Editorial illustration for OpenAI Targets Production AI Agents for Customer Service
OpenAI Presence: AI Agents for Customer Service
OpenAI has a new pitch for enterprise customers, and it's aimed at a problem the company itself admits is unsolved: getting AI agents to actually work outside a demo. The product is called Presence, and it's built for two jobs businesses care about most, customer service and internal workflows, rather than the one-off internal tools that OpenAI's existing Workspace Agents were designed for.
Those Workspace Agents are essentially customizable GPTs, useful for individual teams but not built to survive contact with a real production environment. Presence is OpenAI's attempt to close that gap. For companies that need more than the standard setup, OpenAI sends in its Forward Deployed Engineers, staff who work directly with a client to map out workflows, wire up existing systems, write the operating guidelines, and shepherd the whole thing through testing to launch.
Access is limited for now. OpenAI is rolling out Presence to qualifying enterprise customers rather than opening it to everyone, and several practical questions are still unanswered, including how it squares with rules like the EU AI Act. OpenAI has talked about building in trust mechanisms but hasn't spelled out what that means legally.
Getting AI agents to work reliably in real business settings is still a messy problem. OpenAI is trying to fix that with Presence, a new offering aimed at enterprise customers.
Why this matters
OpenAI drawing a hard line between Workspace Agents and Presence tells us something about where the company thinks the real money is. GPTs were a playground for internal tinkering; Presence is a bet that businesses will pay for agents that actually survive contact with paying customers, not just internal Slack channels. For founders building on top of OpenAI's stack, the Forward services layer is the detail worth watching: it's an admission that "production-ready" out of the box still has limits, and OpenAI wants to be the one filling those gaps rather than ceding that work to consulting shops or competitors like Salesforce and Microsoft.
For developers, that means Presence could become the default scaffolding for customer service bots, which raises the usual lock-in question: build fast on OpenAI's rails, or keep optionality and build your own orchestration. Researchers should note this as more evidence that reliability, not raw model capability, is the current bottleneck in enterprise AI. Whether Presence actually closes that gap, or just repackages the same agent problems with better branding, is the thing to track over the next few enterprise rollout cycles.
Related Reading
Browse all OpenAI news →
Business & Startups Nordic pilot adds Gemini for Education, NotebookLM to boost AI literacy
Business & Startups Kling launches Video O1, all-in-one model with MVL bridge using transformer
Business & Startups DeepSeek Seeks More Capital Weeks After USD 7B Funding Round
Related Topic OpenAI researcher quits, citing distrust over ad‑driven engagement metrics
LLMs & Generative AI
Related Topic OpenAI launches GPT-Image 1.5 with precise editing for enterprise visuals
LLMs & Generative AI
From Archives Fenix Flexin' New Single Sparks AI Slop Debate Over Vocal Style
From Archives Apple May Charge for Advanced Siri AI Features
From Archives OpenAI to Publish Report on AI Solving Ten Unsolved Math Problems
Open Source
From Archives Sources: More OpenAI Agents Reportedly Escaped Sandboxes
Open Source
Common Questions Answered
What is the key difference between OpenAI's Workspace Agents and the new Presence product?
Workspace Agents are customizable GPTs designed for individual teams and internal use cases, while Presence is specifically built for enterprise customers to handle customer service and internal workflows at production scale. Presence represents OpenAI's effort to create AI agents that can reliably work in real business settings with paying customers, whereas Workspace Agents were designed for one-off internal tools and internal Slack channels.
Why does OpenAI acknowledge that getting AI agents to work in production is an unsolved problem?
OpenAI admits that getting AI agents to work reliably outside of demos and in real business settings is still a messy problem that hasn't been fully solved. This acknowledgment is the core reason the company developed Presence, as they recognized that existing solutions like Workspace Agents weren't equipped to handle the demands of production environments with actual paying customers.
What specific business functions is Presence designed to address?
Presence is built for two primary jobs that businesses care about most: customer service and internal workflows. This focus on high-impact, customer-facing and operational functions distinguishes Presence from earlier offerings that were primarily designed for internal team experimentation.
What does OpenAI's distinction between Workspace Agents and Presence reveal about the company's business strategy?
OpenAI's hard line between Workspace Agents and Presence indicates where the company believes the real revenue opportunity lies in the AI agent market. The company is betting that businesses will pay premium prices for production-ready AI agents that can reliably handle customer interactions, rather than just internal tools for experimentation.
Further Reading
- Introducing OpenAI Presence - OpenAI
- OpenAI Presence: Enterprise Platform for AI Agents - Agenccy AI
- OpenAI Presence: enterprise AI agents, engineers included - Artificial Intelligence News
- OpenAI Launches Presence for Trusted Enterprise AI Agents - HyperAI
- Introducing AgentKit - OpenAI
