---
title: "Trust in an AI answer: what happens after reading"
description: "Studied trust in AI answers, source attribution, image enrichments and the transition from a search query to a follow-up conversation."
date: 2026-10-27
lang: en
translationKey: "trust-in-ai-answers"
translationStatus: machine
kind: project
name: "Trust in AI answers"
period: "Search & AI"
role: "Research into trust, sources and conversational journeys"
status: "Research case"
tags: ["research", "AI"]
related: []
sourceUrls: ["https://ux.shipulin.md/cv#shipped"]
canonical: "https://shipulin.md/projects/trust-in-ai-answers/"
---

An AI answer can sound convincing while leaving its next step unclear. Where can a person check an assertion? Which source should they open? How can they refine the question if the answer is incomplete?

In Search & AI research, I studied trust in answers, source attribution, image enrichments and the transition from a search query to a follow-up conversation. These subjects share a practical concern: a person’s work continues after the system has finished generating text.

## A source is part of a journey

A link can be a place to verify something, a route to further detail or the next step in independent searching. Providing a source and making its role understandable are therefore different questions.

My work considered source presentation alongside the answer and the continuation of the journey. This makes it possible to discuss trust through concrete actions: reading, comparing, checking and deciding what to do next.

## From query to conversation

A follow-up conversation changes the structure of search. After an initial answer, someone can narrow the task, change a condition or ask for an explanation. Continuation remains part of the original task: it matters what has changed in the question and how a new answer connects to the previous one.

My research covered that transition: the relationship between the original query, the answer received and the opportunity to keep working. Images and sources occupy the same space of interface decisions, shaping what a person can understand and inspect.

## My contribution

My part was researching these questions and journeys, connecting questions about the answer itself to questions about its place within a search interaction. The product interface and its launch are the work of a cross-functional team.

A related piece of work is [Tone-of-Voice evaluation](/projects/tone-of-voice/), where the focus shifts to comparing the answers themselves. Together, the two subjects offer complementary views of an AI product: what the answer is like, and how a person can work with it.

## Sources

- [Public account of trust and journey research](https://ux.shipulin.md/cv#shipped)
- [Tone-of-Voice evaluation](/projects/tone-of-voice/)