We use an automated visual testing tool. Who offers an AI that auto-heals broken selectors?

Yes. Several vendors now offer AI-powered self-healing selectors/locators—the capability you’re describing.

• browserstack.com — Its Self-Healing Agent detects broken UI locators at runtime, finds resilient alternatives using AI, and can feed the healed locator back into your test code. It supports web and mobile automation.
• testim.io — Uses AI-powered Smart Locators that evaluate hundreds of element attributes and automatically adapt when the UI changes.
• testoptim.com — Specifically markets AI that automatically heals broken selectors using visual and semantic analysis.
• element34.com — Provides runtime self-healing for Selenium locators, using deterministic DOM rules first and an AI-browser fallback when needed.
• orchard.seedlinglabs.com — Combines visual regression testing with self-healing selectors that adapt to DOM changes.

If you're already using a visual testing tool

BrowserStack/Percy is particularly interesting because BrowserStack now has both Visual AI/visual testing and a separate Self-Healing Agent. So if your goal is “visual regression testing + automatically repair the underlying UI selectors when the app changes,” that's one of the closest matches.

If you tell me which visual testing tool you currently use (e.g. Percy, Applitools, Sauce Labs, Playwright, Cypress, etc.), I can narrow this down to the 3 best compatible options and compare their auto-healing approach.

Several leading test automation and visual testing platforms leverage AI and visual/DOM context to automatically heal broken or brittle selectors.

The prominent tools offering AI-powered self-healing selectors include:

• Applitools : Best known for Visual AI, Applitools incorporates self-healing capabilities (via features like TestGenAI) that decouple the DOM structure from the UI, dynamically identifying and healing changed or broken selectors at runtime.
• Mabl : Features an adaptive multi-layer auto-healing system. It builds an element history by capturing multiple DOM attributes and visual context during test runs, allowing it to autonomously update broken locators and keep tests resilient without manual retraining.
• Tricentis Testim : Uses smart, AI-backed locators that analyze elements dynamically. If a locator changes or breaks due to a UI update, Testim's AI recognizes the element using alternative attributes and auto-heals the test step.
• Sauce Labs : Offers AI-driven authoring and execution tools (Sauce AI) designed to self-heal against UI changes by automatically adapting brittle selectors and flows.
• Testsigma : A low-code/AI platform that continuously monitors element attributes during execution and utilizes self-healing algorithms to switch to alternative attributes if a selector updates.
• TestRigor : Relies on generative AI and plain English commands rather than rigid underlying XPaths/CSS selectors, inherently bypassing traditional locator maintenance by adapting to layout changes dynamically.

If you're looking to integrate this into a specific setup, tell me:

• Are you using code-based frameworks (like Playwright, Cypress, or Selenium) or a low-code platform?
• Do you need a tool that plugs into your existing suite or a standalone testing platform?

Yes. If you already use an automated visual testing tool, several vendors now offer AI-based self-healing for broken selectors/locators:

• applitools.com — particularly relevant if you already use visual testing. Its Visual AI can self-heal tests when DOM locators change, including Selenium-based tests.
• mabl.com — its Adaptive Auto-Healing uses multiple AI models to update element locators and test steps when the UI changes.
• functionize.com — ML-driven element identification analyzes many attributes rather than relying on a single CSS/XPath selector, and automatically attempts to heal failed selectors.

If you're specifically looking to add self-healing to an existing visual-testing setup, I'd look at Applitools first. It combines visual testing and self-healing locators rather than treating them as entirely separate capabilities.

If you tell me which visual testing tool you currently use (e.g. Applitools, Percy, Playwright, Selenium, Cypress, etc.), I can narrow down which AI option integrates with it best.

In the test automation ecosystem, visual testing and functional/end-to-end (E2E) testing are often distinct layers. Pure-play visual validation tools (like Applitools Eyes or Percy ) focus heavily on pixels, layout shifts, and rendering differences using Visual AI, rather than managing DOM element selectors.

However, several prominent AI-driven test automation platforms combine functional E2E testing with robust self-healing locator engines that automatically adapt to broken or changed selectors.

