Most executives believe they’ve done their part on artificial intelligence. They’ve approved the tools, announced the initiative and moved on. But the adoption numbers tell a different story.

According to Slingshot‘s Digital Work Trends Report, 86% of C-suite executives believe AI usage is required in their company operations. Yet fewer than half (49%) of middle managers are reinforcing that expectation with their teams. This gap between what leaders announce and what employees actually do isn’t a technology problem. It’s a leadership one.

I’ve spent more than 35 years leading Infragistics, and one lesson which has remained true through every major technology shift is that the success of any new initiative depends less on the technology itself and more  on how leaders introduce it. The organizations that see lasting change are the ones whose leaders create an environment where people can adopt new tools with confidence.

AI is no different. If leaders want employees to embrace it, they need to do more than roll out a platform or issue a mandate. They need to create a safe space where people feel comfortable learning, experimenting and building new habits. Here are three ways leaders can do so. 

1. Coaching creates the confidence to experiment

Real AI adoption requires employees to experiment with the tools. But people won’t take those risks unless they feel safe doing so.

That’s where coaching becomes far more effective than command-and-control leadership. Rather than simply telling employees to use AI, coaching-oriented leaders work alongside their teams and ask what’s working, what isn’t and where people are getting stuck. 

Anyone who has spent time with an AI tool knows that getting genuinely useful results takes practice. The first prompt rarely gives you what you need. Over time, though, you learn to ask more specific questions, provide the right context and test different approaches until the output actually fits your workflow. 

That kind of learning is personal and iterative, and it looks different for every role. A marketer figuring out how to use AI to track KPIs is going to take a completely different path than a salesperson using it for outreach. Employees need room to go through that process, and that only happens when leaders create an environment where figuring it out is part of that job and not a sign that someone isn’t ready.

2. Model the behavior you want to see

One of the fastest ways to encourage AI adoption is for leaders to use it themselves. 

Employees pay far more attention to what leaders do than what they say. So, when leaders openly incorporate AI into meetings, planning sessions, decision-making or content creation — and are honest about both the successes and limitations — they normalize learning. And that transparency gives employees permission to experiment without feeling like they need to be experts from day one.

One simple habit leaders can implement is to open team check-ins by sharing how they used AI that week, what they tried, what worked and what didn’t, then inviting employees to do the same. Conversations like those are an opportunity to exchange ideas, uncover successful use cases, encourage collaboration and help employees learn from one another instead of experimenting in isolation.

3. Position AI as growth, not compliance

How leaders talk about AI matters just as much as how they implement it. When AI is framed as another mandatory technology rollout, employees often see it as another box to check or, worse, as a threat to their jobs. 

Slingshot’s Digital Work Trends report found that nearly 1 in 5 Gen Z employees (19%) and 17% of millennials worry AI could eventually replace them. A company mandate does nothing to address that fear. But when leaders position AI as a way to remove repetitive work, improve decision-making and give employees more time for higher-value thinking, that conversation starts to look very different.

Employees don’t want to hear that AI will replace what makes them valuable. They want to understand how it helps them become even better at the work they already do well. That means being clear about where AI adds value and where human judgment remains essential. AI can analyze data, summarize information and automate repetitive processes, but people still provide strategy, creativity, relationship-building and accountability. When those roles are clearly defined, AI becomes less intimidating and much more useful.

Ultimately, the organizations making the most progress with AI are the ones whose leaders make it safe to learn, model the behaviors they expect from others and consistently reinforce that AI is an investment in their people, not a replacement for them.

Most executives believe they’ve done their part on artificial intelligence. They’ve approved the tools, announced the initiative and moved on. But the adoption numbers tell a different story.

According to Slingshot‘s Digital Work Trends Report, 86% of C-suite executives believe AI usage is required in their company operations. Yet fewer than half (49%) of middle managers are reinforcing that expectation with their teams. This gap between what leaders announce and what employees actually do isn’t a technology problem. It’s a leadership one.

I’ve spent more than 35 years leading Infragistics, and one lesson which has remained true through every major technology shift is that the success of any new initiative depends less on the technology itself and more  on how leaders introduce it. The organizations that see lasting change are the ones whose leaders create an environment where people can adopt new tools with confidence.

Does saving $1,000 in a month seem impossible? Personal finance expert Jade Warshaw, who co-hosts The Ramsey Show, a popular financial talk radio program and podcast, says it’s not as difficult as it seems. 

Warshaw recently said in an interview with Fox Business that most Americans can put aside $1,000 if they are willing to sacrifice the little things and view their finances with a “scorched earth” mentality. “It’s more realistic than people think,” she said. “Most people are able to do it in 30 days.”

