Purpose-Driven, Responsible and Directed: Approach to AI Critical for Maximizing Potential, Value

The rate of advancements and adoption in artificial intelligence (AI) are downright jaw-dropping and unprecedented. Globally, a number of sources have reported that 94% of companies use AI in at least one business function. Resourca paints a vivid picture—of the estimated 444 million companies around the world, 417-plus million have incorporated it or are actively investigating how it can be leveraged to their advantage.

According to recent statistics from Axis Intelligence, global AI investment eclipsed $581 billion in 2025, a 130% year-over-year bump. Even with the pervasiveness of adoption, just 39% of companies report enterprise-level EBIT impact. One of the more significant and compelling trends cited by Stanford HAI’s 2026 AI Index report found that “Generative AI reached 53% global population adoption within three years of mainstream availability.” To put it in perspective, it outpaced the adoption of the personal computer and the internet.

There may be cause to pump the generative brakes, so to speak. The same Axis report, citing Forrester/Anaconda 2026 data, found that 88% of AI agent pilots never reached production in 2026. The primary stumbling blocks were evaluation gaps (64%), governance friction (57%) and model reliability (51%).

In the interest of self-reporting, we asked Microsoft Copilot to share what it believed to be AI’s greatest advances in the past 12 months. Here’s what it offered:

AI made its biggest leap this past year by shifting from passive chatbots to autonomous, reasoning-capable agents. Open-source models like DeepSeek R1 matched top proprietary systems in complex reasoning tasks, democratizing high-end AI capabilities and accelerating global competition. Google advanced multimodal intelligence with Gemini 3, delivering major gains in reasoning, efficiency and cross-modal understanding. The Keyword Meanwhile, OpenAI and others pushed agentic workflows into the mainstream, with AI systems increasingly able to plan, act and interact with tools—reshaping enterprise automation.

Video generation also hit a milestone with OpenAI’s Sora, producing highly realistic, coherent 60-second scenes from text prompts and transforming creative industries. Across sectors, AI became deeply embedded in products, from safety systems to enterprise operations, marking a transition from experimentation to infrastructure.

On the other side of the coin, we turned to Anthropic’s Claude AI for an overview of the year’s top challenges:

Businesses face several significant hurdles when adopting AI. Data quality and availability remain foundational challenges—many organizations lack the clean, well-structured data needed to train reliable models, or struggle with fragmented systems that keep data siloed. Integration with legacy infrastructure adds further complexity and cost, as older systems were rarely designed with AI in mind.

The talent gap is equally pressing. Skilled professionals in machine learning and data science are scarce and expensive, leaving many companies unable to build or maintain solutions internally. Trust and explain-ability concerns also slow adoption, particularly in regulated industries where “black box” decisions are unacceptable.

Beyond the technical, businesses grapple with unclear ROI, making it hard to justify investment. Security, privacy and regulatory compliance create additional risk. Finally, organizational resistance and cultural inertia often undermine initiatives, as employees fear displacement or simply distrust unfamiliar tools. Successful adoption requires addressing both technological and human dimensions thoughtfully.

This article kicks off the July State of the Industry report on AI, although as you’ll see, it’s one of several we’ve baked into this month’s editorial package. Throughout the year and into the foreseeable future, we’ll be offering more content centered on developments, guidance and best practices. We encourage you to absorb all the content possible via articles, videos, webinars, in-person presentations—there’s a cornucopia of information available online. For now, take a look at the AI journeys embarked on by your fellow office technology dealerships.

Google-ish Quality

The early days of AI were reminiscent of Google, notes Sam Stone, president of Stone’s Office Equipment in Richmond, Virginia. A healthy mix of skepticism and curiosity guided the dealer’s foray into AI four years ago, and it was leveraged on a superficial level—assistance in drafting marketing content, enhancing email tone, ferreting out social media ideas and quickening the pace of repetitive administrative work. Claude was essentially an AI-generated intern, minus the ability to pick up lunch or dry cleaning.

“What changed for us was when we realized AI wasn’t just a novelty tool — it was a force multiplier,” Stone noted.

From an internal standpoint, Claude’s current role at Stone’s Office Equipment has grown to address:

• Marketing content creation and SEO blogs
• Proposal and agreement drafting
• Financial modeling and forecasting
• KPI dashboards and reporting
• Workflow documentation and process improvement
• Brainstorming operational efficiencies
• Training outlines and job descriptions
• Customer communication refinement

We treat AI like a calculator for knowledge work: incredibly powerful but dangerous if you stop checking the math.

