AI Experiments
Evaluating reasoning, emotional comprehension, and perception capabilities of Frontier AI models through advanced and complex prompting approaches.
Featured
March 6, 2025
Can Frontier AI Models Solve Concept Mapping Problems?
AI models show strong STEM reasoning but struggle with abstract concepts. Our latest insights reveal key gaps in AI understanding. Discover more!
February 27, 2025
Capabilities Test #5: Associative Reasoning
AI models struggle with similarity judgments. See how context boosts accuracy, the trade-offs of reasoning time & how to optimize AI for better decisions.
February 20, 2025
Capabilities Test #4: Strategic Ideation Under Uncertainty
Explore how AI models navigate uncertainty in high-stakes scenarios. Discover insights into risk perception and strategy. Read the full test results now!
February 13, 2025
Capabilities Test #3: Complex Problem-Solving
We tested four AI models on complex problem-solving with social dynamics. Check what we found about their reasoning, decision-making, and adaptability!
February 6, 2025
Capabilities Test #2: Emotion Classification Task
Can AI classify human emotions in complex text? We tested five models on an emotion classification task. See how they responded and what it implies!
January 30, 2025
Capabilities Test #1: Heinz Dilemma Variations with OpenAI's o1 Model
Explore how OpenAI’s o1 adapts moral reasoning in varied ethical dilemmas. Read our experiment and uncover AI’s evolving moral choices. Start now!