Warshaw said to start with building a budget. “If you don’t have a budget, it’s not going to work,” she said. That means calculating net monthly income, tracking expenses and making sure expenses are less than income. 

From that point, Warshaw suggests addressing the problem from two angles by both spending less and bringing in more. That may involve picking up extra shifts at work, starting a side hustle such as driving for Lyft or Uber or selling items online. 

At the same time, cutting back on routine spending, from streaming service subscriptions to dining out, can free up additional cash. 

Warshaw, who claims to have paid off $460,000 in debt, said groceries and dining are often among the fastest areas where households can reduce spending. Frequent restaurant visits and delivery fees can add hundreds of dollars to a monthly budget. Packing lunch at home instead of buying it during the workday can make a meaningful difference, she said.

“For the average person, they could spend anywhere between $12 to $15 going out for lunch, but making that same lunch at home, you could save half and only spend $5 or $6,” Warshaw said.

Baby steps

Ramsey Solutions, the financial education company founded by Dave Ramsey in 1992, recommends seven baby steps to help people take control of their finances. The first step is what Warshaw described: Saving $1,000 for a starter emergency fund. After that comes paying off all debt, saving three to six months of expenses and investing 15% of household income in retirement. 

Warshaw said that people trying to get out of debt could consider temporarily pausing their retirement contributions until they pay everything off. She also recommended being realistic about expenses, especially during the holidays.

“If a family says, ‘I’d like to save $1,000 in the month of December,’ well, it’s going to be tough because you’ve got Christmas going on,” she said.

These baby steps are difficult for most Americans to follow. According to a Bankrate survey released in February, only 47% of Americans said that they had enough money to cover a $1,000 emergency expense. Around 30% indicated that building an emergency fund and decreasing their credit card debt were equally important to them. 

“Most American households want to grow their savings, but few are making meaningful progress right now,” Stephen Kates, Bankrate financial analyst, said in the report. “Rather than trying to tackle everything at once, I recommend focusing on the single most important financial priority in 2026 and making consistent progress there first.”

Does saving $1,000 in a month seem impossible? Personal finance expert Jade Warshaw, who co-hosts The Ramsey Show, a popular financial talk radio program and podcast, says it’s not as difficult as it seems. 

Warshaw recently said in an interview with Fox Business that most Americans can put aside $1,000 if they are willing to sacrifice the little things and view their finances with a “scorched earth” mentality. “It’s more realistic than people think,” she said. “Most people are able to do it in 30 days.”

Warshaw said to start with building a budget. “If you don’t have a budget, it’s not going to work,” she said. That means calculating net monthly income, tracking expenses and making sure expenses are less than income. 

A few years ago, a candidate who had a poor experience during recruitment might tell a few friends. Today, they post about it. They leave a Glassdoor review and share screenshots. Hiring has become public in a way most organizations weren’t built to handle.

What I’ve seen, both through building a company and through analyzing how people find and evaluate jobs, is that trust has become the deciding factor in hiring. Job seekers evaluate employers by how clearly and consistently an organization communicates before anyone has been offered a role.

The cost of getting that wrong is measurable: Companies with a poor employer reputation pay at least 10% more per hire, and that’s before accounting for the offers declined, the referrals lost and the reviews that don’t stop.

The hiring process is part of your reputation

Those signals start earlier than most organizations realize. By the time someone applies, they’ve already formed an impression — shaped as much by what a company hasn’t said as by what’s written in the job description.

The data supports this — 66% of candidates say a positive hiring experience influenced their decision to accept an offer. Poor communication or unclear expectations led 26% of job seekers to decline offers, and up to 72% say they’ll share a bad hiring experience publicly. The numbers reflect a pattern, and patterns compound.

The specific ways unstructured hiring damages trust

Ghost jobs are one of the most visible symptoms of broken hiring communication. When organizations post roles that aren’t actively being filled, or allow listings to sit long after hiring has paused, they erode the credibility of everything else. People apply, hear nothing and draw conclusions about the company, its honesty and its regard for people’s time.

Ghost jobs are only one piece. Vague timelines, repeated interview rounds without explanation and inconsistent updates between stages all create the same corrosive effect: uncertainty. And uncertainty, for most candidates, reads as disrespect.

A 2023 survey found that 40% of applicants were ghosted after multiple interview rounds. Another 35% didn’t receive any acknowledgment of their application. The frustration this generates becomes the story people tell about your brand.

What transparency looks like before the first interview

The recruiters who consistently close strong candidates tend to operate from the same principle: Clarity before the interview is more valuable than a polished pitch during it. What applicants want to understand early is how the process works and what the company genuinely expects. Vague or overly optimistic job descriptions tend to omit all that.

