Episode Details

This story was originally published on HackerNoon at: https://hackernoon.com/ai-is-guessing-we-stopped-paying-extra-for-better-guesses.
The industry's answer to AI writing bad code is a bigger model. Mine is a deterministic compiler and 97 validators, which is why a build costs $2.
Check more stories related to web3 at: https://hackernoon.com/c/web3. You can also check exclusive content about #decentralize-ai, #llms, #ai-inference, #ai-agent-token-usage, #ai, #ai-assisted-coding, #ai-coding, #verifiable-ai, and more.

This story was written by: @rickardhansson. Learn more about this writer by checking @rickardhansson's about page, and for more stories, please visit hackernoon.com.

The industry's answer to AI writing bad code is to buy more expensive guesses. We shrank the target until a cheap model could hit it, put a compiler behind it that cannot guess at all, and made being wrong free. Builds cost $2. Also, one thing came back broken, and it's in here.