Telecom software provider Circles has announced it is using OpenAI's technology to power personalization across its telco offerings, according to a case study published by OpenAI. The integration is aimed at tailoring plans, promotions and customer interactions for telecom operators that use Circles' platform to serve their own end customers.
Circles builds cloud-based software for telecom operators, enabling them to launch and manage digital-first mobile brands. According to OpenAI's writeup, the company is applying OpenAI's models to better understand customer intent and usage patterns, then surface more relevant plan recommendations and support responses. Specific technical details — such as which models are in use, how the system is architected, or measured impact on conversion or retention — were not disclosed in the source material and should be treated as unconfirmed pending further detail from either company.
For an intelligent automation consultancy audience, the significance of this announcement isn't the telecom vertical itself — it's the underlying shift it represents. Personalization at scale has historically required either large data science teams or rigid rules-based segmentation that breaks down as soon as customer behavior gets even slightly unusual. What cases like Circles demonstrate is that large language models can now sit in that gap: ingesting messy, real-world customer data and producing individualized outputs — a recommendation, a message, a next-best-action — without a small army of analysts hand-tuning segments.
That has direct relevance to B2B companies far smaller than a telecom operator. Sales teams juggling a few hundred accounts often default to generic outreach because true personalization at that volume has been operationally expensive. Support teams answer the same handful of ticket types with slightly-off canned responses because writing a truly tailored reply for every ticket doesn't scale on a small team. Operations teams send the same renewal or upsell sequence to every customer regardless of usage signals, because building segment-specific logic is a project, not a quick fix.
The Circles case is a reminder that the technology enabling more granular personalization is now accessible outside of enterprise telecom and consumer tech — but the amount of infrastructure work required to get there responsibly should not be underestimated. Companies considering a similar move should start by asking a narrower question than "how do we personalize everything": which single, repeatable customer touchpoint — a renewal email, a support triage step, a lead-scoring decision — has enough volume and enough available data to make personalization worth automating first. Broad ambition without that starting discipline tends to produce projects that stall before they reach production, regardless of company size.
No pricing, timeline, or rollout details for the Circles-OpenAI integration were included in the source announcement.