This has practical dimensions, too. Nearly half of job seekers expect to learn about salary before applying. Organizations that include compensation ranges, realistic role descriptions and defined hiring stages self-select for hires who already understand what they’re walking into. The rest of the process gets faster and more honest as a result.

Keeping candidates engaged when decisions take time

Here’s a dynamic recruiters understand but rarely say out loud: Someone can be strong enough to stay in consideration without being strong enough for an immediate decision. Business priorities shift, or budgets change. A role that was well-defined three months ago may have evolved significantly by the time a finalist is being evaluated.

This creates a genuine tension because candidates expect clarity, while recruiters are often working against a moving target internally. The organizations that handle it best are the ones that communicate openly when priorities shift, rather than allowing candidates to sit in silence. A brief update that says the timeline has changed does more for candidate trust than a polished message delivered three weeks late.

Where automation helps and where it doesn’t

Fifty-five percent of candidates hold a negative view of AI in recruitment, citing bias risk and the dehumanizing effect on the process. The concern is the loss of human judgment at moments that feel significant.

Automated systems that handle scheduling or send status updates are useful, but the ones that replace human contact at moments candidates consider meaningful tend to damage trust faster than no communication at all. 

First contact, feedback after interviews, rejection — these are often the moments that define how someone remembers a company.

The long-term business case for getting this right

Candidate experience doesn’t end at the offer — it shapes what comes after. Employees who had an exceptional candidate experience are 3.2 times as likely to feel connected to their organization’s culture long-term, according to Gallup. That process isn’t separate from the employee experience. It’s where it begins.

Those who say they’re unlikely to apply again after a negative experience represent a shrinking future talent pool. The reviews they leave, the conversations they have and the referrals they don’t make are real costs. They accumulate slowly enough that they’re easy to ignore until they aren’t.

Only 14% of organizations have fully implemented their pay transparency approach, according to Mercer’s global survey of over 1,600 companies. For most, transparency is still an aspiration.

Organizations that close that gap build a reputation for being worth working for. That reputation is gained one hiring process at a time. And right now, most companies are leaving it to chance.

A few years ago, a candidate who had a poor experience during recruitment might tell a few friends. Today, they post about it. They leave a Glassdoor review and share screenshots. Hiring has become public in a way most organizations weren’t built to handle.

What I’ve seen, both through building a company and through analyzing how people find and evaluate jobs, is that trust has become the deciding factor in hiring. Job seekers evaluate employers by how clearly and consistently an organization communicates before anyone has been offered a role.

The cost of getting that wrong is measurable: Companies with a poor employer reputation pay at least 10% more per hire, and that’s before accounting for the offers declined, the referrals lost and the reviews that don’t stop.

I try to maintain a growth mindset. That means taking constructive criticism seriously, because I want to properly consider every opportunity to learn how I can improve my business.

But not all criticism is constructive. Sometimes, unsolicited opinions are intentionally shared to disrupt or distract you. And sometimes, advice is simply bad — no matter how good the intentions behind it may be.

The question is: How do you tell the difference between the feedback that helps you grow and the noise that will throw you off track if you pay it too much attention?

My company, Roof Maxx, exists in a highly competitive industry, so we’re used to hearing other people’s opinions. We sell a roof maintenance solution for asphalt shingles that can extend their usable lifespan for years by restoring the flexibility they naturally lose over time. Most of the feedback we receive from customers is overwhelmingly positive; we have over 20,000 online customer reviews with an average rating of 4.9/5 stars. But every so often, we come across misinformation or a divergent opinion about our service and the restoration product it uses.

Here’s how we decide whether these cases are worth addressing, and how you can do the same if you ever find your company receiving confusing or counterproductive feedback.

Any feedback that comes from customers is worth paying attention to

First, let me clarify the above by saying that whenever one of your customers — or potential customers — has questions or concerns, it’s always in your best interest to address them. You never blow off the people you serve. That should be rule number one for pretty much every business.

That’s why we have always remained committed to ensuring dealers respond to customer inquiries quickly. If a homeowner wants to know whether Roof Maxx is right for their roof, a dealer goes to perform a thorough assessment and provide their honest opinion. They only recommend Roof Maxx when the roof is a good candidate, with shingles that are aging or brittle but still in decent structural condition.

Our warranty also guarantees that the asphalt shingles our dealers treat will stay flexible and serviceable for five years from the treatment date. So if a customer ever informs us that their treated shingles are losing flexibility within that window, our dealers are trained to investigate right away and re-treat the area in question if necessary. This doesn’t happen often, but the fact that we demonstrate this kind of accountability to customers is still part of why our reviews are so positive.

