New Microsoft tool lets devs spin up AI behavior tests using text descriptions Microsoft has released ASSERT, an open-source framework designed to simplify the evaluation of application-specific AI behavior. ASSERT converts high-level, natural-language descriptions of intended behaviors into structured tests that are run against the AI system, providing scored results and detailed logs for inspection. This framework aims to fill the gap left by general AI evaluations, enabling developers to create trustworthy AI systems tailored to their product’s context and policies.

• Microsoft launched ASSERT (Adaptive Spec-driven Scoring for Evaluation and Regression Testing), an open-source framework.
• ASSERT simplifies AI behavior testing for specific products and services.
• The framework turns natural-language descriptions of goals and policies into scored tests.
• It generates problem scenarios and test cases, runs them, and scores the results, recording AI system paths.
• Developers can customize evaluations with system context, tools, and constraints.
• ASSERT can be used during development, after deployment, and for continuous monitoring.
• This release aligns with a broader industry shift towards repeatable testing and regression checks in AI. Continue reading https://techcrunch.com/2026/06/02/new-microsoft-tool-lets-devs-spin-up-ai-behavior-tests-using-text-descriptions/