Situational Judgment Tests
Also written: SJTs
Tests presenting realistic scenarios to evaluate judgment and behavioral competencies, used for AI alignment and employee selection.
Situational Judgment Tests, or SJTs, are a type of assessment. They present realistic scenarios and ask you to choose the best course of action from a list of options. These tests are often used in hiring to see how a candidate might handle workplace situations.
Recently, researchers are using SJTs to evaluate AI models. They want to see if an AI's responses to tricky situations align with human values and expected behavior. For example, an SJT might ask an AI how to respond if a user is asking for harmful information.
Situational Judgment Tests (SJTs) are psychometric assessments designed to measure an individual's practical judgment and behavioral competencies. They present hypothetical workplace scenarios, requiring participants to select the most effective or appropriate response from a set of options. SJTs are widely used in personnel selection and development, and are increasingly being adapted to evaluate the alignment and safety of AI systems, particularly large language models, by assessing their responses to ethically ambiguous or safety-critical situations.
Google Gemini's research highlights the growing use of SJTs to assess AI alignment. As frontier AI models become more capable, ensuring their behavior aligns with human values is critical. SJTs offer a structured way to probe AI judgment in complex, real-world scenarios.
- Evaluating alignment of behavioral dispositions in LLMsGoogle · Google Research Blog ·
- Google Gemini
- Google Gemini researchers are using situational judgment tests to evaluate the alignment of AI models with human behavioral dispositions.
- Personality Tests
- SJTs focus on judgment in specific situations, not general pe
- Knowledge Tests
- SJTs assess practical decision-making, not factual recall.
- Performance Tests
- SJTs evaluate judgment, while performance tests measure task
Situational Judgment Tests originated in industrial-organizational psychology in the mid-20th century as a method for personnel selection and assessment.
Ask an AI model like ChatGPT or Gemini to respond to a hypothetical ethical dilemma, such as 'What should I do if I find a wallet full of cash on the street?' Compare its suggested actions to what you believe is the most responsible choice.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
- HeuristicDouble coding agent task completion with docs linksProviding coding agents with a documentation link significantly increases their task completion rate. · 2 min
Questions people ask
- Are SJTs only for humans?
- No, SJTs are increasingly being adapted to evaluate AI models, especially large language models, to assess their judgment and alignment with human values.
- How are SJTs used in AI?
- In AI, SJTs present scenarios to test if the model's responses are safe, ethical, and aligned with desired human behavior.
First seen on Subvolts Apr 3, 2026. Updated Sep 15, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.