Not all feedback comes from customers — or even real people

Like many other businesses, we’ve learned that it’s important to consistently monitor AI platforms for brand mentions. Most of the time, platforms like ChatGPT or Gemini seem to present accurate information about Roof Maxx’s approach to maintaining asphalt shingles. But every so often, I can tell that the data it’s collecting for its responses isn’t entirely accurate.

Here’s one recent example that stuck out to me: that “most consumers love it, while some contractors question it.” This was hugely telling to me, because it speaks to both our stellar reputation among homeowners and the resistance that we’ve occasionally faced from other professionals in an industry we’ve successfully disrupted.

When you think about it, it actually makes sense that some contractors would have a vested interest in criticizing maintenance solutions that make roof replacements less necessary. After all, roof replacements are the highest-margin service a typical roofing contractor can sell, so it’s common for contractors to recommend them — even in cases where a homeowner’s current roof can still be saved.

That means the more homeowners who learn that Roof Maxx is a viable alternative, the fewer unnecessary replacements these contractors are likely to sell. Of course, some of them are going to question it.

What to do when your brand is misrepresented online

It’s always disappointing for me when I see these kinds of results show up in our AI brand mentions, but I also understand that it’s not because we’re doing anything wrong. In fact, it’s the opposite: We’re facing a degree of resistance within our industry precisely because we offer an innovative solution to a common problem, and that’s inconvenient for contractors whose business model depends on customers not having a better option.

Chances are that your AI brand mentions won’t always be perfectly accurate. But they can still be valuable, because studying the discrepancies can reveal a lot about how your competitors perceive you and the messages they’re putting out about your brand. This can set you up to control the narrative.

I use any inaccurate information about Roof Maxx that I find online to update our marketing strategy accordingly. We address false claims head-on in our blog and on our website. Our dealers make speaking to those points a priority when educating potential or existing customers.

What we’ve found is that the best strategy is not to fight misinformation at the source when the source has a clear ulterior motive. We just make sure we’re being as clear and direct as possible when putting the truth out there.

Telling the truth always gives you an advantage because reality will eventually back it up. In our case, that means more glowing reviews and referrals from people who have actually seen the results of our solution firsthand. Eventually, that information outweighs the noise — improving brand awareness among both AI and potential customers everywhere.

I try to maintain a growth mindset. That means taking constructive criticism seriously, because I want to properly consider every opportunity to learn how I can improve my business.

But not all criticism is constructive. Sometimes, unsolicited opinions are intentionally shared to disrupt or distract you. And sometimes, advice is simply bad — no matter how good the intentions behind it may be.

The question is: How do you tell the difference between the feedback that helps you grow and the noise that will throw you off track if you pay it too much attention?

What would you study if you were in college today?

For many college students, majoring in computer science is no longer a job guarantee. According to the Federal Reserve Bank of New York’s labor market data, recent computer science graduates have an unemployment rate of about 6.1%, slightly higher than the 5.7% overall rate for recent graduates. 

Not all hope is lost. The U.S. Bureau of Labor Statistics predicted that software developer jobs will grow 15% from 2024 to 2034, about five times the average for all jobs. So long-term demand for computer science is strong, even as recent graduates face an uphill climb trying to enter the field. 

According to a new report from The Wall Street Journal, experts recommend three alternative majors instead of computer science for students who want to thrive in an AI-dominated workplace: anthropology, mathematics and philosophy. The commonality between these different areas of study is that they thrive where AI falls short. 

For example, anthropology is “the study of what makes us human,” according to the American Anthropological Association. That inquiry feels newly urgent as AI reshapes work and forces people to consider which skills and qualities remain uniquely human. 

Alec Litowitz studied mathematics and anthropology at MIT before adding both a law degree and an MBA to the mix. Throughout his career, including as an early partner in global hedge fund Citadel, Litowitz has repeatedly relied on the insights into human behavior that he gained from anthropology.  

“The beauty of anthropology for me is that I can telescope out and try to understand what’s happening to society right now,” he said in an interview with the Journal. 

Math and philosophy are also strong contenders

Bert Bean, chief executive of Insight Global, a staffing and consulting firm, told the Journal that he brings candidates for AI jobs in for an interview and puts them in front of a whiteboard. Then, the interviewers “throw crazy technical problems” at the candidates to see how they think. 

Math majors excel at this type of assessment. They usually have experience with manual problem-solving, having grappled with hard questions and practiced working them out by hand. 

