By Shree Amujala, Founder & CIO, Raise Networks. Shree brings 30+ years of IT leadership experience and advises growing companies as a Fractional CIO on cloud, security, and AI adoption.
Key takeaways
- Most small business owners are already using or testing AI. The real gap is between buying licenses and changing how work gets done.
- Businesses move through four stages of AI maturity: hidden AI use, licenses without a plan, workflow driven, and connected and governed.
- The biggest wins come from one recurring workflow at a time, not a company wide launch.
- Simple guardrails, such as approved business accounts, no payroll or HR data, and human review, speed adoption up rather than slow it down.
- AI amplifies your IT foundation, so identity, access, and data ownership are part of the adoption work.
For decades, small businesses followed a familiar pattern with new technology. Large enterprises went first, spent the money, absorbed the risk, and worked out the kinks. Everyone else waited until the tools got cheaper and simpler.
AI has broken that pattern.
The questions we hear from small and growing businesses have changed noticeably this year. It used to be “Should we be looking at AI?” Now it sounds more like “Half our team is already using ChatGPT or Claude. How do we get this under control and actually make it useful?”
That is a very different conversation. It is no longer about picking a tool. It is about how your business runs.
The numbers back it up
What we are seeing with clients shows up clearly in the research.
- Most owners are already in. Bluevine’s 2026 Small Business AI Trends Report found that 74% of small business owners are using or actively testing AI. ChatGPT and Gemini lead in reported usage, with Copilot and Claude behind them.
- Momentum is shifting fast. The same report found Claude’s user base among Bluevine customers grew 729% year over year, and by April 2026 Claude had passed ChatGPT in total monthly transactions among those customers. Owners are not just trying AI, they are choosing tools deliberately and using them more heavily.
- Buying is not the same as using. A Business.com study found 57% of U.S. small businesses are investing in AI, up from 36% in 2023. Yet only 30% of employees use it daily.
That last gap is the one that matters most. Access is easy. Changing how work actually gets done is the hard part.
The AI vendors have noticed too. Anthropic recently launched a small business offering built around Claude Cowork, with integrations into tools small businesses already rely on, like QuickBooks, DocuSign, and HubSpot. Small businesses account for 44% of U.S. GDP, and the market is finally building for them.
ChatGPT, Claude, or Copilot: which fits your small business?
Most owners start with ChatGPT for small business tasks, and for good reason. It is the most widely used assistant and handles everyday writing, research, and brainstorming well. But the right choice depends on the work, not the brand.
- ChatGPT is a strong all round assistant for drafting, research, customer communication, and quick analysis. Its business plans add admin controls and connectors to tools like Google Drive and SharePoint.
- Claude stands out for long documents, careful writing, contract and policy review, coding, and multi step workflows connected to business apps such as Microsoft 365, Google Workspace, QuickBooks, DocuSign, and HubSpot.
- Microsoft Copilot makes the most sense if your team lives in Microsoft 365, since it works directly inside Outlook, Word, Excel, and Teams with your existing permissions.
Many small businesses end up using more than one. What matters is that every tool runs on approved business accounts, connects to your systems with the right permissions, and supports specific workflows your team actually uses. That is where most of the value, and most of the risk, sits.
Where is your business today? The four stages of AI maturity
Across our client work, we see businesses fall into one of four stages. Knowing your stage matters more than picking the “best” tool, because each stage has a different problem to solve.
Stage 1: Hidden AI use. Employees are using personal accounts on their own, and leadership has no visibility into what is being pasted into a chat window. Research from BlackFog, covered by Forbes, found 49% of workers use AI in ways their employer has not approved. This is not a discipline problem. It is a clear signal that your people want these tools, paired with a real data risk.
Stage 2: Licenses without a plan. The business buys seats, shares a getting started guide, and waits for adoption to happen. It usually doesn’t. A few people experiment, most don’t, and eventually someone concludes that “AI didn’t really work for us.”
Stage 3: Workflow driven. AI is mapped to real, recurring work. A small number of specific workflows are built, tested, and proven before expanding. This is where the time savings become real.
Stage 4: Connected and governed. AI is integrated with the systems the business already runs on, with identity controls, permissions, audit trails, and clear ownership. Few small businesses are here yet. Many will be pushed here by customers, regulators, or cyber insurers. We would much rather help them get there by design than after an incident.
Most of the demand we see today is businesses trying to move from Stage 2 to Stage 3.
What good adoption actually looks like
The most successful rollouts we have run are not flashy. They start with one real task that people already do every week.
