More than 15,000 M&T Bank employees now have AI copilots at their desks — a scale that puts the Buffalo-based regional lender among the most aggressive adopters of enterprise AI in American banking. The deployment comes after years of internal technology modernisation, and it signals how mid-sized banks are quietly reshaping daily operations with generative AI.
How M&T Bank is using AI across daily operations
The bank's AI tools now analyse call-centre conversations to identify customer needs, draft internal reports, generate software code, and flag potential risks in loan portfolios, according to Fast Company. For employees, this means routine tasks like summarising customer interactions or preparing memos are increasingly handled by AI assistants.
American Banker reported in September 2025 that roughly 16,000 of M&T's approximately 22,000 employees were already using Microsoft Copilot. That means nearly three-quarters of the workforce has access to AI tools for tasks ranging from drafting emails to synthesising call-centre transcripts.
Why a regional bank's AI rollout matters beyond M&T
Large money-centre banks like JPMorgan Chase have dominated headlines about AI in finance. But M&T's deployment shows that regional banks — often seen as technology laggards — are now embedding AI into core operations at scale. This matters because regional banks collectively serve millions of American customers, and their efficiency choices directly affect service quality and branch staffing.
For customers, AI-driven call-centre analysis could mean faster issue resolution. For employees, it means shifting from repetitive tasks to higher-judgment work — or, in some cases, concerns about job displacement.
The technology overhaul that made this AI expansion possible
M&T's AI rollout did not happen in isolation. The bank spent years modernising its technology infrastructure before deploying AI broadly. This foundation work — including data standardisation and cloud migration — created the conditions for safe, controlled AI adoption.
Notably, M&T initially restricted employee access to public large language models. The chief data officer has indicated this cautious approach allowed the bank to understand risks before enabling enterprise-wide tools.
What 15,000 employees with AI copilots means for banking jobs
The human impact of this rollout cuts both ways. For customer-service representatives, AI that analyses call conversations can reduce note-taking and let them focus on the caller. For software developers, AI-generated code can accelerate delivery timelines. But the same tools raise legitimate questions about which tasks remain human-only.
M&T's approach appears to be augmentation rather than replacement — AI copilots assist employees rather than operate independently. The bank's exploration of agentic AI in cybersecurity and fraud detection, however, suggests a future where AI takes more autonomous actions.
What M&T's chief data officer has said about the AI strategy
The bank's leadership has framed the rollout as a deliberate, staged process. By initially restricting access to public large language models, M&T sought to avoid the data-leakage and compliance risks that have worried regulators across the banking sector.
This measured approach reflects a broader industry reality: banks face strict regulatory scrutiny over data privacy, model risk, and fair lending. Any AI tool touching customer data must be validated for bias and accuracy before wide deployment.
Why M&T's AI bet is a test case for mid-sized banks
M&T's scale — roughly 22,000 employees and over $200 billion in assets — makes it a meaningful reference point. If the bank demonstrates measurable productivity gains without regulatory missteps, other regional lenders may follow. If problems emerge, the cautious approach many banks have taken will be vindicated.
The bank's focus on internal operations first — rather than customer-facing AI chatbots — suggests a priority on employee productivity and risk management before public-facing experimentation.
Confirmed facts versus what remains unclear about M&T's AI rollout
Verified: M&T has deployed AI copilots to more than 15,000 employees. The bank uses AI for call-centre analysis, report drafting, code generation, customer-need identification, and portfolio risk flagging. Microsoft Copilot is a primary tool, with 16,000 users reported in September 2025.
Unclear: The specific productivity gains achieved, the exact timeline for agentic AI deployment in cybersecurity and fraud detection, and how M&T measures AI's impact on customer satisfaction or employee workload remain undisclosed. The chief data officer's full comments were not available in the source material.
M&T Bank's competitive position in banking technology
M&T's technology overhaul gives it a foundation that many peers lack. Years of infrastructure modernisation — including data standardisation and cloud readiness — created the conditions for enterprise-wide AI. This technological moat matters because AI tools are only as good as the data they access.