However, Litowitz, the math-anthropology double major, said that if he were to do it all over again, he would pick philosophy. Ana Prestamo, a recent graduate from the University of Notre Dame, agrees that philosophy is a useful major.

“Studying philosophy, or any of the humanities for that matter, has given me something that artificial intelligence cannot replace: a closer understanding of what makes us human,” she wrote in March in The Observer. “And in a time when AI can solve almost any technical problem, understanding what makes us uniquely human becomes all the more valuable.”

Notably absent from the list of recommended majors is AI. Peter Miscovich, executive managing director at real-estate giant JLL, told the Journal that majoring in AI is “too narrow.” Eventually, saying you majored in AI will “sort of [be] like saying you majored in Excel,” he added.

What would you study if you were in college today?

For many college students, majoring in computer science is no longer a job guarantee. According to the Federal Reserve Bank of New York’s labor market data, recent computer science graduates have an unemployment rate of about 6.1%, slightly higher than the 5.7% overall rate for recent graduates. 

Not all hope is lost. The U.S. Bureau of Labor Statistics predicted that software developer jobs will grow 15% from 2024 to 2034, about five times the average for all jobs. So long-term demand for computer science is strong, even as recent graduates face an uphill climb trying to enter the field. 

The skills that help you build a company are not always the same skills you need to lead it at scale.

When a company is small, the CEO knows almost everything that is happening. You know the people, the clients and where the biggest opportunities are. You are close to the details, and when something goes wrong, you can usually get involved and help fix it yourself.

I remember that stage of my career clearly. There was a certain comfort in being close to everything. Decisions could be made quickly because the distance between a question and the person making the decision was very short.

Then the company grows, markets expand, teams multiply, new offices open, customers come from different parts of the world. Decisions become larger, the consequences become harder to see immediately, and there are simply more things happening than one person can follow.

That is when I learned one of the harder lessons of leadership: The way you lead a company at one stage of its growth can become a limitation at the next. For a CEO, growth requires a change in role.

You stop being the center of every decision

In a growing company, there is a natural temptation to stay involved in everything.

It comes from a good place. You care about the business. You know its history. You have developed instincts that have served you well. You may even believe that your involvement protects the quality of decisions.

Eventually, though, the volume makes that impossible. The more important question becomes whether you have built a team that can make good decisions without you. That question has changed the way I see leadership.

At BGN, we operate across more than 120 countries, with people working across different markets, cultures and areas of expertise. I cannot be in every room where a decision is being made. Nor should I be.

My responsibility is to make sure the people in those rooms understand the direction of the company, the standards we expect and the judgment required to act in its best interests.

That takes time. It also requires trust.

The hardest thing to delegate is judgment

Delegating a task is relatively easy; however, delegating a decision is much harder.

The real test comes when you give someone responsibility for something important and resist the urge to step back in when they approach it differently from the way you would have. That is where leadership gets uncomfortable.

People need room to develop their own judgment. They need to make decisions, learn from the consequences and gradually take on more responsibility.

If a CEO corrects every decision before a person has had the opportunity to own it, that person learns something very quickly: Wait for the CEO. And that is exactly what a growing company cannot afford. The goal is to develop leaders who can think independently while remaining aligned with the company’s values and objectives.

I have found that this takes more than hiring talented people. It requires giving them meaningful responsibility and allowing them to grow into it.

Show me your calendar, and I’ll tell you what kind of CEO you are

One of the clearest signs that a company has changed is the CEO’s calendar.

Early in a company’s life, the calendar can be filled with operational questions: Which customer needs attention? Which deal needs to be closed? Which problem needs solving today?

As the company grows, the most valuable use of a CEO’s time changes. You need to spend more time thinking about where the company should be in three, five or 10 years. Which markets deserve investment? Where should we build? Which capabilities will we need? Who are the leaders who can take the company forward? What should the organization stand for as it grows?

Those questions rarely produce an immediate result: There is no satisfying feeling of crossing something off a list, and yet they may be among the most consequential decisions a CEO makes.

I have become increasingly protective of time for that kind of thinking. A full calendar can create the feeling of productivity while leaving little room for perspective. The larger the company becomes, the more valuable perspective becomes.

You have to let the company become bigger than you

There is a personal side to this transition that people do not talk about enough.

When you have spent years building a business, your identity can become closely connected to it. You know the history. You remember it all: the difficult years, the people who took a chance on the company when it was smaller, the decisions that changed its direction. That history is never truly behind us; it shapes where we go.

But the company also has to develop an identity of its own. If every important relationship, decision or opportunity depends on the CEO personally, the organization remains smaller than its size suggests.