In one engagement with a commercial real estate lender, the legal team used AI for a first pass review of leases and third party agreements. The tool flagged unusual clauses and explained why they deserved a closer look, while final judgment stayed with the attorney. The marketing team turned a single funded deal summary into a blog post, social copy, and an email blurb, then reviewed each piece before it went out.
None of this is dramatic. But it saves real hours every week, because the AI fits the job instead of forcing employees to become prompt experts.
Just as important, we set clear guardrails before expanding:
- No payroll or HR data goes into AI tools.
- A human reviews anything client facing or judgment based.
- Everyone works from approved business accounts with proper data protections, not personal logins.
Those rules did not slow anyone down. They gave employees confidence about what was allowed, and adoption picked up because of it.
A practical playbook for business owners
If your team is already experimenting with AI, here is where we recommend starting.
- Find out what is already happening. Before buying anything, get visibility into which tools your people are using and what data is going into them.
- Set simple, clear policies. A one page policy covering approved tools, off-limits data, and human review requirements goes a long way.
- Start with one workflow that matters. Pick a recurring task that eats up real time. Build it, prove it, then expand.
- Build security in from day one. Identity, access controls, and data protection are part of the adoption work, not something to bolt on later. Our AI-ready IT infrastructure checklist is a good place to start.
- Measure independence, not attendance. Workshop attendance and login counts tell you very little. The real test is whether an employee can run a workflow correctly on their own and teach it to a colleague. That is how AI adoption survives long after the training ends.
AI amplifies whatever foundation you already have
This is the part that often gets missed. AI does not fix a messy environment. It amplifies it. If identity and access are loosely managed, data is scattered, and nobody owns key processes, AI will make those problems more visible, not less. We covered this in more depth in Where AI fits, and where it doesn’t.
That is why the businesses getting the most out of AI treat it as a people, process, and governance challenge, not just a software purchase. Licenses take minutes to buy. Moving a team from scattered experiments to safe, repeatable workflows takes discovery, training, clear data decisions, and real follow-through, and most small businesses simply don’t have that capacity in house.
How Raise Networks helps
At Raise Networks, AI enablement sits alongside the rest of what we do: managed IT, cloud, cybersecurity, identity, and compliance. That matters, because safe AI adoption depends on all of it working together.
We help businesses assess where they are today, put the right guardrails in place, build practical workflows around real work through our AI workforce enablement program for Claude and ChatGPT, and connect AI securely to the systems they already use through AI automation and intelligent IT operations. Then we stay with them as the program grows, acting as an extension of their team rather than another vendor dropping off a license.
Small businesses no longer have to wait years for enterprise technology to trickle down. The lag is now measured in months. The businesses that win will be the ones that move quickly and safely at the same time.
Frequently asked questions
Should my small business use ChatGPT or Claude?
Both are strong choices, and many businesses end up using both. ChatGPT is the most widely used AI assistant among small business owners, while Claude has been growing quickly, especially for writing, document review, coding, and connected business workflows. The better question is which workflows you want to improve. We help clients pick the right tool per workflow, then standardize on approved business accounts so usage is visible and secure.
Can Claude and ChatGPT integrate with our existing IT stack?
Yes. Both offer business plans with connectors to common tools such as Microsoft 365, Google Workspace, and popular CRM, finance, and document platforms. The real work is deciding what data AI can access, setting permissions through your identity platform, and building workflows around the systems your team already uses. That is where most small businesses need help.
Is it safe to put company data into AI tools?
It depends on how you set it up. Personal accounts give you no visibility or control. Business plans for Claude and ChatGPT do not use your data to train their models by default and include admin controls. Pair that with a simple AI use policy, clear limits on sensitive data like payroll and HR records, and human review for client facing work.
What is AI workforce enablement?
AI workforce enablement is a structured program that gets employees using AI productively in their real daily work. It goes beyond a one time training session and includes discovery, hands on coaching, shared workflows, guardrails, and ongoing support so adoption actually sticks.
Where should a small business start with AI?
Start by finding out which AI tools your team already uses. Then set a one page policy, pick one recurring task that eats up real time, and build a secure workflow around it. Prove the value on that one workflow before expanding.
How does Raise Networks help small businesses adopt AI?
Raise Networks is a Technology Services Provider and managed IT partner based in San Jose. We assess where you are today, set up secure access and governance, run AI workforce enablement programs for Claude and ChatGPT, and connect AI to your Microsoft 365, cloud, and business applications. Because we also manage IT, security, and compliance, AI adoption fits into your whole technology stack instead of sitting on its own.
Is your team already using AI without a plan? Talk to Raise Networks about building a secure, practical AI adoption roadmap for your business.
