{"id":813,"date":"2026-11-16T11:14:53","date_gmt":"2026-11-16T11:14:53","guid":{"rendered":"https:\/\/melento.ai\/blog\/?p=813"},"modified":"2026-11-16T11:15:35","modified_gmt":"2026-11-16T11:15:35","slug":"why-banks-are-building-agentic-ai-centres","status":"publish","type":"post","link":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres","title":{"rendered":"Why Banks Are Building Agentic AI Centres of Excellence"},"content":{"rendered":"<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">It\u2019s a Monday review at a mid-sized bank, and the slide on screen says it all: eleven AI pilots, nine teams, five different tools.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">The collections team has a voice experiment that calls customers and sounds promising. Operations has a chatbot that nobody trusts with anything important. Compliance has asked the same question twice and still has no answer: when an agent acts on its own, who is accountable? And right in the corner, a relationship manager quietly wonders whether getting good at these tools is the same as making herself redundant.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Then the CEO asks the simplest question in the room: \u201cWhen does this pay back?\u201d Nobody can say.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Here is the curious part. Some banks sit through that same meeting a year later and have agents running in production, with owners, guardrails, and numbers to show. They didn\u2019t buy a better model. So what did they change?<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">They built a structure around the work. More and more banks are doing it through an agentic AI Centre of Excellence (CoE), and that is why we\u2019re talking about it. This post covers what a CoE actually does, why readiness has to come first, and what it needs from the platform underneath.<\/p>\n<h2><span style=\"color: #2c5363;\"><b>The pilot problem: plenty of experiments, little production<\/b><\/span><\/h2>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">The numbers point to a wide gap between trying agentic AI and running it. EY\u2019s 2025 banking AI research, discussed in its<a href=\"https:\/\/www.ey.com\/en_dk\/insights\/financial-services\/how-governed-intelligence-can-scale-agentic-banking\"> <span style=\"font-weight: 400;\">governed intelligence paper<\/span><\/a><span style=\"font-weight: 400;\">, found that 52% of banks had piloted agentic AI, but only 16% had fully deployed use cases.<\/span><\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Gartner expects the gap to cost money. It<a href=\"https:\/\/www.gartner.com\/en\/newsroom\/press-releases\/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027?id=106\" class=\"broken_link\"> <span style=\"font-weight: 400;\">predicts that more than 40% of agentic AI projects will be cancelled by the end of 2027<\/span><\/a><span style=\"font-weight: 400;\">, citing rising costs, unclear business value and inadequate risk controls. Analyst Anushree Verma<\/span><a href=\"https:\/\/w.media\/over-40-percent-of-agentic-ai-projects-could-face-the-axe-by-end-of-2027-gartner\/\"> <span style=\"font-weight: 400;\">describes most current projects<\/span><\/a><span style=\"font-weight: 400;\"> as \u201cmostly driven by hype and often misapplied.\u201d<\/span><\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">The pattern inside banks is familiar. In its<a href=\"https:\/\/blogs.infosys.com\/infosys-consulting\/wp-content\/uploads\/2026\/06\/IC260429-02_A-novel-operating-model-for-AI-Transformation-in-Financial-Services_WEB.pdf\" class=\"broken_link\"> <span style=\"font-weight: 400;\">operating model for AI transformation in financial services<\/span><\/a><span style=\"font-weight: 400;\">, Infosys Consulting describes business units experimenting independently, each with its own technology stack and governance, until the lab becomes \u201can expensive showcase rather than a delivery engine.\u201d<\/span><\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A CoE exists to break that pattern.<\/p>\n<h2><span style=\"color: #2c5363;\"><b>Agents act, so the stakes are different.<\/b><\/span><\/h2>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A chatbot answers. An agent does something. McKinsey puts it plainly in its<a href=\"https:\/\/www.mckinsey.com\/capabilities\/tech-and-ai\/our-insights\/the-rise-of-the-agentic-coworker-a-banking-cios-guide-to-the-next-wave-of-ai\" class=\"broken_link\"> <span style=\"font-weight: 400;\">guide for banking CIOs<\/span><\/a><span style=\"font-weight: 400;\">: unlike most copilots, agentic coworkers \u201cdo not only suggest; they execute.\u201d<\/span><\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">That changes what banks have to put in place first. McKinsey notes that there is no separate rulebook for agentic AI, and that regulators are applying existing supervisory expectations today. Each agent needs an explicit identity, a named owner and a formal life cycle, and every action it takes should be logged against that identity.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Melento\u2019s engineers add a practical point from building agentic systems for customers: auditing agent decisions can\u2019t wait until production. If a team only starts asking why an agent chose go-live, capturing that reasoning has come too late. Checkpoints and human-in-the-loop steps belong in the design from day one, so people can see why choices are made instead of treating the agent as a black box.