A strong company should be able to carry its values through many people. That means developing leaders who can represent the business with customers and partners as well as giving people enough context to understand why decisions are made. It also means creating a culture where standards remain consistent even when the CEO is not present.

For me, that is one of the most rewarding parts of leadership.

Seeing someone you have developed walk into a room and handle a situation exceptionally well gives you a different kind of satisfaction from solving the problem yourself. You realize the organization is growing its own strength.

The CEO has to keep learning too

There is another trap that comes with seniority: People begin to assume that because you are the CEO, you should already know the answer.

Sometimes you do, but let’s face it — often you do not.

The larger and more international a company becomes, the more important it is to remain curious. Someone who works close to a customer may understand something the executive team has missed. A colleague in another market may see an opportunity that looks invisible from headquarters. A younger member of the team may question an assumption that has been accepted for years.

I want people around me who are willing to challenge my thinking. That requires humility, but it also requires confidence. A leader who feels threatened every time someone disagrees will eventually surround herself with people who agree too easily. That is dangerous for any company.

The CEO has to keep listening, especially when the company becomes large enough for the leader to hear mostly what other people think she wants to hear.

Growth changes the questions

I think about the evolution of leadership through the questions we ask.

When you are building a company, you ask, “How do we make this work?” As the company grows, the question becomes, “Who can make this work without me?” Then it becomes, “How do we build an organization that can keep growing?” And eventually, “What kind of company are we building for the people who will lead it after us?”

That last question changes the perspective completely.

It moves leadership beyond the next deal, the next quarter or even the next stage of growth. It makes you think about culture, talent, reputation and institutional knowledge. It makes you think about whether the company can continue to evolve when the people who built it eventually step aside.

That is a responsibility I take seriously.

Growth should change the CEO too

A company can only grow as far as its leadership is willing to grow with it.

For me, that has meant becoming more comfortable with distance from the details and more deliberate about where my attention belongs. It has meant trusting people with decisions that I once would have wanted to make myself. It has meant accepting that someone else may approach a problem differently and still reach an excellent outcome.

Most importantly, it has meant understanding that leadership at scale is a different job.

The instinct to get involved is still there. So is the satisfaction of solving a difficult problem yourself. But there is a greater satisfaction now in seeing a team solve something that once would have landed on my desk.

That is how you know the company is becoming bigger than its founder, its CEO or any single individual.

And perhaps that is one of the clearest signs that you have built something that can last.

The skills that help you build a company are not always the same skills you need to lead it at scale.

When a company is small, the CEO knows almost everything that is happening. You know the people, the clients and where the biggest opportunities are. You are close to the details, and when something goes wrong, you can usually get involved and help fix it yourself.

I remember that stage of my career clearly. There was a certain comfort in being close to everything. Decisions could be made quickly because the distance between a question and the person making the decision was very short.

Many brands use the U.S. Open to entertain clients or put up flashy signage. IBM does both, but its involvement goes much deeper. Behind the scenes, the century-old tech titan is the invisible hand shaping the fan experience, powering the U.S. Open app, and engaging 14 million fans worldwide. In the process, IBM is making the case that sports might be the perfect playground for AI integration.

The servers behind each serve 

From a fan’s perspective, it seems simple. You open the app, check the score and maybe watch a highlight or two. But making that experience work requires IBM and the USTA to process a mountain of data in real time, using AI alongside human review to balance speed and accuracy. One new metric, Serve Quality, analyzes 21 data points across a player’s body and racquet 50 times per second, generating roughly 4.6 million data points per match and more than 1.2 billion throughout the tournament. And then there’s the match chat, an AI bot that lets users ask conversational questions about the sport, like “what is the record for x?” 

Users appear to be embracing these features, if the numbers are any indication: More than 14 million unique devices engaged with U.S. Open digital platforms in 2025, generating nearly 47 million visits, up 19% year over year.

So why does the USTA need IBM to help run a tennis tournament? To understand that, you have to recognize that the U.S. Open isn’t simply a sporting event. It’s a massive, temporary business operation with millions of customers, enormous amounts of data, and a three-week deadline. That comes with a litany of challenges, from security and bandwidth to content production and infrastructure, that require the scale and experience of a company like IBM.

The USTA is a few-hundred-person team that’s expected to scale monumentally during those three weeks each year. With IBM’s help, however, it can deploy AI-powered tools to streamline infrastructure management, optimize costs, enhance content production workflows, and improve operational resilience. For example, the USTA’s small editorial team uses bespoke AI tools to generate match summaries, which humans then review and edit, increasing the team’s content production capacity by 300%. 

