Blog

Notes on what comes next.

What we're learning about models, agents and the cost of thinking.

October 12, 2026Consequence models: AI that knows what happens nextAgents can write the code. What they can't do is predict what running it will do. A model that answers one question, "will this action work?", could be what lets agents run without someone watching.Read the post· 6 min

Agent view · paste this into your agent

Open in Claude Codeif installed Open in Claude Desktopif installed Open in Codexif installed

Opens your agent with the prompt typed. You press Enter.

Deep links into apps you have installed. Claude and Claude Code are trademarks of Anthropic; Codex is a trademark of OpenAI. ScaledThought is not affiliated with or endorsed by either.

Agents can read the whole site as markdown at scaledthought.com/llms.txt.