PwC and OpenAI Launch Agentic AI for Contact Centres

What happened

PwC and OpenAI have jointly launched a suite of agentic solutions targeting contact-centre and customer-service operations, marking one of the most significant enterprise deployments of autonomous AI agents in the professional-services sector to date. The collaboration brings OpenAI's large language model capabilities together with PwC's industry and transformation consulting expertise to deliver AI-driven agents capable of handling end-to-end service interactions with reduced human intervention.

The solutions are designed to move beyond simple chatbot automation, enabling agents that can reason across multi-step service workflows — querying back-end systems, resolving complaints, processing requests and escalating edge cases — all within a single conversational thread. PwC is positioning itself as both an implementer and a co-developer of these tools, embedding them into client engagements across sectors including financial services, healthcare and retail.

Why it matters

The shift from scripted bots to genuinely agentic AI represents a structural inflection point for customer-experience design. Where previous automation reduced handle time on simple queries, agentic systems can now own complex, multi-turn service journeys — fundamentally changing the role of the human agent from primary responder to exception handler and empathy layer. For CX leaders, this is not an incremental efficiency play; it is a redesign of the service operating model itself.

From a behavioural-economics perspective, the implications are equally significant. Customers' trust in automated service is fragile and highly context-dependent — shaped by perceived competence, consistency and the availability of a human fallback. Deploying agents that can reason and act autonomously raises the stakes on every one of those dimensions. Organisations that treat this as a cost-reduction exercise without re-engineering the experience around it risk triggering the very dissatisfaction they are trying to eliminate.

The Renascence take

The PwC–OpenAI announcement will be read by most operators as validation to accelerate automation rollouts. That instinct is understandable — and largely wrong in its framing.

The real design challenge here is not technical — it is psychological. Agentic AI can resolve a query faster than any human team, but speed is not the primary driver of customer trust or loyalty; perceived care and felt control are. The operators who will win are those who use agentic capability to create more space for meaningful human moments, not fewer. Before deploying autonomous agents at scale, a customer-obsessed organisation should map precisely where in its service journey customers most need to feel heard — and protect those touchpoints fiercely, regardless of what the automation can technically handle. Efficiency that erodes emotional connection is not a service improvement; it is a slow brand liability.