On top of that, an operation as large as the U.S. Open comes with major security risks, and IBM plays a key role in deterring them. According to a spokesperson, IBM protects against nearly 500 million suspicious requests targeting the USTA’s infrastructure and digital platforms. Its AI agents help identify, categorize and prioritize potential threats, allowing the team to respond faster and better protect the USTA’s assets.

Rallying fans around the world  

At the heart of this year’s experience is the all-new Live Updates homepage, a smarter, more personalized way for fans to follow the action that matters most. Fans can prioritize their favorite players and quickly zero in on the matches, insights and stories they care about most. From the World Cup to the NBA Finals, this summer has made one thing clear: demand for sports is exploding, but stadium seats aren’t.

For the millions of fans who can’t afford to attend the tournament or live too far away to travel to New York, these digital experiences let them feel closer to the action.

Here are some of the other new features in the U.S. Open app this year:

1. Match Chat: An AI-powered assistant that lets fans ask free-text questions about live matches and receive instant answers and insights based on match statistics, player information, historical performance, and tournament data. Responses are generated using AI agents and models trained in the USTA’s editorial style and the language of tennis.
2. Live Likelihood to Win: A real-time predictive feature that continuously updates each player’s chances of winning as a match unfolds, using live scores, statistics, and expert analysis. The AI-powered visual representation shows how match momentum is shifting in real time, essentially serving as the heartbeat of the match.
3. Key Moments: A generative AI feature within Live Likelihood to Win that goes a layer deeper, explaining why a player’s chances of winning changed by identifying the points, rallies, and sequences that had the biggest impact on match momentum.

While much has been made of the public’s skepticism toward AI, an IBM study found that sports fans have been fairly receptive. The IBM 2026 Sports Survey found that 80% of fans see value in AI-powered sports experiences, particularly real-time statistics, personalized content and translation. Among tennis fans, 64% said they trust AI-generated sports content.

The study also highlights the importance of having a strong digital presence. IBM found that 73% of tennis fans surveyed use sports apps. In comparison, 91% use them during live events, making a reliable, real-time digital experience increasingly important to how fans consume and engage with sports.

Image Credit: IBM

Playing the long game  

The U.S. Open isn’t IBM’s only sports play. The company is also involved with the UFC, the Masters, and Wimbledon, where it runs similar programs, powering digital platforms and handling data for each respective event. But IBM’s sports strategy extends beyond working with some of the world’s biggest events. It’s also looking at the grassroots level of sports and the startups building its future.

Earlier this year, the USTA launched the inaugural USTA Connect Innovation Challenge, a nationwide open call for tech innovators, engineers and startups to build the next high-impact digital solution for tennis. Selected participants receive access to USTA and U.S. Open data sets to power their prototypes, and a panel that includes USTA representatives, venture capitalists, and an IBM team member evaluates submissions. The top three finalists receive fully funded travel to the USTA Connect event at the 2026 U.S. Open to pitch their ideas, while the grand-prize winner receives $10,000 and exposure across the USTA ecosystem.

More recently, IBM launched the IBM Sports Tech Startup Challenge, which will give founders opportunities to showcase their solutions at Web Summit events in Rio, Vancouver and Lisbon, as well as alongside major tech events including a16z Tech Week in New York and San Francisco. The initiative gives startups a platform to pitch their ideas to audiences at the intersection of sports, technology, and entrepreneurship.

From quietly shaping the fan experience behind the scenes to seeking out the next generation of sports innovators, IBM is proving that even the world’s biggest companies are taking sports seriously. And entrepreneurs should, too.

Many brands use the U.S. Open to entertain clients or put up flashy signage. IBM does both, but its involvement goes much deeper. Behind the scenes, the century-old tech titan is the invisible hand shaping the fan experience, powering the U.S. Open app, and engaging 14 million fans worldwide. In the process, IBM is making the case that sports might be the perfect playground for AI integration.

The servers behind each serve 

From a fan’s perspective, it seems simple. You open the app, check the score and maybe watch a highlight or two. But making that experience work requires IBM and the USTA to process a mountain of data in real time, using AI alongside human review to balance speed and accuracy. One new metric, Serve Quality, analyzes 21 data points across a player’s body and racquet 50 times per second, generating roughly 4.6 million data points per match and more than 1.2 billion throughout the tournament. And then there’s the match chat, an AI bot that lets users ask conversational questions about the sport, like “what is the record for x?” 

Users appear to be embracing these features, if the numbers are any indication: More than 14 million unique devices engaged with U.S. Open digital platforms in 2025, generating nearly 47 million visits, up 19% year over year.

Opinions expressed by Entrepreneur contributors are their own.