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Indian BFSI teams face the same direction of travel. The RBI\u2019s<a href=\"https:\/\/rbi.org.in\/scripts\/PublicationReportDetails.aspx?ID=1306\"> <span style=\"font-weight: 400;\">FREE-AI committee report<\/span><\/a><span style=\"font-weight: 400;\">, published in August 2025, sets out seven guiding principles, the first of which is \u201cTrust is the Foundation.\u201d It makes 26 recommendations across six pillars, including a board-approved AI policy. One<\/span><a href=\"https:\/\/chambers.com\/articles\/rbi-committee-report-on-responsible-ai-in-the-financial-sector-free-ai-framework\"> <span style=\"font-weight: 400;\">law-firm summary of the report<\/span><\/a><span style=\"font-weight: 400;\"> also highlights capacity building, including in-house expertise and AI Centres of Excellence.<\/span><\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">In short: the more an agent can do, the more a bank has to be able to show about how, why, and under whose authority it did it.<\/p>\n<h2><span style=\"color: #2c5363;\"><b>Why readiness comes before the CoE?<\/b><\/span><\/h2>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A CoE is a response to a deeper problem, and it works best when a bank knows where it stands first. In a Melento discussion on AI readiness, the team made the same point the numbers hint at: pilots rarely fail because of the model. They fail because of everything around it.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Three gaps came up again and again:<\/p>\n<ul>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Systems built for a pre-AI world.<\/b><span style=\"font-weight: 400;\"> Legacy platforms weren\u2019t designed with AI in mind, so plugging agents into them is harder than it looks.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Data that isn\u2019t AI-friendly.<\/b><span style=\"font-weight: 400;\"> Most organisations hold plenty of data, but it is fragmented and was stored for analytics or front-end systems, not for language models. That is why context management matters.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>People who don\u2019t feel safe or skilled.<\/b><span style=\"font-weight: 400;\"> Leaders want AI, yet the mid-level teams who will use it often lack training, and many quietly wonder whether using it well will cost them their jobs.<\/span><\/li>\n<\/ul>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Process maturity matters just as much. Without it, as one Melento speaker put it, AI will \u201conly automate the randomness.\u201d<\/p>\n<h3><span style=\"color: #2c5363;\"><b>MARS: a readiness framework that points the way forward<\/b><\/span><\/h3>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Melento\u2019s answer is MARS, its AI readiness framework. Instead of asking only whether the technology is in place, it looks at five dimensions:<\/p>\n<ul>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>AI awareness<\/b><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Data readiness<\/b><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Integration<\/b><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Productivity impact<\/b><span style=\"font-weight: 400;\">, including quality gains and reduced friction<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>AI governance and culture<\/b><span style=\"font-weight: 400;\">, from privacy awareness to how AI output is double-checked<\/span><\/li>\n<\/ul>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Results map onto five stages:<\/p>\n<table>\n<tbody>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Stage<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>What it looks like<\/b><\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>1. Exploring<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">No formal process. Use is individual and low-key.<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>2. Experimenting<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Small pockets of experimentation, nothing solidified.<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>3. Operationalising<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A few departments run AI inside workflows, with room to scale.<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>4. Scaling<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Business outcomes are measured: revenue, time saved, productivity, or quality.<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>5. AI native<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">AI is part of every step, from decisions to final output.<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone wp-image-824\" src=\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic2aimaturitystaircase-300x157.jpg\" alt=\"The five MARS stages, from Exploring to AI native, with scoring at company, vertical, and individual levels.\" width=\"394\" height=\"206\" srcset=\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic2aimaturitystaircase-300x157.jpg 300w, https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic2aimaturitystaircase-1024x535.jpg 1024w, https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic2aimaturitystaircase-768x401.jpg 768w, https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic2aimaturitystaircase.jpg 1200w\" sizes=\"auto, (max-width: 394px) 100vw, 394px\" \/><\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Scores come at three levels: company, business vertical, and individual, drawn from surveys of employees, vertical heads, and executives. That grassroots view is what sets MARS apart from top-down assessments. And the output is more than a number. It names the next step for each stage and sets out a roadmap to the return the organisation is after.