Found at “Computer-Using Agent | OpenAI” on 2025-01-24 10:59:49 +01:00.

OpenAI’s announcement of their Computer-Using Agent (CUA) mentions APIs a few times. However, it shows that not needing to use APIs is a benefit and something to aim for. The first mention states that there’s an increased flexibility in bypassing any APIs and instead using Web apps directly.

This gives it the flexibility to perform digital tasks without using OS- or web-specific APIs.

The second mention of APIs shows examples of how flexible the CUA is because it doesn’t need to interact with APIs.

CUA (…) uses a virtual mouse and keyboard to complete actions. This enables CUA to act in a wide range of digital environments, performing tasks like filling out forms and navigating websites without needing specialized APIs.

The third mention of APIs hints at how the CUA can grow in usage over time. And, not using APIs is what makes that growth possible.

By moving beyond specialized agent-friendly APIs, CUA can adapt to whatever computer environment is available

Finally—and ironically—the announcement mentions that the CUA itself will be available through OpenAI’s own API.

We’re also working to make CUA available in the API, so developers can use it to build their own computer-using agents

Something else I also found interesting was their video announcement where Reiichiro Nakano reinforces the fact that not needing an API is an added benefit.

(…) before, if you wanted to build something like operator without CUA you’d need to use some specialized APIs. For example if you wanted your model to buy stuff from Instacart you’d need to figure out if Instacart had an API. You’d need to figure out if that API had all the functions that it needed and you need to give your model the specs of that API. But if your site, like most other websites did not have an API then you’re out of luck.