OpenAI quietly upgraded ChatGPT Work on August 25 with two practical capabilities that move it closer to a real always-on operator for small teams and solo founders. Scheduled tasks can now be triggered by webhooks from Gmail, Slack, and GitHub instead of only running on a fixed timer. At the same time, the browser agent can complete multi-step work on websites that require a logged-in session, provided the user confirms sensitive actions.
Previously, Work could schedule recurring jobs, but it still needed you to check in or manually kick off the next step when something changed in your inbox or channels. Now you can tell it, for example, “When a new email from a client arrives in Gmail, summarize it, draft a reply in my tone, and flag anything that needs a decision.” Or “When a pull request is updated in GitHub, review the changes and post a short status update in the relevant Slack channel.” Free users gained limited sharing of these tasks; Plus and Pro users get the full webhook and sign-in browser features.
For SMB owners and solopreneurs the difference is immediate.
Many of us already live inside Gmail, Slack, and a handful of web tools. Being able to hand those recurring loops to ChatGPT Work removes the constant context-switching tax that eats hours every week. A solo consultant can let the agent triage inbound leads overnight. A small agency can keep project status current without someone copying updates between tools. The confirmation step on signed-in sites keeps risk manageable while still allowing the agent to move work forward.