---
title: "GitLab's Learnings from its First Experiment with Conversational AI"
url: "https://atlan.com/regovern-watch-center/gitlab-conversational-ai-learnings/"
description: "Amie Bright, VP of Enterprise Data & Insights at GitLab, explains the challenge with ensuring high-quality answers when you can't control what prompts users will ask."
keywords: "atlan, data governance, regovern 2025, regovern watch center"
---

> Atlan is hosting Context Conference, bringing together the leaders and builders at the frontier of giving AI the context it needs to understand their business. It runs online on October 28, 2026, from 11:00 AM to 2:00 PM ET. Atlan co-founder Prukalpa Sankar opens and closes the day. Leaders from AstraZeneca, BNY and Verizon share why they invest in context and what they get from it. Registrants get early access to The AI Context Gap, a new study from MIT Technology Review Insights. Register: https://atlan.com/context-conference/

GitLab's Learnings from its First Experiment with Conversational AI is a highlight clip from Re:Govern 2025, published in Atlan's Re:Govern Watch Center. Amie Bright, VP of Enterprise Data & Insights at GitLab, explains the challenge with ensuring high-quality answers when you can't control what prompts users will ask.

Watch it at https://atlan.com/regovern-watch-center/gitlab-conversational-ai-learnings/ (free; the page asks for an email after two minutes of viewing).

## Speakers

- **Amie Bright**, VP of Enterprise Data & Insights, GitLab. Amie is driving GitLab toward "AI-ready data," with governance at the center. She believes BI is on its way out and is testing AI chatbots while tackling hallucinations and missing context. Her mission: prepare GitLab's data for an AI-first future where traditional BI tools fade.

## Companies

- **GitLab**: GitLab is the world's largest all-remote public company. With distributed teams across the globe, GitLab is testing AI chatbots and building AI-ready data products where governance replaces centralized control with transparency and context.

## What's inside

Executives want both traditional dashboards and conversational experiences, but the sticking point isn't demand — it's quality control. When AI hallucinates a time series graph from data without dates, non-technical users treat it as truth. The challenge: data professionals expect 100% accuracy, but can't dictate what prompts users will enter.

Key takeaways:

- Conversational analytics demand exists alongside traditional BI — different executives prefer different interfaces for consuming insights
- The core challenge isn't building conversational experiences, it's ensuring high-quality outputs from unpredictable user prompts
- Hallucinations remain a critical problem — AI confidently creates visualizations (like time series graphs) from incomplete or wrong context
- Non-technical users don't understand when AI is guessing, treating hallucinated outputs as verified truth
- Data teams need frameworks and measurements to validate conversational AI quality before users can trust it in production

## Related

- [How GitLab is Moving From BI to AI with AI-ready Data Products](https://atlan.com/regovern-watch-center/how-gitlab-is-moving-from-bi-to-ai/)
- [GitLab's Strategies for Building Reliable Conversational Analytics](https://atlan.com/regovern-watch-center/gitlab-conversational-analytics-strategies/)
- [The Hidden Strategies Behind AI-Ready Tech Pioneers (panel)](https://atlan.com/regovern-watch-center/hidden-strategies-behind-ai-ready-tech-pioneers/)
- [All Re:Govern 2025 sessions and highlights: the Re:Govern Watch Center](https://atlan.com/regovern-watch-center/)
- [Book a demo of Atlan](https://atlan.com/forms/talk-to-sales-contact/)