Society · Trend

Plurality of Human Preferences for AI Systems

Formal model shows local pairwise comparisons fail to capture global priorities like proportionality and egalitarianism under internal pluralism.

Trend strength 6/10

Momentum +6/q

Confidence medium

Status new

Forecast horizon

Culturally tagged training datasets may become a standard requirement for globally deployed AI systems.

Connections

Connections · 1

How this node ties into the rest of the map, and the evidence behind each link.

from · informs 3/10

Emergent Misalignment from Narrow Fine-Tuning

Both findings challenge the assumption that RLHF and fine-tuning reliably produce aligned models that reflect diverse human preferences.

+3 growth

Signal sources

Dated facts from primary sources in this direction.

AI in government Sep 2025

An OECD review of ~200 government AI use cases found 57% support automating or tailoring public services and 45% enhance decision-making — most still stuck in pilots.

OECD — Governing with AI →

Labor signal Nov 2025

A Stanford payroll study found a 13% relative decline in employment for workers aged 22–25 in AI-exposed occupations since late 2022, while older peers held steady.

Stanford Digital Economy Lab →

Information integrity Oct 2025

Microsoft reported AI-enhanced phishing reached a 54% click-through rate (4.5× traditional) and AI-generated fake IDs grew 195% globally in 2025.

Microsoft — Digital Defense Report 2025 →