planning and reasoning
Capabilities of agentic systems that enable iterative problem-solving: the ability to break down problems, assess state, decide on actions, and loop based on changing conditions
As defined in How Domain-Specific AI Agents (DXA) Will Shape the Industrial World in the Next 10 Years
Where it comes up · 2
- How Domain-Specific AI Agents (DXA) Will Shape the Industrial World in the Next 10 Years4 claims
- Pat Grady - Relentless Application of Force - [Invest Like the Best, EP.378]2 claims
Most important claims · 6 of 6Search all claims for “planning and reasoning” →
- Agentic AI systems (models with planning and reasoning capabilities) are necessary because models alone lack the ability to iterate, loop back, and reconsider problems—agentic systems add a reasoning loop that allows for multi-step problem solving and adaptation as the environment changes.How Domain-Specific AI Agents (DXA) Will Shape the Industrial World in the Next 10 Years
- If frontier model capabilities froze at today's level, optimization work alone to make models cheaper, faster, and easier would revolutionize almost every industry, because current capabilities are powerful and have only begun to be harnessed.Pat Grady - Relentless Application of Force - [Invest Like the Best, EP.378]
- The illusion that systems like ChatGPT have planning and reasoning is created by external planning frameworks wrapped around the model—the models themselves lack these capabilities, which are added by the system architecture.How Domain-Specific AI Agents (DXA) Will Shape the Industrial World in the Next 10 Years
- The planning and reasoning code written at high-level abstraction (not at the neuron level of models) should be viewed as part of the knowledge paradigm, not as a crutch; higher-level abstractions are appropriate for representing structured knowledge.How Domain-Specific AI Agents (DXA) Will Shape the Industrial World in the Next 10 Years
- If returns to scale are diminishing, the AI industry could pivot to a small team of geniuses designing new architectures, shifting compute from training toward inference for planning and reasoning, and aggressive optimization, creating a cash-generative business with a dominant position but with the risk of not having reached diminishing returns.Pat Grady - Relentless Application of Force - [Invest Like the Best, EP.378]
- Open SSA (Small Specialist Agents) is a project that implements planning and reasoning for agentic AI and is available open-source for download and use.How Domain-Specific AI Agents (DXA) Will Shape the Industrial World in the Next 10 Years