– Sam Stone, Stone’s Office Equipment

From a client-facing perspective, AI has facilitated clear and consistent communications for Stone’s, whether it’s copier-driven or hydration-focused courtesy of its Pure Water Virginia partnership. One of the great parts is its complementary fit with employees, and not as a substitute. Stone feels its greatest benefit is removing friction.

“AI gives our team leverage. It helps us move faster, communicate better and spend more time solving real customer problems instead of staring at blank screens or reinventing documents,” he said. “And frankly, in a family-owned business environment where everyone already wears five hats, getting a sixth invisible assistant is pretty useful.”

ChatGPT joins Claude as the dealer’s independent AI tools, in addition to applications where the technology is baked into platforms such as CEO Juice. Stone’s follows an “AI assists, humans approve” mantra. It’s not enabled to operate autonomously in sensitive areas such as financial decisions, contract approvals, HR matters, customer-specific commitments or legal/compliance-sensitive communications. Elements that are customer-facing or operationally critical require human review.

Other staples that govern Stone’s AI applications include:

• No confidential customer information entered into public AI systems
• AI-generated content must be fact-checked
• Employees should use AI to enhance expertise, not bypass thinking
• Internal transparency about where AI is being used

“We treat AI like a calculator for knowledge work: incredibly powerful but dangerous if you stop checking the math,” Stone added.

While the AI push is leadership driven, it’s also highly collaborative, he added. Stone has taken the lead in exploring how AI can improve operations, marketing, customer communication and strategic planning. Multiple departments are engaged so that it’s not approached as an IT project.

There’s an organic quality to how AI has been blended into the dealer’s quest to address tangible challenges, whether it’s sales, admin, service or marketing. Stone encourages “experimentation with accountability,” and the driving motivation should always be its application in solving problems, not following a trend.

“There’s a huge difference between innovation and chasing shiny objects,” Stone added. “One builds businesses. The other builds conference keynote slides.”

With Intent

Eakes Office Solutions traipsed down the AI road in a manner matching many other firms, adopting it in a broad and decentralized way. It employed tools such as ChatGPT, Copilot and Gemini to help beef up email creation, refine ideas and organize thinking while also devising early concepts for strategy and problem-solving. That’s given way to a more intentional approach, notes Nate Schaf, chief technology officer for the Grand Island, Nebraska-based dealership.

According to Schaf, Eakes has targeted use cases that feature efficiency, speed and decision support, and the dealer has consolidated most AI activity into trusted, internal platforms. “The strongest returns are coming from code development, business process automation, pricing analysis and data comparison work that previously required significant manual effort,” he said. “We’re also using AI to improve prospecting and cross-sell efficiency to our customer base.”

The go-to platform for Eakes is Copilot as part of the Microsoft 365 ecosystem because it furnishes strong data controls and security guardrails. It’s particularly critical for handling customer, pricing and contract data. Unmanaged use of external large language models is frowned upon, and the dealer monitors for compliance.

Regular collaboration has accelerated learning, shortened adoption cycles, and kept our efforts anchored in real business value, which can be a struggle if AI initiatives are siloed.

– Nate Schaf, Eakes Office Solutions

Embedded AI capabilities in tools such as HubSpot (Breeze AI), Excel Power Query, Power Automate and Sharp’s Synappx platform are leveraged by Eakes. Agent development falls to manager-level oversight within the various departments. As a result, the dealer is assured that solutions dovetail with operational needs while being centrally governed.

Rather than anointing a point person, Schaf notes that Eakes has developed a cross-functional AI committee. The group is responsible for evaluation, prioritization and knowledge sharing, and it accomplishes this by incorporating technology, operations, service, sales and leadership. It’s another example of guaranteeing AI initiatives go hand in hand with Eakes’ operational mission.

“One of the committee’s key roles is identifying where individuals or teams are using AI in unique, effective ways and then finding opportunities to share and scale those practices across the organization,” Schaf said. “A real challenge with AI today is that a lot of great use is happening organically, but companies don’t always know it’s going on. By creating visibility and encouraging knowledge sharing, we’re able to turn isolated wins into broader capability.”

It’s helped to stave off rogue experimentation as well. “Regular collaboration has accelerated learning, shortened adoption cycles, and kept our efforts anchored in real business value, which can be a struggle if AI initiatives are siloed,” he added.

Purpose Driven

As impressive and awe-inspiring as AI has proven itself to be, dealers and their customers need to temper that enthusiasm and employ the same MO as they would with any other newer technology that has the propensity to change how business is done. While enterprise-level clients have stepped up their AI game and devised impressive applications, Spectrum Technologies saw a golden opportunity to assist small- and medium-sized business (SMB) clients. The challenge for SMBs, notes Jake Elliott, vice president and CRO for the El Paso, Texas-based firm, is to parlay that power into bona fide improvements in processes, workflows and day-to-day operations.

