Février 2026

12 février 2026

Nouveautés

Autopilot

Autopilot is a collection of AI-powered digital systems, also known as agents, designed to boost the productivity of testers throughout the entire testing lifecycle. These capabilities are integrated into UiPath Studio Desktop and UiPath Test Manager.

Important :

Autopilot is available in Test Manager, only when delivered through Test Cloud.

Autopilot propose des fonctionnalités qui peuvent être regroupées dans les catégories suivantes :

• Agentic test design: AutopilotTMAutopilotTM in Test ManagerTest Manager supports you to evaluate requirements for quality aspects such as clarity, completeness, and consistency. Autopilot also helps you generate manual test cases for requirements (such as user stories) and SAP transactions.
• Agentic test management: AutopilotTMAutopilotTM in Test ManagerTest Manager allows you to get actionable insights into test results through reports.

For an overview of the Autopilot capabilities, visit the Autopilot overview guide.

Consultez la page Licences Autopilot pour obtenir des informations sur la façon dont les activités Autopilot sont mesurées et attribuées.

Les fonctionnalités suivantes sont actuellement disponibles :

• Générer des cas de test
• Importer des cas de test
• Évaluer la qualité des prérequis
• Générer des rapports
• Rechercher des cas de test obsolètes

Supported LLMs:

• Anthropic.Claude-4-5-sonnet
• gpt-4o-2024-11-20
• Gemini-2.5-pro
• Gemini-2.5-forg

User access management with Autopilot:

• Pour travailler avec les fonctionnalités d’Autopilot dans votre organisation, vous devez d’abord apporter votre propre abonnement LLM.
• Before configuring the AI Trust Layer policy, go to the LLM Configuration tab in the AI Trust Layer of your organization, and set up your preferred model or subscription. This step is required to enable AI-powered features, including Autopilot, within your organization.

The AI Trust Layer governance policy allows you to manage the use of AI-powered features within your organization. Although all members have default access to these features, you can use this policy to restrict access as needed. The AI Trust Layer governance policy empowers you to limit a user's access to certain AI-powered features or all of them, at a user, group, or tenant level. Additionally, it gives you the ability to decide which AI products users can access. You can create, modify, and implement this governance policy in Automation Ops. If you want to deploy an AI Trust Layer governance policy and still use the AI-powered testing capabilities, ensure that, within the policy's Features Toggle, you select Yes for Enabling Test Manager features. To create, configure, and deploy a governance policy for your organization, do the following:

• Créer une politique de gouvernance
• Configure the settings for AI Trust Layer policies
• Déployer les politiques de gouvernance

Modifications à venir

Upcoming support for Claude Sonnet 4.5

To ensure access to its AI-enabled features, Test Manager is currently working on enabling support for Claude Sonnet 4.5 in view of the upcoming retirement of the Claude Sonnet 3.7 model. We will announce a set of recommendations to our customers in the upcoming weeks.