OpenAI is building AI agents for everything. Will everyone use them?

OpenAI is building AI agents that can autonomously complete multistep work tasks across various professional fields, not just software engineering. The company released a product available on its lowest subscription tier that gives workers AI tools capable of accessing and controlling their digital workflows, from email to collaboration apps. Adoption has been slow outside the company itself, with internal usage vastly outpacing external adoption, presenting a challenge for OpenAI to make these tools intuitive enough for mainstream users while maintaining the advanced functionality that power users need. The commercial stakes are significant because agents that work longer burn through more computational tokens, making them more profitable, and expanding AI agent usage beyond software engineering is crucial for justifying the industry's massive investments in training and computation.