While Elliott knows it’s easy to get caught up in the AI hoopla, there’s a true danger in spending “$100,000 solving a $5,000 problem.” He’d seen real-life examples. That’s why Spectrum adopted the philosophy of approaching AI with the goal of tying it to a clear business case, process improvement or ROI.

“Initially, we approached the market fairly broadly,” Elliott said. “We were helping customers think through everything from marketing use cases to more complex operational initiatives. Over time, we realized we couldn’t be all things to all people. We needed to become much more focused on where we could create the most meaningful value.”

It was also important for Spectrum to heed its own admonishment and embrace AI in a similar manner. It’s being used to transform the dealer’s contact center, allowing clients to more immediately address routine needs such as ordering toner, checking unpaid invoices, requesting service and accessing the same information that Spectrum’s internal team has traditionally handled.

“That’s been a strong proving ground for us because we understand our own customer interactions, our internal systems, the metrics we’re trying to improve and the operational pain points,” Elliott noted. “It’s allowed us to build AI around real business context rather than theoretical use cases.”

The upshot has been enhancements to efficiency, consistency and the customer experience. AI takes the friction out of repetitive tasks and allows people to focus on more core objectives, particularly those where judgment, relationship building and problem solving truly matter.

Anthropic and OpenAI tools are primary to Spectrum’s efforts, with variables such as data sensitivity levels, system connections and oversight requirements dictating its choice. With an eye toward security, control, transparency and fit for purpose, team members can leverage AI tools for general productivity applications—drafting communications, organizing ideas, summarizing information and the like.

Anything pertaining to sensitive client data, operational systems, financial information or automation that could impact a customer experience calls for a more controlled framework. On a systems level, Spectrum has designed permissions, integrations and escalation paths for prompts to ensure AI knows where it can impact, where it can’t and situations where a human needs to be added to the equation.

A collaborative personnel structure governs Spectrum’s AI endeavors, with a small group dedicated to helping drive innovation and pinpoint opportunities where it can create meaningful value. It’s important, Elliott says, to involve key people from the various departments who are actually doing the work. They, in turn, can help ensure applications mesh with how the business operates.

“The people closest to the work understand the exceptions, the customer nuances, the manual steps and the reasons certain processes exist,” Elliott added. “Bringing them into the conversation early helps us avoid building something that looks impressive but doesn’t solve the real problem. We also involve process-focused team members who are constantly looking for areas where work is highly manual, repetitive or a good candidate for improvement. From there, we pair the business opportunity with the right technology approach.”

We were helping customers think through everything from marketing use cases to more complex operational initiatives. Over time, we realized we couldn’t be all things to all people.

– Jake Elliott, Spectrum Technologies

Taking Charge

In the past two years, as AI has garnered momentum from a mainstream perspective, dealerships such as Braden Business Systems saw the importance of getting out in front of how it’s introduced within the company and setting standards to govern its internal use by team members. The Fishers, Indiana-based dealer knew employees were experimenting with the technology on the consumer end, so CEO and Managing Partner Erik Braden wanted to foster responsible adoption. Given the dealer is held accountable for client data, print fleets, networks and their entire security infrastructure, that shaped Braden’s approach to the tech.

Braden felt it was important to be the test subject before rolling out any tools to clients. That would help the dealer understand the security implications, data-handling realities and day-to-day friction of introducing something new to a team. By doing the taste test, he felt it lent credibility during the course of client conversations.

The dealer’s AI strategy addresses three main areas. For productivity, Copilot is embedded throughout the team for drafting communications, meeting summaries and mundane-but-necessary email/document work that’s generally a time burgular. Claude Enterprise is the go-to for advanced reasoning and analysis work such as long-form document review, proposal development, deeper research and complex client scenarios. The third area is service delivery—diagnostics, ticket triage and documentation across managed IT and managed print practices.

“We chose that combination because the strengths line up cleanly,” Braden said. “Copilot wins on ambient productivity and Microsoft integration. Claude Enterprise wins on depth, document handling and the kind of analysis we used to assign to a senior team member with a free afternoon. We deploy and support Microsoft 365 Copilot for our clients because that’s what fits most client environments, and we’ve built deep capability around it.”

Braden sees the company’s customer-facing role with AI is enabling it to safely adopt tools. The sticking point for many end-users, he said, is that their data environment, permissions and security posture aren’t ready to support Copilot at scale. It falls upon the dealer to remove that blocker.

