Stanford + Meta just dropped the paper that flips everything about AI agents.
It's called "Code as Agent Harness."
Right now, we treat large language models as text generators. When they need to solve a complex problem, they rely on a "chain of thought."
But natural language is slippery. It's vague. It loses context. When an agent hallucinates in English, it just keeps talking.
So they introduced a framework that changes the entire architecture of autonomy: "Code as Agent Harness."
They stopped asking the AI to reason in words, and forced it to reason in code.
Code isn't just the final output anymore. It is the memory. It is the environment. It is the boundary.
Instead of writing a paragraph about how to solve a problem, the agent writes a script, executes it, and reads the output.
Tests become its senses. Execution logs become its memory. Sandboxes become its physics.
If an agent makes a mistake in English, it apologizes and hallucinates again.
If an agent makes a mistake in code, the compiler throws an error. The trace tells it exactly what broke. The system forces it to fix it.
This is where prompt engineering dies, and systems engineering takes over.
The paper proves that reliability doesn't come from a smarter base model. It comes from the "harness" wrapped around it:
- The model proposes.
- The harness executes.
- The environment returns feedback.
- The verifier checks.
When the Industrial Revolution made mass production possible, the value of art shifted to who made it, why, and the life they lived. AI is the same—it cannot generate history. True art will stay in demand, while mass-produced garbage will just flood the market.
However, even AI art can hold value if it reflects the creator's uniqueness or the motivation behind it.