Building trust in agentic tools: What we learned from our users

Source: GitLab Blog

Building Trust in Agentic Tools

As AI agents become increasingly sophisticated partners in software development, a critical question emerges: How do we build lasting trust between humans and these autonomous systems? Recent research from GitLab’s UX Research team reveals that trust in AI agents isn’t built through dramatic breakthroughs or flashy features alone. Instead, it hinges on consistent, transparent, and user‑centric experiences that align with developers’ existing workflows and expectations.

The study highlights several key factors that influence trust, including:

• Reliability
• Explainability
• Control
• The ability to recover gracefully from errors

By understanding these dimensions, product teams can design AI‑driven tools that not only boost productivity but also foster a collaborative relationship where developers feel confident delegating tasks to their intelligent assistants.