OpenAI CEO Sam Altman may introduce platform for AI agents that work 24/7: What we know

HIGHLIGHTS

OpenAI may bring new AI agents to ChatGPT that can handle tasks like emails and reports on their own.

The leak suggests these tools could run on a schedule and even help inside apps like Slack.

Nothing is confirmed yet, but it shows AI may soon work more actively instead of just replying to users.

OpenAI CEO Sam Altman may introduce platform for AI agents that work 24/7: What we know

OpenAI CEO Sam Altman may be working on a major upgrade for ChatGPT that changes how people use it. A recent social media report says the system may soon include AI agents that can work in the background instead of only answering one question at a time. The information was shared by Tibor Blaho, who often studies OpenAI-related code. The leak talks about a feature called Hermes that might let users create and control small AI assistants. These assistants could take care of tasks like emails, reports, and updates on their own, without needing to be asked again and again. If true, this could make AI more active in daily work. It may change how users complete everyday digital tasks significantly.

Digit.in Survey
✅ Thank you for completing the survey!

The leak points to a feature called Hermes. It may let users build small AI helpers and put them to work without having to guide them every step of the way. These helpers could check emails, write short reports, and pull together updates on their own. Users may also be able to set a schedule, telling an agent when to run and what to do during that time.

Also read: Pixel battery draining overnight? Google confirms issue, fix coming soon

Moreover, the agents may be able to join workplace chats like that on Slack, answer common questions, and find useful information quickly. This can reduce repeated messages and help teams save time.

These agents can also interact with files, applications, and user preferences, which enables them to provide more relevant assistance to individual users.

The leak also mentions a few other tools. One is an image feature that may let users create pictures that look more like themselves using reference photos. Another is an audio tool that could turn long text into short audio clips, like mini podcasts or quick study notes.

Also read: New online gaming rules: Govt proposes dedicated authority to monitor the space

OpenAI has not confirmed any of this. Sam Altman has said nothing publicly about these features, and the company could change or drop them before anything goes live.

Still, the pattern is clear, as the AI tools are moving away from simply waiting to be asked something, and they are being built to take action on their own. That opens up real possibilities for saving time at work, but it also raises fair questions about trust, accuracy, and what role people will play alongside these tools.

Bhaskar Sharma

Bhaskar Sharma

Bhaskar is a senior copy editor at Digit India, where he simplifies complex tech topics across iOS, Android, macOS, Windows, and emerging consumer tech. His work has appeared in iGeeksBlog, GuidingTech, and other publications, and he previously served as an assistant editor at TechBloat and TechReloaded. A B.Tech graduate and full-time tech writer, he is known for clear, practical guides and explainers. View Full Profile

Digit.in
Logo
Digit.in
Logo