Aesthetic Value and the AI Alignment Problem

Alice C. Helliwell · 2024 · Philosophy & Technology 37:129 background low priority coded

Main argument

Thesis: value alignment targets ALL human values, not only morally relevant ones; explores whether an alignment problem arises for aesthetic value and what aligning AI with human aesthetic values would involve - extending the alignment problem's scope to non-moral evaluative domains.

Why it matters here

Extends the alignment problem beyond moral values to AESTHETIC value - a domain-extension proof that 'human values' in alignment covers all evaluative domains. Marginal but useful for the S&B point that overall deontic verdicts span multiple normative domains.

Reading notes

Compact treatment. Abstract read.

Helliwell, A. C. (2024). Aesthetic Value and the AI Alignment Problem. Philosophy & Technology, 37, 129.

Close reading — 1 coded units

#1 · pp. 1 · claim

“solutions to the value alignment problem target all human values, not only morally relevant ones. Is there a value alignment problem in other domains? In this paper, I explore whether the AI value alignment problem extends [to aesthetic value].”

VC-INTRA-VALUE ?

Synthesis-matrix row

supports T4-ROSSIAN-DEMAND

cross-domain value plurality extends the weighing problem

Memos (1)

comparison · unit #1

One-cite support for STEINGRUEBER_BAUM unit 5's claim that AI's normative constraints exceed morality (moral + legal + social + now aesthetic reasons) - the cross-domain weighing problem is broader than ethics, which strengthens the case for a general reasons-weighing (Rossian-style) architecture over any single-domain solution.