SYNTHESIS NOTE
Topics›Psychology Users›this note

Do consumers actually want to delegate shopping decisions to AI?

Accenture's survey reports 74% of consumers trust AI agents with purchases, but the data reveals selective delegation bounded by brand loyalty and safeguard requirements. The question asks whether stated willingness reflects genuine appetite or reflects survey framing and vendor bias.

Synthesis note · 2026-10-15 · sourced from Psychology Users

Accenture reports a survey of 25,590 people across 16 countries in which "nearly three in four (74%) would trust a personal AI agent more than their best friend to make a purchase on their behalf." The firm frames this as consumers being "ready to delegate decisions to AI agents, from discrete tasks to fully autonomous agentic purchases." But the same survey shows delegation is bounded, not total: "More than half (56%) would tell their agent which brands to consider, yet 37% of loyal consumers would let an agent switch for a better fit." Accenture also reports that "once protections are in place, 80% would consider delegating at least one daily activity," and that among weekly generative-AI users, gen AI has "already overtaken the physical store as the leading discovery channel."

Accenture's own explanation for the delegation is practical overload, not technological enthusiasm: consumers are "overwhelmed by too many choices, too little time and too much fine print," so they hand off "what feels like hard work" while holding on to "what carries personal meaning." The 80%-once-protected figure signals that reported willingness to delegate is conditional on trust infrastructure existing first, not a baseline preference independent of safeguards.

This selective-delegation pattern complicates Does the personal assistant model actually serve most users?, which found that "automated email, calendar, and routine task management" appeals only to a narrow slice of users. Accenture's figures suggest broader appetite for delegation in the purchasing domain specifically — though the two findings may not conflict: shopping tasks that feel like "hard work" (comparison, fine print) are a different kind of chore than the routine admin tasks the narrower finding addressed. The conditional nature of the 80% figure also echoes Why do capable AI agents still fail in real deployments?: trustworthiness as a precondition for adoption, not capability alone, is exactly what Accenture's "once protections are in place" caveat implies.

What this excerpt does not establish is whether reported trust translates into actual delegated behavior, since the figures are self-reported attitudes and stated intentions from a consultancy survey, not measured transaction data. Accenture is also a vendor that sells agentic-AI strategy and transformation services to the brands this report is addressed to, which gives it a direct interest in finding — and publicizing — high consumer readiness to delegate. The appropriate inference is that consumers report comfort with agent-mediated purchasing under hypothetical conditions, not that agent-mediated purchasing is currently occurring at this scale.

Inquiring lines that read this note 18

This note is a source for these research framings, grouped by the broader line of inquiry each explores. Scan the bold lines of inquiry; follow any specific question forward.

What systematic biases emerge in AI shopping agents' decision-making? Why do consumers overtrust AI preference agents despite limited understanding?

Related concepts in this collection 5

This note in its neighbourhood — explore the map, then jump to a related concept in the list below.

Concept map
14 direct connections · 85 in 2-hop network ·medium cluster Open in graph ↗

Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph

your link semantically near linked from elsewhere

Related papers in this collection 8

Papers most semantically related to this note, ranked by cosine similarity in the embedding space.

Original note title

Accenture's survey finds consumers trust AI agents more than their best friend but delegate selectively