OpenAI has published new research examining how AI agents are changing the way people work, with a growing shift from conversational chatbots to autonomous systems that execute complex, multi-step tasks.
Drawing on internal Codex usage, the report shows increasing adoption across engineering, legal, finance, marketing, and operations teams, with non-technical employees rapidly expanding their use of AI agents.
Rather than simply answering questions, these agents plan, execute, and iterate on work while keeping humans in supervisory roles. OpenAI says this transition marks a broader move toward agentic workflows that improve productivity, reduce manual effort, and reshape knowledge work across organizations.
