The Complete Guide to Using AI in the Financial Services Industry in Raleigh in 2025

Raleigh, North Carolina is uniquely positioned to scale AI in financial services in 2025: state and private activity is converging - from a 12‑week pilot where the N.C. Department of State Treasurer and OpenAI used ChatGPT to identify “millions of dollars” in potential unclaimed property (North Carolina Treasurer and OpenAI ChatGPT pilot identifying unclaimed property) to Governor Stein's announcement that Amazon will invest $10 billion to build major AWS cloud and AI infrastructure in North Carolina (Amazon $10B AWS AI infrastructure investment in North Carolina), providing the compute, jobs, and training pipelines financial firms need.

Local deployments - like Raymond James' Raleigh rollout of Zoom AI Companion meeting summaries - show operational wins, while industry reports warn that governance and regulatory scrutiny are accelerating.

For teams ready to move from experiment to impact, Nucamp's Nucamp AI Essentials for Work 15-week bootcamp offers a 15‑week, nontechnical path to practical prompts, tooling, and workplace AI skills that turn pilot projects into measurable productivity gains.

“Our team set out to find out how we could modernize our department, while still providing top notch service to folks across the state… As this pilot program wraps up, we are thrilled to say our divisions were able to take that publicly available information and utilize ChatGPT in ways that resulted in tangible and measurable improvements to their daily workflow.” - State Treasurer Brad Briner

Raleigh's financial services firms should be preparing for a 2025 where agentic AI moves from neat pilots to everyday work: multi‑agent systems will orchestrate document‑heavy workflows, surface high‑value exceptions, and even draft actionable summaries so teams start each week with a prioritized to‑do list rather than a backlog; industry surveys show adoption is widespread - nearly 80% of organizations are using AI agents and 96% plan to expand this year, with many expecting outsized returns (62% project ROI above 100%) (Multimodal agent adoption statistics and ROI forecasts).

That momentum maps directly to three strategic priorities that matter for Raleigh institutions - workflow‑level efficiency, stronger risk controls, and hyper‑personalized customer experiences - captured in vendor and industry guidance as the path from pilots to production (nCino 2025 AI trends analysis for financial services).

The practical takeaway for local teams: invest in data, open orchestration, and governance now (interoperability and auditability are repeatedly flagged as deployment blockers) so Raleigh can turn compute investments and local pilots into measurable savings and faster, safer customer outcomes.

Raleigh teams should act now to invest in data, orchestration, and governance to realize these benefits in 2025.

Raleigh financial teams moving from pilots to production in 2025 are gravitating toward enterprise‑grade building blocks that deliver fast, auditable reasoning and flexible deployment: NVIDIA's Nemotron family offers open, high‑accuracy reasoning models that think up to 9x faster and can cut reasoning costs by roughly 60%, making agentic assistants and retrieval‑augmented workflows more affordable at scale; Nemotron excels at instruction following, tool calling, visual reasoning and is available as enterprise‑ready NIM microservices or via DGX Cloud for on‑prem or cloud launches (NVIDIA Nemotron family of reasoning models).

For firms needing advanced vision or synthetic data for model fine‑tuning, NVIDIA Cosmos provides world foundation models, tokenizers and data pipelines that accelerate video and visual‑language workloads - useful where large document sets or multimodal records meet compliance and auditability requirements (NVIDIA Cosmos world foundation models and data pipeline).

The practical payoff for Raleigh: proven enterprise tooling, multi‑cloud delivery options (Bedrock/SageMaker, Azure/Vertex), and microservice deployment paths that let teams iterate quickly while keeping data control and governance front and center - so AI moves from an experiment into a repeatable, cost‑controlled operational capability.

Major capital is arriving in North Carolina in 2025, and Raleigh's financial ecosystem stands to benefit from the spillover: Amazon's plan to invest $10 billion in an AWS cloud and AI innovation campus in Richmond County will add the compute, storage, and networking that regional banks and fintechs need for generative AI workloads and at least 500 high‑paying jobs - announced as one of the largest single capital investments in state history (Amazon $10B AWS AI campus in North Carolina); alongside that, major corporate moves like Schneider Electric's planned U.S. investment (which includes a new Robotics & Motion Center in Raleigh) are boosting energy and automation capacity that data centers and AI operations depend on (Schneider Electric $700M+ U.S. expansion and Raleigh robotics center).

