OpenAI “o” Assistant: What Is the New Always-On AI?

A new OpenAI product appears to be emerging under a very short name:

o.

Recent reports have identified references to “o, your always-on assistant” in ChatGPT interface code, suggesting that OpenAI may be preparing an always-on assistant experience. The company has not publicly announced the product, so the details should still be treated as preliminary.

The phrase “always-on assistant” is more interesting than the name itself.

It suggests a shift from AI that waits for prompts to AI that is available continuously.

That could mean an assistant capable of handling tasks across a user's digital life.

For example:

Reading information.

Managing reminders.

Working with files.

Handling communications.

Monitoring ongoing tasks.

Connecting different applications.

That is very different from traditional chatbot usage.

Today, users generally open an AI tool when they want something.

An always-on assistant reverses the relationship.

The assistant becomes part of the user's environment.

That also creates significant privacy questions.

An AI that is continuously available potentially has access to far more information than a chatbot.

Which conversations can it access?

What data can it remember?

Can it send emails?

Can it act without confirmation?

What happens when it makes a mistake?

Those questions will matter more than the branding.

If OpenAI does launch an always-on assistant, the product could represent another step toward agentic computing.

The interface may become less about asking AI questions and more about delegating tasks.

For now, the available evidence points to an upcoming product rather than a fully announced release.

But the direction is clear.

The next generation of AI assistants may not simply answer when called.

They may stay with you throughout the day.