<\/p>\n<h3><span style=\"color: #2c5363;\"><b>Where the CoE comes in?<\/b><\/span><\/h3>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">For a bank, this is where a CoE earns its keep. Readiness results show which departments can pilot first, and the CoE turns the roadmap into shared standards, owners and guardrails. Spreading investment thin across every team is rarely the answer. Proving value in a ready department, measuring it, then bringing the others along is.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Three practical lessons follow:<\/p>\n<ul>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Don\u2019t force-fit AI.<\/b><span style=\"font-weight: 400;\"> Some roles need it at a different scale than others. The goal is to apply AI where it clearly pays off, not to hit an adoption target.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Respect boundaries.<\/b><span style=\"font-weight: 400;\"> Agents can cross departmental lines, and organisations have those lines for good reasons. Unless every department moves at a similar pace, a bank can\u2019t call itself AI native.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Be realistic about ROI.<\/b><span style=\"font-weight: 400;\"> Some outcomes people expect within a month more realistically take around six, and saying so early makes for better boardroom conversations.<\/span><\/li>\n<\/ul>\n<h2><span style=\"color: #2c5363;\"><b>What an agentic AI CoE actually does?<\/b><\/span><\/h2>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A CoE is not a committee that approves slide decks.<a href=\"https:\/\/www.lumenova.ai\/blog\/ai-center-of-excellence-for-banks\/\"> <span style=\"font-weight: 400;\">Lumenova<\/span><\/a><span style=\"font-weight: 400;\">, a vendor, makes a useful distinction: a governance committee sets policy and approves use cases, while a CoE is operational, building and maintaining the tooling, standards, and support teams use every day.<\/span><\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">In practice, five jobs matter most.<\/p>\n<ul>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Set the guardrails.<\/b><span style=\"font-weight: 400;\"> Define which actions an agent may take alone and which need a human to confirm, then apply those tiers everywhere.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Keep the register.<\/b><span style=\"font-weight: 400;\"> Track every agent, its owner, its permissions, and its lifecycle in one place.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Build the foundations once.<\/b><span style=\"font-weight: 400;\"> Data access, integrations, and context are shared, not rebuilt for each pilot.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Choose use cases on value and risk.<\/b><span style=\"font-weight: 400;\"> Gartner advises pursuing agentic AI only where it delivers clear value or return on investment. A CoE is where that discipline lives.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Produce the evidence.<\/b><span style=\"font-weight: 400;\"> Auditors and supervisors want to see what happened, so the record should be a by-product of the work.<\/span><\/li>\n<\/ul>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">The third job deserves emphasis. Celonis argues that when agents share a contextual layer and bespoke integrations are minimised,<a href=\"https:\/\/www.celonis.com\/news\/article\/the-operational-context-imperative-why-banks-are-failing-to-scale-ai-and-how-they-can\"> <span style=\"font-weight: 400;\">time to production shortens and governance costs fall<\/span><\/a><span style=\"font-weight: 400;\">. Its conclusion is that leading banks will be those with \u201cthe most AI-ready architecture,\u201d not necessarily the best models.<\/span><\/p>\n<table>\n<tbody>\n<tr>\n<td><\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Pilots in silos<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>With an agentic AI CoE<\/b><\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Policy<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Set team by team<\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">One board-aligned policy, applied everywhere<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Ownership<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Unclear who answers for an agent<\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A named owner and a register for each agent<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Integration<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Rebuilt for every pilot<\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Built once, reused<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Use cases<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Chosen by enthusiasm<\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Chosen by value and risk<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Evidence<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Reconstructed on request<\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Recorded alongside the work<\/p>\n<\/td>\n<\/tr>\n<tr>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Human oversight<\/b><\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Varies by team<\/p>\n<\/td>\n<td>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Defined confirmation points<\/p>\n<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><img loading=\"lazy\" decoding=\"async\" class=\"alignnone wp-image-820\" src=\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic1coehubandspoke-300x157.jpg\" alt=\"An agentic AI Centre of Excellence at the centre, connected to business teams that each own their agents.