The practical takeaway: these projects deliver not just servers but workforce pipelines, energy resilience, and local supplier opportunities - turning Raleigh's pilot‑scale AI experiments into infrastructure‑backed production systems capable of supporting regulated financial services at scale.

“Artificial intelligence is changing the way we work and innovate, and I am pleased that North Carolina will stay at the forefront of all that's ahead as we continue to attract top technology companies like Amazon.” - Governor Josh Stein

Raleigh's finance teams should treat infrastructure as a strategic playbook for 2025: hybrid IT and colocation are the practical path to run latency‑sensitive AI (think real‑time fraud detection and model inference) while keeping costs and data control in check, especially as regional capacity tightens and hyperscalers push compute into the Southeast - Amazon's $10B AWS campus in Richmond County is a case in point (Amazon $10B AWS campus in North Carolina announcement).

Industry research shows hybrid adoption is effectively universal (98% planning or using hybrid models) and enterprises are migrating generative and augmented AI workloads into colo for performance, cost predictability and interconnection benefits (CoreSite hybrid IT and colocation trends report); meanwhile the Flexential 2025 State of AI Infrastructure survey finds 90% of organizations deploying generative AI and flags infrastructure constraints as a top barrier (44%), with many teams now planning capacity 1–3 years ahead to avoid being locked out of GPU, power, and space.

Practical implications for Raleigh: prioritize colo relationships with robust cloud interconnects, map workloads by latency and egress cost, design for higher rack power densities (now commonly ~17 kW) and build governance that lets compute scale without surprise bills - because in AI, a missed server lease or a month‑long procurement delay can shave months off a project's ROI.

“Artificial intelligence is changing the way we work and innovate, and I am pleased that North Carolina will stay at the forefront of all that's ahead as we continue to attract top technology companies like Amazon.” - Governor Josh Stein

For Raleigh's banks and fintechs, improving customer and employee digital journeys isn't optional - it's how to protect deposits, cut support costs, and speed approvals; Pendo's suite of in‑app analytics, guides, and feedback tools (Pendo is even “Made in Raleigh, NC”) lets teams map the exact paths top performers take, then replicate those paths for everyone so onboarding stops being a lottery and becomes a predictable outcome - for example, a long 11‑step guide with only a 5% completion rate can be replaced by short, segmented flows that hit 45–70% completion in real deployments (Pendo banking onboarding white paper for improving bank client onboarding).

That same data‑driven approach reduces time‑to‑value (82% of customers see ≥10% effort/time savings) and can boost feature discoverability and compliance across homegrown apps - the practical payoff for Raleigh teams: fewer support calls, faster loan processing, and in‑app nudges that turn hesitant users into confident ones.

Start by instrumenting funnels, surfacing NPS and qualitative feedback, and shipping short, role‑based guides; Pendo's financial‑services playbook and case studies show how to do it without heavy engineering lift (Pendo solutions for financial services to improve product adoption and compliance).

“With Pendo, we will be able to drive operational efficiency through self-service and always-on training in our business critical workflows.” - Pendo customer testimonial

Regulatory risk is no longer a side conversation for Raleigh's finance teams - 2025 has turned AI compliance into a board-level priority as federal and state rules advance in parallel, and the choices North Carolina makes now will affect access to federal incentives and even where firms locate compute and talent.

State legislatures are prolific - NCSL documents that every state introduced AI bills in 2025 and that dozens enacted measures this year - so Raleigh organizations must watch local bills (and pending NC measures) closely and build governance that's portable across jurisdictions (NCSL Artificial Intelligence 2025 legislation).

At the federal level, new White House priorities and agency scrutiny are reshaping risk: America's AI Action Plan signals incentives for states that ease regulatory friction while agencies like the SEC, FTC and CFPB keep AI, cybersecurity and consumer harms squarely on their enforcement lists - enforcement has led to six‑ and seven‑figure penalties already, so governance failures can be expensive (America's AI Action Plan overview from Consumer Finance Monitor).

