The Triadic Cognitive Architecture: Bounding Autonomous Action via Spatio-Temporal and Epistemic Friction

1. 1Identify the limitations of current autonomous AI agents driven by Large Language Models (LLMs)
2. 2Introduce spatio-temporal friction to provide an intrinsic sense of network topology and temporal pacing
3. 3Implement epistemic friction to establish epistemic limits and prevent excessive tool use or prolonged deliberation
4. 4Evaluate the effectiveness of the Triadic Cognitive Architecture in interactive environments

AI researchers and engineers working on autonomous agents can benefit from this framework to improve the robustness and efficiency of their systems, while product managers can utilize this knowledge to develop more effective AI-powered products

💡 The Triadic Cognitive Architecture can mitigate failure modes in autonomous AI agents by providing a more robust and efficient framework for decision-making

💡 Introducing spatio-temporal & epistemic friction to bound autonomous action in AI agents