\" width=\"447\" height=\"234\" srcset=\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic1coehubandspoke-300x157.jpg 300w, https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic1coehubandspoke-1024x535.jpg 1024w, https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic1coehubandspoke-768x401.jpg 768w, https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/infographic1coehubandspoke.jpg 1200w\" sizes=\"auto, (max-width: 447px) 100vw, 447px\" \/><\/p>\n<h2><span style=\"color: #2c5363;\"><b>A CoE sets the standard, but doesn\u2019t own the outcome.<\/b><\/span><\/h2>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">One caution keeps a CoE honest. McKinsey is explicit that ownership of an agent should not default to a generalised AI centre of excellence. Every agent needs a clearly accountable human who answers for the result it produces, and in most cases that person sits in the business.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">The operating model also depends on scale. EY observes that a central AI factory can give smaller or less mature banks focus, scarce-skill leverage and a repeatable path from idea to production, while larger banks may spread delivery more widely. The common thread is that standards are shared and accountability stays close to the work.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Mastercard\u2019s recent move shows the pattern at a global scale. It announced<a href=\"https:\/\/www.mastercard.com\/news\/ap\/en\/newsroom\/press-releases\/en\/2026\/mastercard-powering-asean-s-ai-ambitions-in-the-future-of-payments\/\" class=\"broken_link\"> <span style=\"font-weight: 400;\">authenticated agentic transactions across ASEAN markets<\/span><\/a><span style=\"font-weight: 400;\"> alongside a planned regional AI Centre of Excellence in Singapore, bringing its innovation hub, cybersecurity work and AI expertise together, with governance and auditability stated as part of the offer.<\/span><\/p>\n<h2><span style=\"color: #2c5363;\"><b>What a CoE needs from its platform?<\/b><\/span><\/h2>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A CoE can write excellent policy and still stall if the platform underneath can\u2019t enforce it. The platform has to let teams build quickly, keep work visible, reach customers where they are, and ask for a human decision at the right moments.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">That is the role Mel AI plays on the Collaborative Intelligence Platform (CIP). It is Melento\u2019s AI layer across the platform: teams state what they want in plain language, and the platform acts on it, with guidance along the way. Four capabilities map directly to a CoE\u2019s needs.<\/p>\n<ul>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Workflows by conversation.<\/b><span style=\"font-weight: 400;\"> Teams can draft and refine MStream workflows through chat, with guidance on nodes, expressions, and product capabilities, which makes standard patterns quicker to create and reuse.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Work with a visible home.<\/b><span style=\"font-weight: 400;\"> In MWork, each use case can live in a space with dots, comments, and attachments, so progress and decisions sit in one place.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Insight on request.<\/b><span style=\"font-weight: 400;\"> Ask about work items, overdue tasks, or workflow execution and get summaries and visuals, useful for CoE reviews.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Confirmation before consequence.<\/b><span style=\"font-weight: 400;\"> Before an impactful action, Mel AI stops and checks with you. For a CoE, that is the human confirmation tier made concrete.<\/span><\/li>\n<\/ul>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">\u201cMel AI moves you from navigating individual modules to expressing your intent and getting work done.\u201d<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">From the Mel AI launch announcement<\/p>\n<h3><span style=\"color: #2c5363;\"><b>Voice belongs inside the governed workflow<\/b><\/span><\/h3>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">For banks, some of the highest-value use cases are conversations: reminders, follow-ups, and collections. Melento\u2019s voice agents are built into workflows rather than bolted on beside them. An agent can place a phone call as a step in the process; voice and text can be combined, and every regional language is supported, which matters for customers who prefer to be reached in their own language.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">A collections cycle shows how this plays out. The workflow flags overdue accounts, the voice agent calls, the customer\u2019s response is captured, the tracker is updated, and the next action is scheduled. Anything sensitive, such as an escalation, goes to a person for approval. The call is a governed step, not a side channel, which is exactly what a CoE wants from voice.