Banks with fragmented legacy systems cannot simply switch on AI copilots. M&T's years of preparation position it ahead of regional competitors that are still consolidating their data environments.
Risks and balanced view on enterprise AI in banking
The benefits of M&T's AI rollout — efficiency, faster code delivery, better risk detection — are clear. But risks deserve equal attention. AI models can produce biased outputs, hallucinate facts, or mishandle sensitive customer data. Regulatory bodies including the Federal Reserve and OCC have signalled heightened scrutiny of AI in banking.
Employee concerns about job security also remain unresolved. While M&T frames AI as assistive, the long-term impact on headcount in call centres and back-office functions is unknown. Critics of banking AI argue that cost-cutting pressures will eventually translate AI productivity gains into workforce reductions.
The wider trend: how US banks are adopting generative AI
M&T is part of a broader wave. Major banks including JPMorgan, Bank of America, and Wells Fargo have all deployed generative AI tools internally. What distinguishes M&T is the speed and scale relative to its size — a regional bank moving with the ambition of a money-centre institution.
The banking industry's AI adoption is being watched closely by regulators, technology vendors, and competitors. Microsoft's Copilot has become a default choice for financial institutions given its enterprise security features and existing Office integrations.
What M&T employees and customers should watch for next
For M&T employees, the practical step is to understand which tasks AI can handle and where human judgment remains essential. For customers, the visible changes will likely be faster call resolution and more personalised service — though the bank has not disclosed customer-facing AI features.
Investors and industry watchers should monitor M&T's quarterly earnings commentary for mentions of AI-driven efficiency gains or cost savings. The bank's exploration of agentic AI in fraud detection could also signal future product announcements.
Future outlook: where M&T's AI journey goes from here
The next phase for M&T appears to be agentic AI — systems that take actions rather than merely assist. The bank is examining applications in cybersecurity and fraud detection, areas where autonomous AI could flag suspicious activity in real time.
Any such deployment will require careful regulatory navigation. But if M&T succeeds, it could become a template for how regional banks adopt advanced AI without compromising safety or customer trust.
Our Take
M&T Bank's enterprise AI expansion is significant not because it is unique, but because it is representative. A regional bank with 22,000 employees has done what many assumed only the largest financial institutions could achieve — deploying AI copilots to the majority of its workforce in a controlled, staged manner.
The real story is the years of technology overhaul that preceded this moment. AI is not a shortcut; it is a payoff for infrastructure work that began long before generative AI became mainstream. For other banks still modernising their systems, M&T's journey offers a clear lesson: prepare the data foundation first, then layer AI on top.
The unanswered question remains workforce impact. M&T frames AI as augmentation, but the banking industry's history suggests efficiency gains eventually reshape staffing. How M&T manages that transition — transparently or otherwise — will define whether this rollout is seen as a model or a warning.
Frequently Asked Questions
How many M&T Bank employees are using AI tools?
M&T Bank has deployed AI copilots to more than 15,000 employees. American Banker reported in September 2025 that 16,000 of the bank's roughly 22,000 employees were using Microsoft Copilot for tasks including drafting emails, reports, and summarising call-centre conversations.
What does M&T Bank use AI for?
M&T uses AI to analyse call-centre conversations, draft internal reports, generate software code, identify customer needs, and flag portfolio risks. The bank is also exploring agentic AI applications in cybersecurity and fraud detection.
Is M&T Bank replacing employees with AI?
M&T's current AI deployment is framed as assistive — AI copilots support employees rather than replace them. The bank initially restricted access to public large language models before the controlled rollout, suggesting a cautious approach to workforce impact.
Why did M&T Bank restrict public AI access before the rollout?
M&T's chief data officer indicated the bank initially restricted employee access to public large language models to understand risks before enabling enterprise-wide tools. This reflects regulatory concerns about data privacy, model bias, and compliance in banking.
