a HUMANISTIC ethics for AI has three interrelated features (the three Ps): (1) PLURALISM - values are plural (well-being elements AND moral components) and INCOMMENSURABLE - there is often no single optimal decision, only a limited array of rationally eligible alternatives - which undermines any optimizing function and any master concept (human rights can't cover environmental impact; trustworthiness is parasitic on more basic values); consequently much apparent 'noise' in human judgment (Kahneman/Sibony/Sunstein) may be ACCEPTABLE VARIABILITY within the eligible range, and bail decisions are multivalue problems wrongly reduced to abscond-probability predictions; (2) PROCEDURES not only outcomes - incommensurability grounds reasons to assign decisions to humans (autonomy in choosing among eligible paths), and even given a single correct answer, process values differ by domain: cancer diagnosis (soundness is all) vs criminal sentencing (a robot judge sacrifices reciprocity among citizens central to the rule of law; ML 'explanations' may not offer intelligible reasons a defendant can grasp; machines 'do not have a share in human solidarity and cannot be held accountable... in the way that a human judge can'); (3) PARTICIPATION - well-being is not a passive end-state (the happiness-drug case) but requires agency, individually and as self-governing democratic citizens.

Why it matters here

The three-Ps manifesto against preference-utilitarian AI ethics: Pluralism (values plural AND incommensurable - no master concept, not even human rights or trustworthiness), Procedures (process values matter beyond outcomes - the robot-judge vs cancer-diagnosis contrast), Participation (agency in decisions, not passive end-states). The Oxford philosophical anchor for the dissertation's pluralist posture, with the sharpest published statement that 'noise' may be acceptable variability among rationally eligible alternatives.

Reading notes

Targeted extraction from the Daedalus volume (essay pp. 232-243). Tasioulas directs the Oxford Institute for Ethics in AI. His robot-judge/cancer-diagnosis contrast and the noise-as-eligible-variability argument are directly usable in all three case chapters.

Tasioulas, J. (2022). Artificial Intelligence, Humanistic Ethics. Daedalus, 151(2), 232-243.

Close reading — 6 coded units

#1 · pp. 235 · argument

“This pluralism of values abandons the comforting notion that the key to the ethics of AI will be found in a single master concept, such as trustworthiness or human rights. How could human rights be the comprehensive ethical framework for AI when, for example, AI has a serious environmental impact beyond its bearing on anthropocentric concerns? [...] Being parasitic on compliance with more basic values, trustworthiness cannot itself displace those values.”

NF-PLURAL-OTHER VC-INTRA-VALUE

#2 · pp. 235–236 · argument

“Beyond the pluralism of values is their incommensurability. [...] although some decisions will be superior to others, there may be no single decision that is optimal [...] This incommensurability calls into question the availability of some optimizing function that determines the single option that is, all things considered, most beneficial or morally right.”

NF-CONSEQ VC-INTRA-VALUE

#3 · pp. 236 · argument

“such decisions typically address multivalue problems, and there is no guarantee that there is a single best way of reconciling the competing values in each case. This means [...] that much of what looks like noise may be acceptable variability of judgments within the range of rationally eligible alternatives.”

TU-METHOD

#4 · pp. 237 · argument

“When it comes to cancer, generating the most accurate diagnosis may be all-important [...] In criminal sentencing, however, being sentenced by a robot judge—even if the sentence is likely to be less biased or less 'noisy' than one rendered by a human counterpart—appears to sacrifice important values, such as the ideal of reciprocity among fellow citizens that is central to the rule of law.”

VC-ROLE

#5 · pp. 237 · argument

“How does it feel to contemplate the prospect of a world in which judgments that bear on our deepest interests and moral standing have, at least as their proximate decision-makers, autonomous machines that do not have a share in human solidarity and cannot be held accountable for their decisions in the way that a human judge can?”

AG-MORAL-CON RL-LEGITIMACY

#6 · pp. 238 · argument

“[Participation:] These end states could in principle be brought about through a process in which the person who enjoys them is passive: for example, by the government putting a happiness drug into the water supply. Contrary to this passive view, it would stress [active participation in decision-making, individually and as self-governing democratic citizens].”

RL-LEGITIMACY

Synthesis-matrix row

supports T2-PREFERENTISM-BROKEN

preference-utilitarian default attacked; incommensurability

complicates T3-PROCEDURALISM-INCOMPLETE

procedures matter beyond outcomes - but as plural values, not as replacement for ethics

supports T4-ROSSIAN-DEMAND

pluralism + incommensurability + eligible-alternatives structure

supports T5-AGENCY-DENIED-EVALUABILITY-KEPT

no solidarity-share, no accountability-capacity

supports T6-RESPONSIBILITY-UNALLOCATED

accountability-incapacity of machine decision-makers as dignity loss

Memos (2)

theoretical · unit #2

Tasioulas is the philosophical anchor the dissertation's pluralism should cite FIRST: the three Ps state, with Oxford authority and in 2022, the exact combination the dissertation builds on - plural incommensurable values (Rossian structure), procedures mattering beyond outcomes (Baum's execution scope), participation (the stakeholder corpus's rationale). Unit 3 (noise as eligible variability) additionally arms the methodology chapter against a subtle objection: folk disagreement in the corpus is NOT necessarily error to be averaged away (contra Condorcet-style treatments, cf. S&K unit 8) - some of it is rational variability within the eligible set, and the convergentist method's job is to distinguish eligible-range variation from genuine conflict. That distinction - operationalized via the corpus's reasoning codes - could be an original methodological contribution.

thesis-link · unit #4

Units 4-5 give the case chapters their sharpest evaluative contrast: the SAME accuracy-first logic that vindicates AI in diagnosis (AI Scribe's home domain) DELEGITIMATES it in sentencing-like contexts where reciprocity/accountability are constitutive (AI Interviewer for hiring; Chibook for immigration status - both closer to sentencing than to diagnosis in what they do to a person's standing). Tasioulas's question - what is lost when the proximate decision-maker 'cannot be held accountable... in the way that a human judge can' - IS the dissertation's responsibility question asked from the dignity side. And his explanation-the-defendant-can-grasp point anticipates the intelligible-reasons requirement (S&K's justification criterion, J&N's contestability).