The study investigates the application of artificial intelligence (AI) for adaptive testing aimed at assessing the knowledge of future engineers. The methodology for determining test difficulty parameters based on Bloom’s Taxonomy is examined. A formula is proposed for calculating the overall task difficulty, considering the volume of information, degree of knowledge integration, time for completion, and level of creativity. The weighting coefficients for these parameters can be adjusted to emphasize the importance of specific evaluation aspects. An example of creating an adaptive test in Google Forms using AI (ChatGPT, Gemini, or Copilot) is provided. The stages of test creation are described, including generating questions with multiple answer options and configuring section transitions based on responses. To determine the current state of AI usage in evaluating the learning outcomes of future engineers, a survey was conducted among engineering students and educators. The role of adaptive testing using AI (ChatGPT, Gemini, or Copilot) and the tool for creating online surveys and questionnaires (Google Forms) is identified as an effective means for assessing the knowledge of future engineers.