UiPath Test Cloud Adds Autonomous Execution and Playwright Support in October 2026

UiPath is moving Test Cloud toward more autonomous testing. In its September 23, 2026 announcement, the company described capabilities arriving around October: AI-powered execution of manual tests, agent-led application exploration, Playwright integration, and automation for routine test-run tasks. Several are now available or entering public preview. This is a broader release than another round of AI-assisted test generation.

The model is straightforward. A tester supplies a manual test, a goal, or an existing automation, and UiPath's agents take on more of the execution and maintenance. Testers still review findings, resolve ambiguity, and decide whether to release. That makes agentic testing an operating model, not a substitute for quality-engineering judgment. ([uipath.com](https://www.uipath.com/newsroom/uipath-enhances-test-cloud?utm_source=openai))

What is shipping in October

The clearest change is Autonomous Test Execution. In Test Manager, an existing manual test case can run automatically with an AI model and an execution target; teams do not need to convert it first into traditional scripted automation. UiPath's documentation says the feature uses Autopilot and requires Test Cloud delivery, a suitable Orchestrator runtime, and related permissions. It is not available in Test Manager delivered through the standard Automation Cloud service. ([docs.uipath.com](https://docs.uipath.com/test-manager/automation-cloud/latest/user-guide/running-autonomous-test-execution?utm_source=openai))

UiPath is also introducing Autonomous Exploration. The tester sets an objective, and an AI agent explores the application, investigates potential issues, and reports its findings for review. That is exploratory testing, not a replay of a fixed regression script. Its usefulness will turn on how well teams set goals, limit environments, and separate meaningful discoveries from noise. ([uipath.com](https://www.uipath.com/newsroom/uipath-enhances-test-cloud?utm_source=openai))

For teams already running coded web automation, the planned Playwright integration could be the most practical part of the release. UiPath says Playwright tests created outside the platform will run through Test Manager and connect with broader test planning and orchestration. Teams therefore will not need to replace an existing Playwright suite with a proprietary authoring model. The integration was announced for public preview in October 2026. ([uipath.com](https://www.uipath.com/newsroom/uipath-enhances-test-cloud?utm_source=openai))

Pricing is enterprise-oriented

UiPath lists Test Cloud Standard and Enterprise as "Contact Sales" offerings instead of publishing a simple per-user subscription price. Standard includes test management, low-code and coded automation, built-in AI agents, governance controls, Bring Your Own Model support, and one hosting region. Enterprise adds multi-region or on-premises deployment, customer-managed encryption keys, premium connectors, and included Screenplay functionality. ([uipath.com](https://www.uipath.com/pricing?utm_source=openai))

The add-on prices show more of the commercial model. Performance Testing is listed at $20,000 for 2,500 virtual users, Screenplay costs $10,000 for 50,000 runs on Standard, and Self-Healing costs $5,000 for 5,000 heals on Standard. UiPath says Enterprise includes Screenplay, while Performance Testing remains an additional purchase. Those prices put Test Cloud squarely in the enterprise testing market, rather than the budget Playwright-dashboard category. ([uipath.com](https://www.uipath.com/pricing?utm_source=openai))

What QA leaders should evaluate

• Evidence quality: Require the agent to retain screenshots, logs, test steps, environment details, and a clear explanation for every failure classification.
• Human approval points: Set rules for actions the agent can take on its own and those requiring tester confirmation, especially destructive data changes and production-like environments.
• Coverage measurement: Compare autonomous exploration with existing risk-based testing, mutation testing, and production defect data. More activity does not equal better coverage.
• Model governance: Record which model handled each test, what data it accessed, and how changes to prompts, policies, or models may affect repeatability.
• Migration cost: Playwright teams should check how results, traces, retries, fixtures, secrets, and CI pipeline status will map into Test Manager before expanding the rollout.

The larger shift

UiPath's update points to a broader change in test automation. The competition is no longer limited to script generation. Vendors want to manage the entire loop: understand intended behavior, run tests, search for risk, interpret results, maintain automations, and send evidence to the people responsible for release quality.

That claim needs operational evidence, not a product demonstration. Teams assessing Test Cloud should start with a narrow, representative workflow and track escaped defects, investigation time, false positives, maintenance effort, and the share of agent decisions needing human correction. If it improves those measures without sacrificing traceability, UiPath's October release could make autonomous testing useful to enterprise delivery teams instead of leaving it as an AI-roadmap headline.