OpenAI Workspace Agents in ChatGPT — Team Automation Layer
OpenAI launched workspace agents in ChatGPT on April 22, 2026. Shared, Codex-powered agents that teams can create once and use together — running 24/7 in the cloud, accessible via ChatGPT or Slack. Available in research preview for ChatGPT Business, Enterprise, Edu, and Teachers plans. Free until May 6, 2026, then credit-based pricing. Described as the evolution from Custom GPTs (single-user) to team automation layer (shared, persistent, multi-workflow). Launched alongside Codex enterprise partnerships with Accenture, PwC, Infosys, and Cognizant.
This is the clearest signal yet that OpenAI's strategic direction is "team automation platform," not "powerful chat interface." For GTM operators: workspace agents that can prepare reports, write code, respond to messages, and run multi-step workflows autonomously represent a real shift in what "your AI stack" does at the team level. The consulting giant partnerships (Accenture, PwC, Infosys, Cognizant) follow the traditional enterprise software playbook: capability → system integrators → Fortune 500 rollout. This means enterprise AI adoption is entering the institutionalization phase — it's no longer about early adopters; it's about deployment at scale through trusted channels. For Seva's buyer-operator audience: the question shifts from "does my team use AI?" to "do we have agents that handle our recurring workflows autonomously?" Teams without shared workspace agents will be doing manually what competitors automate.
- "OpenAI shipped workspace agents in ChatGPT this week. Build once, team uses together, runs in the cloud. It's a subtle shift: chatbot your team chats with → agents your team runs. The dashboard era is ending."
- "OpenAI partnered with Accenture, PwC, Infosys, Cognizant to deploy Codex at enterprise scale. That's the enterprise software playbook: capability → SI → Fortune 500. The AI tools era has entered the rollout phase."
- "ChatGPT workspace agents are free until May 6. After that: credit-based. The preview-to-paid pattern is now predictable. This is also when usage data shows who actually builds workflows vs. who just explores."