<\/p>\n<h2><span style=\"color: #2c5363;\"><b>A starter blueprint for a bank\u2019s first CoE<\/b><\/span><\/h2>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">This is a suggested path, not a rule. Adapt it to your size and regulatory posture.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Before the five steps, take a readiness baseline. A framework such as MARS shows which departments are ready to pilot and which need data or skills work first, so the steps below begin where the bank actually is.<\/p>\n<ol>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Name the sponsor and the owners.<\/b><span style=\"font-weight: 400;\"> One executive sponsor for the CoE, and a named business owner for every agent.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Anchor on policy.<\/b><span style=\"font-weight: 400;\"> Align to a board-approved AI policy and your regulator\u2019s expectations, including incident reporting.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Pick two or three bounded use cases.<\/b><span style=\"font-weight: 400;\"> Reminders, onboarding follow-ups, and collections are natural starting points because outcomes are measurable.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Set the confirmation tiers.<\/b><span style=\"font-weight: 400;\"> Decide which actions run alone, which need a human, and which are off-limits.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Review on a rhythm.<\/b><span style=\"font-weight: 400;\"> Look at status, exceptions, and value on a regular cadence, and retire what doesn\u2019t earn its place.<\/span><\/li>\n<\/ol>\n<h2><span style=\"color: #2c5363;\"><b>So, why are banks building these?<\/b><\/span><\/h2>\n<ul>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>To move from pilots to production<\/b><span style=\"font-weight: 400;\"> with shared standards instead of one-off experiments.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>To make accountability explicit<\/b><span style=\"font-weight: 400;\"> for software that can act, not just advise.<\/span><\/li>\n<li style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>To build the foundations once<\/b><span style=\"font-weight: 400;\"> so every new use case starts further ahead.<\/span><\/li>\n<\/ul>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\">Intent in, governed action out.<\/p>\n<p style=\"text-align: justify; font-size: 17px; font-family: 'open sans';\"><b>Exploring a CoE of your own?<\/b><span style=\"font-weight: 400;\"> See how Mel AI can anchor it. Try Mel AI, or ask us for a walkthrough built around your use cases.<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>It\u2019s a Monday review at a mid-sized bank, and the slide on screen says it all: eleven AI pilots, nine teams, five different tools. The collections team has a voice experiment that calls customers and sounds promising. Operations has a chatbot that nobody trusts with anything important. Compliance has asked the same question twice and [&hellip;]<\/p>\n","protected":false},"author":4,"featured_media":826,"comment_status":"open","ping_status":"open","sticky":false,"template":"page_templates\/blog-new-oldd.php","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[78],"tags":[155,156,157,159,162,158,161,160],"class_list":["post-813","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-cip","tag-agentic-ai-centre-of-excellence","tag-agentic-ai-in-banking","tag-ai-coe-for-banks","tag-bfsi-ai-governance","tag-collaborative-intelligence-platform","tag-governed-agentic-ai","tag-mel-ai","tag-voice-agents-for-bfsi"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v26.2 - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Why Are Banks Building Agentic AI Centres?<\/title>\n<meta name=\"description\" content=\"Discover why banks are building agentic AI centres to automate operations, enhance customer experiences, and drive innovation.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Why Are Banks Building Agentic AI Centres?\" \/>\n<meta property=\"og:description\" content=\"Discover why banks are building agentic AI centres to automate operations, enhance customer experiences, and drive innovation.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres\" \/>\n<meta property=\"og:site_name\" content=\"Melento\" \/>\n<meta property=\"article:published_time\" content=\"2026-11-16T11:14:53+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-11-16T11:15:35+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"627\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/jpeg\" \/>\n<meta name=\"author\" content=\"Neela S\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Neela S\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"10 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"WebPage\",\"@id\":\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres\",\"url\":\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres\",\"name\":\"Why Are Banks Building Agentic AI Centres?\",\"isPartOf\":{\"@id\":\"https:\/\/melento.ai\/blog\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#primaryimage\"},\"image\":{\"@id\":\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#primaryimage\"},\"thumbnailUrl\":\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg\",\"datePublished\":\"2026-11-16T11:14:53+00:00\",\"dateModified\":\"2026-11-16T11:15:35+00:00\",\"author\":{\"@id\":\"https:\/\/melento.ai\/blog\/#\/schema\/person\/6ce1ed6118768da13c88a1962956fc89\"},\"description\":\"Discover why banks are building agentic AI centres to automate operations, enhance customer experiences, and drive innovation.