Practical playbook items for Raleigh firms: inventory models and training data, require meaningful human review for high‑risk automated decisions, log provenance for retraining and audits, and build an intake workflow that ties risk tiers to review gates - approaches featured in recent governance webinars and vendor playbooks that help future‑proof programs (OneTrust webinar on navigating the evolving US AI regulatory landscape).

The so‑what: with states experimenting like regulatory sandboxes and federal funding nudging policy choices, Raleigh teams that treat governance as product infrastructure - not paperwork - will be best placed to scale safe, auditable AI without losing grant eligibility or inviting regulator scrutiny.

“reassert American leadership in artificial intelligence,” - America's AI Action Plan (Consumer Finance Monitor)

Raleigh's AI readiness depends as much on talent as on compute, and local upskilling programs have been moving from theory to practical CPE that finance teams can use tomorrow: the North Carolina Association of CPAs (NCACPA) runs focused courses - like the four‑credit “Artificial Intelligence for Accounting & Financial Professionals” that translates AI into everyday tools (including Excel) - and short workshops such as “ChatGPT & the ELI5 Approach” and a Copilot session that teach concrete prompts and workflows for CPAs; explore these offerings on NCACPA's Artificial Intelligence for Accounting & Financial Professionals course, the NCACPA Current & Emerging Technologies Conference (CETC) information, or request tailored on‑site programs via NCACPA's Customized Continuing Education programs.

The city's training ecosystem is built to be flexible - from an 8‑hour CETC lineup with A‑list speakers to 2‑ and 4‑hour modules - and NCACPA's Raleigh office at 2700 Wycliff Road helps anchor employer partnerships and customized curricula that boost retention (NCACPA notes 94% of employees stay longer when employers invest in career growth).

For finance leaders in 2025, the practical play is simple: map role‑level skills to short, credit‑bearing modules (Copilot, ChatGPT, ethics/data analytics), use NCACPA's customized training to close gaps, and keep a steady pipeline of federally‑aligned CPE so teams can scale safe, auditable AI without pausing client work.

Practical AI in Raleigh's financial shops begins by picking one high‑friction workflow - lending, onboarding, or document‑heavy mortgage processing - and treating it like a product: instrument data, define success metrics, and run a short human‑in‑the‑loop pilot that proves value before scaling.

Start small with a vendor that knows banking workflows (nCino's Banking Advisor and Continuous Credit Monitoring are built to slice document queues and surface at‑risk relationships), or deploy a digital origination stack like Blend to speed account opening and loan quotes and hit production fast; real‑world pilots report dramatic wins (one nCino customer cut document processing by 74%).

Make data quality the first project - bad inputs mean bad models - then wire agents and assistants to the systems of record so outputs are auditable and easily reviewed.

Measure cycle time, queue depth, and customer abandonment; iterate with role‑based human review gates to keep risk proportionate; and, when the pilot proves the ROI, expand with automation playbooks and governance tied to those same metrics so scaling doesn't outpace controls.

These steps - target, instrument, pilot, measure, govern, scale - turn AI from a catchy experiment into repeatable savings and safer customer outcomes for Raleigh institutions.

“a new era in financial services” - Sean Desmond, nCino

Ready-to-run next steps for Raleigh financial teams: treat AI adoption as a program, not a project - start by inventorying models and data, codifying human-review gates for high‑risk decisions, and pairing short pilots with measurable success metrics so wins can scale under audit; with Gartner predicting roughly 90% of finance functions using AI by 2025 (Triangle Business Journal report on AI adoption in finance), the time to build governance and workforce pipelines is now.

Use local guidance (NC State's practical AI best practices and approved‑tool rules are a great place to align data classification and responsible use) (NC State AI guidance and best practices for responsible AI use), and lock in role‑based training so employees move from transactional work to oversight and strategy.

For teams that need structured, job‑focused curriculum, a short, nontechnical pathway like Nucamp's 15‑week AI Essentials for Work bootcamp turns promptcraft and tool fluency into repeatable productivity gains - ideal for banks and fintechs shifting pilots into production (Nucamp AI Essentials for Work bootcamp - 15-week program and registration).