“The most benefit we’ve seen, honestly, isn’t productivity gains. It’s risk reduction,” Braden said. “Clients who go through a thoughtful AI-readiness process avoid the kinds of data exposure incidents that are increasingly showing up in industry reporting. For dealers reading this, my honest advice is that the productivity story sells AI, but the governance story is what keeps clients out of trouble. Both matter. Lead with whichever one your client is actually feeling the pain of.”

The honest truth, and dealers in our channel know this, is that shadow AI isn’t a future problem. It’s already inside almost every business, including the ones that think they have it under control.

– Erik Braden, Braden Business Systems

The dealer’s governance is fairly straightforward. Client data, personally identifiable information or anything covered by HIPAA, financial regulations or contractual confidentiality can’t be entered into any AI tool that hasn’t been vetted and approved for that purpose. Braden noted they extend the same identity and access standards to AI that they apply to every other system, such as multifactor authentication, least-privilege access and visibility into how the tool is being used. Consumer AI accounts can’t be used in conjunction with company or client work.

Braden follows a three-step approval process. Any new AI tool recommended by a team is vetted through an internal review that covers data handling, security posture and redundancy. The dealer has a crystal-clear acceptable use policy that outlines what is/isn’t permitted and offers guidance for gray areas. Lastly, recurring and practical training helps inform user behavior.

“The honest truth, and dealers in our channel know this, is that shadow AI isn’t a future problem,” Braden said. “It’s already inside almost every business, including the ones that think they have it under control. Our job, for ourselves and for our clients, is to give people approved, capable tools so they don’t need to reach for unsanctioned ones.”

Like its contemporaries, Braden Business Systems is adopting AI through a small cross-functional working group consisting of leadership from managed IT and security practices, the office technology and managed print side, operations and marketing. Braden sees a danger in appointing a so-called evangelist, as adoption “becomes about that person’s energy rather than the company’s strategy,” he added. “The moment they leave, change roles or burn out, the program stalls. I’ve seen that pattern in other dealerships, and we wanted to avoid it.”

Braden felt the departmental blend was necessary. It helps stave off moving too fast or slow with application adoption.

“Security ensures we’re not chasing productivity at the expense of risk,” he said. “Operations grounds the conversation in what actually saves time on the service side and at the dealer level. Marketing keeps us honest about how we talk to clients about what we’re doing. And the working group reports up to me, which means AI decisions are tied to the same business priorities as everything else we invest in.”

Desired Outcomes

Stargel Office Solutions eased into AI by employing it as a writing and brainstorming assistant to draft social media captions, emails, campaign ideas and basic marketing content more efficiently (ChatGPT is its platform of choice). It’s quickly taken on a more impactful role, supporting SEO planning, website content optimization, social media strategy, competitive positioning, proposal language, customer communication, internal process development and sales enablement.

T.J. DeBello, vice president of sales at the Houston-based dealer, points out that AI has been a boon in transitioning from idea to execution. “It assists with organizing complex information, turning technical product or service details into clear messaging, developing campaign frameworks, summarizing research and creating first drafts that our team can refine,” he said. “Externally, it helps us communicate more effectively with customers by improving the clarity, consistency and professionalism of our marketing and sales materials.”

[AI] can accelerate research, structure thinking and improve drafts, but we still verify facts, review tone and make sure anything published aligns with our brand and business goals.

– T.J. DeBello, Stargel Office Solutions

One of the greatest benefits Stargel has reaped is the addition of speed without subtracting (or sacrificing) strategy. AI, Stargel notes, allows the dealer to arrive at a stronger starting point faster. Final judgment, positioning and approval, however, are dictated by team members who understand the customers, markets and Stargel’s brand.

The dealer’s position is that AI isn’t a substitute for human review or final approval. All applications that are customer-facing, brand-related, technical, financial or operational must be reviewed by the appropriate person before it’s employed. Like most prudent users, Stargel avoids entering sensitive customer information, confidential financial details, private employee information or proprietary data unless it is appropriate and necessary.

“We treat AI as a support tool, not an authority,” DeBello added. “It can accelerate research, structure thinking and improve drafts, but we still verify facts, review tone and make sure anything published aligns with our brand and business goals.”

Leadership and marketing have been governing the manner in which AI is applied throughout the organization and broached with clients. DeBello has been exploring how the technology can bolster marketing, sales communications, SEO, proposal development and internal efficiency. The common denominator across the network is focusing on tying usage to business outcomes—saving time, enhancing quality, supporting sales, augmenting customer communication and taking efficiency to the next level.

The technology may be fun, even exciting and brimming with possibilities. Without purpose, however, it’s just the next shiny, new object.

“That [business outcomes] mindset has made adoption more approachable,” DeBello added. “Instead of presenting AI as a major operational overhaul, we’ve introduced it as a tool that can help employees complete familiar tasks faster and with more structure.”