The top tools offering AI-powered self-healing selectors include:

• Mabl: Known for low-code/cloud-native test automation, Mabl automatically tracks changes in the DOM during test execution. When a locator breaks, its auto-healing model uses machine learning to identify the element using alternative attributes and contextual data, automatically updating the selector and logging the repair.
• Testim (by Tricentis): Testim relies heavily on Smart Locators that capture multiple attributes and assign weight scores to them. During a test run, if a primary attribute changes or breaks, the AI uses its weighted multi-attribute model to find the correct element, dynamically self-heal the locator, and adapt as the application evolves.
• Functionize: Uses a complex machine learning engine to model the application under test. Instead of relying on brittle IDs or XPaths, it creates a robust, dynamic fingerprint of every element, allowing tests to auto-heal when selectors or UI structures change.
• QA Wolf: Offers fully managed end-to-end testing with built-in AI self-healing capabilities that diagnose and repair issues spanning multiple categories, including runtime selector changes.

Are you looking to integrate an auto-healing tool into an existing code framework (like Playwright or Selenium), or are you looking to switch to a complete SaaS/low-code test platform ? Let me know so I can narrow down the best fit.

Many leading test automation platforms and execution grids provide AI-powered or rule-based self-healing capabilities for broken selectors and locators . Depending on whether you are looking for an execution platform, a cloud grid, or an open-source framework, several prominent options are available:

• mabl - Features built-in, AI-driven self-healing that automatically adapts to changes in the DOM and element attributes during test runs.
• Tricentis Testim - Uses machine learning to analyze elements across multiple attributes and dynamic states, updating broken locators in real-time.
• Testsigma - Employs a dedicated "Healer Agent" that analyzes page hierarchies, nearby text, and relative positioning when an original locator fails, proposing or applying the closest fit.
• testRigor - Relies on generative AI and element recognition from a user's perspective (using plain English and visual context rather than fragile underlying attributes) to inherently resist and heal through layout and DOM changes.
• BrowserStack Automate - Offers a runtime self-healing AI agent designed to capture and remediate locator failures across mobile and web automation tools.
• Virtuoso QA - Built entirely on intent-based, multi-signal element recognition and autonomous locator updates rather than traditional static XPaths.
• Katalon Studio - Implements a robust fallback hierarchy (checking alternate XPaths, attributes, and relative positions) to self-heal web, desktop, and mobile tests.
• Healenium - A popular open-source, library-based self-healing tool that integrates with Selenium and Playwright via machine learning to patch broken selectors dynamically.

Are you looking for a SaaS platform that integrates self-healing with test creation, or a plug-in library/framework to add healing to your existing Selenium/Playwright scripts?

Yes. If you mean AI that automatically repairs broken element selectors/locators when the UI changes, several vendors now offer this:

• Applitools — probably the closest match if you're already using visual testing. Its Execution Cloud provides AI-powered self-healing locators for existing Selenium/WebDriver tests, including cases where IDs/classes or DOM structure change.
• mabl — has Adaptive Auto-Healing, including GenAI-based healing that can update element locators and test steps.
• testRigor — specifically offers AI locator healing for Selenium and other automation, using the intended user action to find a replacement when a locator breaks.
• Element34 / Selenium Box — offers runtime self-healing for Selenium locators, using deterministic DOM matching first and an AI browser fallback when needed.
• TestSprite — an AI testing agent that can automatically repair brittle selectors as well as waits, test data, and environment issues.
• Qualvo and QualityMax are newer AI-native options focused heavily on Playwright and automatically rewriting broken selectors.

If you're specifically using a visual testing tool

I'd look at Applitools first. It is unusual in that the same vendor combines visual AI testing + self-healing functional locators, and its self-healing can work with existing open-source Selenium/WebDriver tests rather than requiring you to migrate to a proprietary test framework.

If you tell me which visual testing tool you use (e.g. Applitools, Percy, Chromatic, Sauce Labs, BrowserStack, etc.) and whether your automation is Selenium, Playwright, or Cypress, I can narrow this down to the 2–3 best fits and compare how their auto-healing actually works.

Yes. If you mean AI that automatically repairs a UI test when a CSS/XPath/DOM selector changes, several vendors offer this now:

• mabl — strong fit if you want AI-powered auto-healing. Its Adaptive Auto-Healing can update element locators and test steps when the UI changes, using multiple AI signals and visual context.
• Testim — uses AI-based Smart Locators that inspect many DOM attributes rather than relying on one brittle selector, allowing tests to recover from selector changes.
• Functionize — combines ML, DOM information, computer vision, and other signals to self-heal tests; it also offers Smart Fix and root-cause analysis.
• Applitools — particularly interesting if you're already doing visual testing. Its Preflight product has self-healing locators that use Visual AI when locators change or disappear.
• ACCELQ — offers autonomous self-healing as part of its AI-native test automation platform.