\",\"breadcrumb\":{\"@id\":\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#primaryimage\",\"url\":\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg\",\"contentUrl\":\"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg\",\"width\":1200,\"height\":627,\"caption\":\"Why Banks Are Building Agentic AI Centres of Excellence\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\/\/melento.ai\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Why Banks Are Building Agentic AI Centres of Excellence\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/melento.ai\/blog\/#website\",\"url\":\"https:\/\/melento.ai\/blog\/\",\"name\":\"Melento\",\"description\":\"\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/melento.ai\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Person\",\"@id\":\"https:\/\/melento.ai\/blog\/#\/schema\/person\/6ce1ed6118768da13c88a1962956fc89\",\"name\":\"Neela S\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\/\/melento.ai\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/secure.gravatar.com\/avatar\/31060a06bc054fc55d667191056b1dbfa07afa61c07c2376f75bdaf190218503?s=96&d=mm&r=g\",\"contentUrl\":\"https:\/\/secure.gravatar.com\/avatar\/31060a06bc054fc55d667191056b1dbfa07afa61c07c2376f75bdaf190218503?s=96&d=mm&r=g\",\"caption\":\"Neela S\"},\"url\":\"https:\/\/melento.ai\/blog\/author\/neela-mel\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Why Are Banks Building Agentic AI Centres?","description":"Discover why banks are building agentic AI centres to automate operations, enhance customer experiences, and drive innovation.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres","og_locale":"en_US","og_type":"article","og_title":"Why Are Banks Building Agentic AI Centres?","og_description":"Discover why banks are building agentic AI centres to automate operations, enhance customer experiences, and drive innovation.","og_url":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres","og_site_name":"Melento","article_published_time":"2026-11-16T11:14:53+00:00","article_modified_time":"2026-11-16T11:15:35+00:00","og_image":[{"width":1200,"height":627,"url":"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg","type":"image\/jpeg"}],"author":"Neela S","twitter_card":"summary_large_image","twitter_misc":{"Written by":"Neela S","Est. reading time":"10 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"WebPage","@id":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres","url":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres","name":"Why Are Banks Building Agentic AI Centres?","isPartOf":{"@id":"https:\/\/melento.ai\/blog\/#website"},"primaryImageOfPage":{"@id":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#primaryimage"},"image":{"@id":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#primaryimage"},"thumbnailUrl":"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg","datePublished":"2026-11-16T11:14:53+00:00","dateModified":"2026-11-16T11:15:35+00:00","author":{"@id":"https:\/\/melento.ai\/blog\/#\/schema\/person\/6ce1ed6118768da13c88a1962956fc89"},"description":"Discover why banks are building agentic AI centres to automate operations, enhance customer experiences, and drive innovation.","breadcrumb":{"@id":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#primaryimage","url":"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg","contentUrl":"https:\/\/melento.ai\/blog\/wp-content\/uploads\/2026\/11\/banneragenticaicoe.jpg","width":1200,"height":627,"caption":"Why Banks Are Building Agentic AI Centres of Excellence"},{"@type":"BreadcrumbList","@id":"https:\/\/melento.ai\/blog\/why-banks-are-building-agentic-ai-centres#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/melento.ai\/blog\/"},{"@type":"ListItem","position":2,"name":"Why Banks Are Building Agentic AI Centres of Excellence"}]},{"@type":"WebSite","@id":"https:\/\/melento.ai\/blog\/#website","url":"https:\/\/melento.ai\/blog\/","name":"Melento","description":"","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/melento.ai\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Person","@id":"https:\/\/melento.ai\/blog\/#\/schema\/person\/6ce1ed6118768da13c88a1962956fc89","name":"Neela S","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/melento.ai\/blog\/#\/schema\/person\/image\/","url":"https:\/\/secure.gravatar.com\/avatar\/31060a06bc054fc55d667191056b1dbfa07afa61c07c2376f75bdaf190218503?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/31060a06bc054fc55d667191056b1dbfa07afa61c07c2376f75bdaf190218503?s=96&d=mm&r=g","caption":"Neela S"},"url":"https:\/\/melento.ai\/blog\/author\/neela-mel"}]}},"_links":{"self":[{"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/posts\/813","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/comments?post=813"}],"version-history":[{"count":12,"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/posts\/813\/revisions"}],"predecessor-version":[{"id":828,"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/posts\/813\/revisions\/828"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/media\/826"}],"wp:attachment":[{"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/media?parent=813"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/categories?post=813"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/melento.ai\/blog\/wp-json\/wp\/v2\/tags?post=813"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}