OpenAI launches Presence, an enterprise platform for deploying managed voice and chat AI agents
Product launch ● Confirmed 85% confidence first seen
OpenAI announced Presence, an enterprise platform designed to help large organizations deploy and manage AI agents for customer service and internal workflows. The platform includes governance controls, policy management, and monitoring tools, and is being rolled out to early design partners including BBVA, SoftBank, and IAG. The product is built on AI agents that OpenAI has been running internally on its own support line, which reportedly resolves 75% of English-language support inquiries without human intervention.
Decision brief
- What changed
- OpenAI launched Presence, an enterprise platform for deploying and managing voice and chat AI agents with built-in governance controls, policy management, and monitoring tools. The platform is built on agent technology OpenAI has used internally on its own support line (reportedly resolving 75% of English-language support inquiries without human intervention) and is rolling out via restricted access to early design partners including BBVA, SoftBank, and IAG.
- Why it matters
- This signals OpenAI's push from general-purpose AI models into packaged enterprise infrastructure for customer service and internal workflow automation, directly competing with contact-center and CX software vendors. For large enterprises, the governance/policy/monitoring framing suggests OpenAI is trying to address compliance and oversight concerns that have blocked prior agent deployments, which could accelerate adoption in regulated industries like banking (BBVA) and insurance (IAG) if the model holds up. However, since it's an early, restricted rollout with unproven external performance, decision-makers should treat this as a pilot opportunity rather than a mature solution.
- Evidence
- Coverage comes from OpenAI's own blog (marketing material with minimal detail) plus three secondary/aggregator outlets (The New Stack, TLDR, TLDR Dev) that largely restate the same launch facts and design-partner list without independent verification of performance claims; only The New Stack provides added context on the internal support-line origin.
- What remains uncertain
- The 75% resolution figure is self-reported by OpenAI for its own internal support line and has not been independently verified or shown to generalize to third-party enterprise deployments, non-English languages, or complex workflows. Pricing, availability timeline beyond design partners, specific governance/monitoring capabilities, and security/compliance certifications are not detailed in any source.
- Monitor next
- Watch for public case studies, performance data, or expanded availability announcements from BBVA, SoftBank, or IAG following their design-partner pilots.
Analytical support, not advice — assumptions and open questions stated above.