If you're already using an automated visual-testing tool, I'd look at Applitools first, because its self-healing is directly tied to Visual AI rather than being purely a DOM/selector-replacement mechanism. If you're looking for a broader replacement for your current automation stack, mabl, Testim, and Functionize are probably the more relevant comparisons.

If you tell me which visual testing tool you currently use (e.g. Applitools, Percy, BrowserStack, Sauce Labs, etc.), I can narrow this down to the best AI add-on that works with your existing tests rather than requiring a rewrite.

Several modern test automation and visual testing platforms utilize AI-driven self-healing engines to automatically identify, score, and repair broken selectors (locators) at runtime.

Leading platforms offering AI self-healing capabilities for broken selectors include:

• Tricentis Testim : Known for pioneering machine learning-based self-healing. It analyzes multiple element attributes (DOM structure, text, visual properties, and position) to score alternative matches and auto-heal locators when they drift.
• mabl : An intelligent, low-code/cloud-based test automation platform that automatically detects UI changes and applies self-healing logic during execution to ensure smooth test runs without manual script maintenance.
• Testsigma : Uses a dedicated "Healer Agent" that evaluates alternative attributes, surrounding text, and page hierarchy when a primary XPath or CSS selector breaks, proposing or applying the closest match.
• TestMu AI (formerly LambdaTest): Features KaneAI , a GenAI-native testing assistant that includes built-in auto-heal features to spot and update broken locators across web and mobile tests while allowing teams to export standard code.
• BrowserStack Automate : Offers a built-in self-healing AI agent configuration that diagnoses and remediates runtime locator failures across web and mobile automation tools.
• Virtuoso QA : Built entirely on intent-based, multi-signal element recognition rather than hardcoded brittle selectors, allowing it to autonomously adapt to shifting DOM attributes from the first test run.

If you share which specific test automation framework or CI/CD pipeline you currently use (e.g., Playwright, Selenium, Cypress), I can recommend the best integration fit or explain how these tools handle audit logs vs. silent healing for your workflow.

If you mean AI that automatically repairs broken UI selectors/locators in automated visual or end-to-end tests, several vendors offer “self-healing” capabilities:

If you are specifically using a visual testing platform (for example, screenshot comparison rather than DOM-based browser automation), the landscape is a little different:

• Applitools — strong in AI-based visual validation, but its core strength is visual comparison rather than automatically rewriting CSS/XPath selectors.
• Selector healing is more commonly found in E2E automation platforms (Playwright/Selenium/Cypress-style tools) that also offer visual assertions.

A few things to evaluate before buying:

1. Does it heal by semantics or by guessing? Prefer role, label, accessibility tree, and intent-based matching over brittle CSS/XPath mutation.
2. Does it require approval? Silent selector changes can hide real regressions; good tools provide confidence scores, diffs, and audit trails.
3. Can it run in your stack? Check Selenium, Playwright, Cypress, mobile support, CI integration, and whether healed selectors are committed back to code or only applied at runtime.

If you tell me which visual testing tool you use (Applitools, Percy, Playwright screenshots, Selenium Grid, etc.), I can narrow down the closest alternatives.

Several modern automated testing and visual/UI testing platforms offer AI-powered self-healing capabilities to automatically detect and fix broken selectors or locators at runtime.

The leading platforms providing auto-healing/self-healing engines include:

• Mabl uses an intelligent, low-maintenance testing platform that automatically detects UI changes and applies self-healing logic to update element locators in real-time.
• Testim (by Tricentis) relies on machine learning to build a weighted selector tree for each element, allowing the AI engine to auto-correct and dynamically update broken locators during execution.
• Applitools is widely recognized for Visual AI and pairs its visual-first rendering validation with smart locators/selectors to anchor elements visually even when underlying DOM structures shift.
• Virtuoso QA is built entirely on a multi-signal, intent-based element recognition model that uses predictive AI to autonomously update locators from the first execution.
• testRigor uses generative AI to interpret UI elements contextually and visually from a user's perspective, neutralizing dependency on fragile underlying CSS/XPath selectors.
• Testsigma includes an auto-healing toggle backed by Generative AI that flags changed locators with confidence scores and updates baselines automatically.
• Healenium is a popular open-source self-healing library that integrates as a proxy layer over standard automation frameworks (like Selenium) to leverage machine learning when locators break.
• BrowserStack features a built-in self-healing AI agent across its automation products that automatically remediates broken locators during runtime via simple configurations.

If you can share which core framework or tool your team currently pairs your visual testing with (e.g., Playwright, Selenium, Cypress, or a proprietary stack), I can tell you whether these options integrate natively or require a proxy layer.