Introducing OpenAI Presence

2026-07-22 · Source: OpenAI News · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, short

Summary

OpenAI introduced Presence on July 22, 2026, an enterprise product designed to deploy reliable AI agents for high-value customer and internal workflows. Presence combines model reasoning with policies, guardrails, and escalation rules to ensure accuracy and performance, adapting as products, policies, and user behavior change. It supports real-time voice and chat agents for tasks like resolving billing issues, supporting insurance claims, and handling IT service requests. Each deployment is customized with specific knowledge and system access, and companies define agent actions, approval needs, and human escalation points. After launch, a Codex-powered improvement process proposes updates based on production sessions and escalations. Presence has been proven internally, handling OpenAI's English-language phone support at 1-888-GPT-0090, resolving 75% of inbound issues without human assistance and reducing human handoffs by 15 percentage points in 10 days. Leading enterprises like BBVA, SoftBank, and IAG are also exploring its capabilities.

Key takeaway

For Directors of AI/ML evaluating agent deployment solutions, OpenAI Presence offers a battle-tested platform to operationalize AI agents reliably. You should consider its integrated policy enforcement, guardrails, and continuous improvement loop to ensure agents adapt to evolving business needs while maintaining control. Contact your OpenAI account team to explore how Presence can resolve 75% of inbound issues or reduce human handoffs by 15 percentage points in your high-value workflows.

Key insights

OpenAI Presence enables reliable, adaptable AI agents for enterprise workflows through integrated policies, guardrails, and continuous improvement.

Principles

• Agent reliability requires systems, evaluations, and deployment expertise.
• Policies, guardrails, and escalation rules verify agent accuracy.
• Continuous improvement adapts agents to changing conditions.

Method

Deployments start with a specific job, granting minimal access. Companies set policies. Production data informs Codex-proposed updates, which teams test and approve for controlled rollout.

In practice

• Use for customer support, outbound sales, or IT service requests.
• Implement for voice and chat agents in mission-critical environments.
• Test agents against edge cases and high-risk scenarios pre-launch.

Topics

• AI Agents
• Enterprise AI
• Customer Support Automation
• Workflow Automation
• OpenAI Presence
• Agent Deployment
• AI Governance

Best for: CTO, VP of Engineering/Data, Executive, MLOps Engineer, Director of AI/ML, Consultant

Related on AIssential

• OpenAI unveils Presence, a new platform that lets enterprises launch and manage realtime voice agents and chatbots — OpenAI's Presence provides a governed platform for deploying and managing reliable enterprise AI agents in production.
• The three disciplines separating AI agent demos from real-world deployment — Reliable AI agent deployment requires addressing data fragmentation, workflow clarity, and robust management beyond initial demonstrations.
• Agents in Production: How OpenGov Built and Scaled OG Assist - Gabe De Mesa, OpenGov — OpenGov's OG Assist demonstrates a robust production AI agent architecture, emphasizing control, safety, and continuous improvement.
• OpenAI launches a way for enterprises to build and manage AI agents — OpenAI Frontier provides an open, end-to-end platform for enterprise AI agent management, integrating external data and applications.
• AI alone won’t change your business. The system running it will. — Enterprise AI success hinges on a governed, continuously improving system for agents, not just powerful models.
• ALERT: OpenAI and Anthropic Just Proved AI Cannot Replace Indian IT After All — AI labs now recognize the critical need for implementation services, shifting Indian IT's role from threat to essential partner.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by OpenAI News.