Psychology of AI 2: AI Accuracy, Bias, and Decision Psychology
About This Course
This two-day standalone course equips participants to evaluate Artificial Intelligence (AI)-generated outputs, language models and AI-assisted decisions from both technical and psychological perspectives. Participants will examine accuracy, reliability, hallucination, bias and consistency, while also exploring how human judgement can be distorted by fluency, overconfidence, anchoring, confirmation bias, automation bias and AI sycophancy. Through hands-on exercises, they will examine model performance under different conditions. The course also covers the design of lightweight AI assurance protocols and methods for communicating AI quality and risk findings to diverse stakeholders.
What You'll Learn
- Explain key evaluation concepts such as precision, recall, bias metrics and hallucination rate.
- Apply critical thinking and evidence-based reasoning to interpret Artificial Intelligence (AI) performance results.
- Conduct hands-on testing of language models using Capabara and AI tools.
- Analyse and compare outputs from different AI systems to detect systematic errors and biases.
- Design a lightweight AI evaluation or assurance audit for ongoing organisational use.
Entry Requirements
Participants should preferably hold a degree or diploma and have at least three years of relevant work experience.