The agentic AI market in financial services is valued at $7.78 billion in 2026 and is projected to reach $43.5 billion by 2031, growing at 41 percent annually, according to Mordor Intelligence. According to Deloitte, among companies already using generative AI, 25 percent were expected to launch agentic AI pilots or proofs of concept in 2025, rising to 50 percent by 2027. By then, 74 percent of enterprises surveyed expect to be using AI agents at least moderately across their operations.
Yet the readiness gap is stark. Only 21 percent of surveyed organizations have a mature governance model for agentic AI, while just 11 percent are actively using agentic systems in production. For banks that fail to adapt, the cost is not just inefficiency. McKinsey’s Global Banking Annual Review 2025 estimates that shifts driven by AI could put up to $170 billion in global banking profits at risk over the next decade.
Why customer communication is the right place to start
Customer communication in BFSI touches every revenue line: acquisition campaigns, cross-sell offers, regulatory disclosures, retention nudges, and advisor briefings. It also carries significant compliance exposure. Today, most of it moves through disconnected teams, manual review cycles, and siloed channels. Personalization rarely goes beyond the first name.
The business case for change is clear. Personalization leaders grow revenue 10 percentage points faster annually than laggards, and financial services firms using personalized marketing have increased revenue by more than 30 percent, according to BCG. The gap between what customers expect and what most institutions deliver is where agentic AI creates immediate, measurable value.
What AI agents do that generative AI alone cannot
Generative AI produces content. Agentic AI orchestrates outcomes. The distinction matters. An AI agent does not wait for a prompt; it takes a goal, breaks it into steps, coordinates with other agents, acts across systems, and learns from the result. In customer communication, that means an end-to-end workflow: understanding a campaign brief, generating multi-channel creative, adapting tone per audience segment, localizing across markets, validating against regulatory guardrails, and surfacing everything for human approval. Not one task at a time, but as a coordinated system.
A BFSI campaign, two realities
Without agents: A bank launches a regulated cross-selling campaign. Briefs move through disconnected teams. Compliance review adds days, sometimes weeks. Messaging rarely moves past the first name. Once the campaign ships, attribution is guesswork. Every handoff is a delay and a compliance risk.
With AI agents: Infosys BPM has introduced ACE, the Agentic Communication Engine, a governed agentic platform purpose-built for enterprise marketing operations that integrates seamlessly with clients’ existing mar-tech environments. Instead of manually assembling campaigns channel by channel over days, ACE starts from a single brief and deploys eight coordinated AI agents to do in hours what previously took weeks.
- A brief analyst agent interprets the campaign objective and offer.
- A content generator produces multi-channel draft creative across email, push, SMS, WhatsApp, social and the RM (relationship manager) advisor screen.
- A tone adapter calibrates messaging for each audience segment: warm, bold, or playful.
- A channel formatter structures every output to the exact specifications of its delivery platform.
- A market localizer adapts language, imagery and cultural context across markets.
- A compliance validator checks every variant against regional regulatory guardrails.
- A campaign launcher deploys each campaign into the specified channel by seamlessly integrating into the email service providers (ESPs) or ad platforms such as Google Ad Manager (GAM).
- A performance learner feeds campaign outcomes back into the next cycle, so each run improves on the last.
A human control tower governs the entire system. Nothing ships on autopilot before a person reviews, approves and publishes every variant. The full loop, from a brief to multi-channel, multi-tone, multi-market creative set, completes in hours rather than weeks. The agentic orchestration sits on top of existing mar-tech systems without the need for an expensive technology change program.
The economics are hard to ignore
Agentic AI could unlock $3 trillion in corporate productivity improvements annually and improve EBITDA by 5.4 percent for the average company, according to KPMG. McKinsey reports that early agentic AI use cases have reduced manual workloads by 30 to 50 percent and estimates that agentic AI could lower banks’ operational costs by 20 percent or more. ACE is designed to deliver 30 to 50 percent lower costs, 40 to 60 percent faster speed-to-market, and fully traceable compliance across every campaign.
The window is narrow
Agentic AI is a major source of both efficiency gains and disruption for banks, with early adopters positioned to gain an advantage before those benefits are taken away by the competition. The institutions that build governed, multi-agent systems now will compound their advantage through every campaign cycle. Those still running pilots when the market moves will face a capability gap that only widens.
ACE is how Infosys BPM brings this governed, multi-agent approach to customer communication to life: a single governed platform that unifies content, compliance and campaign under eight purpose-built agents and a human control tower. The result is communication that is faster, compliant by design, personalized at scale, and finally measurable.
The question for BFSI leadership is no longer whether agentic AI will reshape customer communication. It is whether the organization will be the one doing the reshaping.
