Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is a fake commit since there's not a great way/process to change or discuss proposals pre-formally.
Have a few suggestions and thoughts
The above would then allow me to easily insert nodes where ai systems can call functions that interact with the real world, eg. google search, or deploying a docker container, or hosting a website.
in the spec's current state its very limiting and opinionated with a UI vs having a well defined interface. I also think API interactions are likely to be the dominant interaction for ai systems in the long run and should be first class
there should be a component on auto-evaluation. eg. using auto-evals lets you call smart contracts accurately, and provides verification of input parameters, which is then able to be provided into the payments system as proof of evaluation. without auto-evals I don't think it could be viable for production ai
it should let users add whatever models they want, even closed source, if we're building a truly open marketplace