Article
From Charlotte's financial district to the Research Triangle and Greenville's manufacturing corridor, businesses across North and South Carolina are putting AI to work. Here is what is actually happening on the ground.
The Carolinas do not look like a stereotypical AI story. There is no San Francisco press release, no billion-dollar model launch. What there is: a banking hub that processes more transactions than most countries, a research corridor that has been quietly doing computational science for forty years, a manufacturing base reinventing itself, and a growing number of businesses that have stopped asking “should we do AI?” and started asking “what do we do first?”
We work with businesses across North and South Carolina. The patterns below are what we are actually seeing, not what the conference decks promise.
Charlotte is the second-largest banking center in the United States. That fact shapes the AI conversation here in a specific way. The banks themselves — Bank of America, Truist, Wells Fargo — are running enterprise-scale AI programs with hundreds of engineers. Their vendor ecosystems and regional partners feel the downstream pressure.
Compliance documentation is the first place most Charlotte financial services firms are deploying AI. Regulatory filings, audit documentation, policy summaries. Not AI making the compliance decisions — that remains a human and legal judgment — but AI doing the first draft of a fifty-page document that used to take a paralegal four days. The turnaround compresses to a day. The paralegal reviews and corrects. The output is the same quality, delivered faster.
The second area is client communication at scale. Wealth management firms with hundreds of client accounts are using AI to draft personalized quarterly updates from portfolio data. The relationship manager reviews and sends. The relationship is still human. The drafting is not.
Microsoft 365 Copilot is the tool we see most often in this category, because the data stays inside the Microsoft tenant and satisfies most existing data governance requirements without a bespoke architecture.
In Raleigh, Durham, and Chapel Hill, the question is not whether to use AI — it is which generation of AI tools to migrate to. Research Triangle Park has housed computational biology, simulation software, and machine learning workloads since the 1990s. The Triangle’s tech and life sciences firms are now grappling with a different problem: how to bring the productivity gains of modern generative AI to the people who are not data scientists.
The shift we are seeing is AI moving out of the data team and into operations, finance, HR, and sales. Departments that have always handed off requests to data analysts are now running their own queries against structured data using natural language. Not every analyst job is displaced. The better analysts are now doing harder work, because the first pass at a dataset no longer requires their involvement.
For life sciences companies in the Triangle, document-heavy workflows are the primary target. Clinical trial protocols, regulatory submission packages, standard operating procedures. The structured, formal nature of these documents actually makes AI more useful, not less — the genre conventions are consistent, the expected sections are known, and the review checklist is clear.
Azure OpenAI Service is the common infrastructure choice for Triangle companies that need to keep data on-premises or within a defined Azure boundary, which is most of them in life sciences.
Greenville and the Upstate South Carolina manufacturing corridor — BMW, Michelin, and their supplier ecosystem — represent a different kind of AI deployment. The use cases are less about knowledge work and more about operational continuity.
Predictive maintenance is the most mature application. Sensor data from equipment feeds into models that flag anomalies before they become failures. The technology has been industrially deployed for over a decade in some form, but the cost of the infrastructure and the expertise required has dropped sharply. A mid-size precision manufacturer that could not have justified this in 2020 can deploy it today on Azure IoT Hub and Azure Machine Learning for a fraction of the previous barrier.
The second application is quality control. Computer vision systems on production lines that flag defects at camera speed, reducing the dependency on manual visual inspection for high-volume production runs. The model gets trained on images of acceptable and defective parts. The accuracy targets for this vary by what the part does and what a defect costs downstream.
Neither of these is new technology in the abstract. What is new is the total cost of deployment, the availability of managed cloud services that handle the infrastructure, and the talent pool in Greenville that can now support these systems without flying in a specialist.
Three things stand out across the region, different from the national AI narrative:
The speed of adoption is faster at the edges. The large enterprises — the banks, the major manufacturers — move slowly because they have governance, procurement, and risk management processes that exist for good reasons. The most interesting AI deployments we see are at businesses with 50 to 500 employees, where the decision to try something and the resources to act on it can align in weeks instead of quarters.
Data readiness is the real constraint, not model access. Every business that has struggled to get value from AI has the same problem: the data that would make AI useful is in spreadsheets, locked in legacy systems, or in someone’s email inbox. The AI is not the bottleneck. Getting clean, queryable data into a place where AI can reach it is. This is unglamorous work. It is also the work that determines whether you get a return on the AI investment or not.
Microsoft is the default entry point for most Carolina businesses. The existing Microsoft 365 and Azure relationships mean Copilot and Azure OpenAI have shorter procurement and security review cycles than alternatives. That is not a universal endorsement of Microsoft as the best AI platform for every use case — it is an observation about where most organizations are starting and why.
A few things that get pitched to Carolina businesses and do not, in practice, pay back in the near term:
If you are a business in North or South Carolina and AI is on your agenda for 2026, the sequence that works:
First, identify one process where time-to-output matters, the inputs are relatively consistent, and the quality bar is reviewable by a human. Drafting. Summarizing. First-pass categorization.
Second, get the data into a shape where the AI can reach it. This step is usually longer than expected and less exciting than expected.
Third, pilot with a small team, measure the actual time saved, and make the next decision based on that data rather than the promise.
The Carolinas business environment — practical, industry-grounded, skeptical of hype — is well suited to this approach. The AI moment is real. The way to capture it is the same way you capture any operational improvement: one specific thing at a time.
Devsoft Solutions works with businesses across North and South Carolina on Microsoft 365, Azure, and AI implementation. If you are evaluating where AI fits in your operations, get in touch.
— Continue reading
Most Copilot rollouts skip the measurement step, which means nobody can tell whether the per-user license cost is paying back. A six-week plan with the metrics that matter, and the ones that look impressive but mean nothing.
Both run on Microsoft's stack. Different audiences, different strengths, different cost models. A practical decision guide for picking between them on a custom AI project, with the hybrid pattern that ends up working for most teams.
Skip the AI hype cycle. Four use cases where small and mid-market businesses are getting concrete returns from AI right now, and one category where they are mostly still wasting money.
— Related services
Microsoft Copilot rollout, custom copilots with Copilot Studio, and Azure OpenAI integrations for the parts of your business where AI actually pays back.
Tailored Microsoft 365 plans, deployment, and ongoing support — designed for businesses that need reliable productivity infrastructure without the in-house overhead.
Azure infrastructure, Microsoft 365 environments, and the operational glue between them. Right-sized, secured, and supported.