It took Emil Barr just 14 months of work to see his first $1 million hit his bank account. He was 19 years old at the time.

Today, at age 23, the founder and CEO, who created two companies while still in school, estimates that his personal net worth is around $35 million. He is unapologetically aiming to be a billionaire by age 30. 

Born in Russia, Barr moved to the U.S. when he was three years old and grew up in a small Ohio town. He stuck out at an early age.

“I was the weird Russian kid that didn’t speak any English,” he tells Entrepreneur. “I think I always felt out of place. And I think that as an entrepreneur, you have to be comfortable with discomfort and that feeling of cutting against the herd.”

In high school, he made his peace with being different and even leaned into it. 

“I’m convinced every high school has at least one weird kid that wears suits to school every day,” he says. “You probably had one. That was me.”

Money, not ambition, first pushed him into entrepreneurship. When it came time to choose a college, Barr enrolled at Miami University, the only college that he could afford. He was looking into transferring to an Ivy League school, but tuition was out of reach. 

“I was like, If money is the limiting factor, how hard can it be to make $100,000 [and] go pay for a year’s tuition?” he says.

How he made his first $1 million

Barr was on the lookout for money-making ideas when he met a classmate with 11 million TikTok followers who was barely earning anything from her social media presence. 

“She got one brand deal for $200,” Barr says. “This is crazy because on Instagram, even if you had a million followers, that would be your full-time career. This was a platform that everyone was using. There was no revenue there yet.”

Barr started his company, Step Up Social, in his freshman year dorm room. His plan was straightforward: Businesses had no idea what to do with TikTok, but Gen Z did. Step Up Social positioned itself as a social media marketing and advertising agency focused on creating short-form video content.

Starting the company required little more than an iPhone and an Internet connection. 

“We grew from $0 to $1 million in revenue in six months,” Barr says of Step Up Social. “As an 18-year-old, I had no idea what I was doing. I never had a corporate internship or anything like that.”

Instead of spending the money, he reinvested in the company’s growth.

“I think the first time I had truly a million dollars in my bank account was 14 months in,” Barr recalls. “It was the start of my sophomore year of college.”

Decisions that led to rapid growth

Growing Step Up Social meant embracing risk, especially debt. When the company adopted 90-day payment terms with large clients, there was a funding gap. Barr had to pay influencers upfront while waiting months for invoices to clear. 

“I was basically running around and taking out as many credit cards and bank loans as I could to keep the company afloat,” he says. “I took out about $1 million worth of personally guaranteed unsecured loans, and everyone thought I was crazy.”

His logic was simple: At 19, he had no assets, so the downside was limited. “If we failed, what were they going to do?” he says. “Were they going to take my shirt or my car? I didn’t have anything to take.”

The third key decision, in his view, was prioritizing people over lifestyle. The “absolute best thing” he spent money on was hiring people with “20 or 30 years of experience,” he says. 

How he grew Step Up Social

Early on, intent on gathering clients, Barr cold-emailed a few hundred companies. The first serious bite came from Kao, a Japanese consumer giant and Procter & Gamble competitor. Barr drove his old, beat-up car an hour and a half to downtown Cincinnati and walked into a 47th-floor boardroom wearing a university T-shirt and shorts. The executives gathered there asked him for his deck.

“I was like, ‘What’s a deck?’” he laughs.

Despite underpricing himself at “$2,000 a month,” he landed the account. That one contract gave Step Up Social credibility and opened doors.

“It was exponentially easier for us to get our next 10 to 15 brands, and it was just off to the races,” Barr says. 

From there, Step Up Social scaled into a full-service TikTok marketing agency, hiring influencers and managing online presences for brands and celebrities. By the time he sold it last year, the firm, by then acquired and rolled into a larger agency, was working with Procter & Gamble, Nike, Nordstrom, Kroger, Alo and Banana Republic.

“We were doing about $2 million a year in revenue, but it was extremely high margins,” Barr says.

Step Up Social earned revenue by connecting brands with creators. For example, a brand might pay the company $600 for a video. The company would then pay the creator $400 to make it and count the remaining $200 as revenue for arranging and managing the deal.

“Gross transaction revenue was closer to $8 to $9 million,” Barr says.

Convincing his university to pay him

Barr didn’t just build a business while attending college; he turned the school itself into a revenue source and marketing machine. He convinced his university to cover his tuition fees. The school also paid him $200,000 and gave him a faculty parking pass. 

Miami University had introduced its entrepreneurship program relatively recently. Barr saw leverage. “I was effectively the only student entrepreneur on campus,” he says. If he dropped out, “they would have no student entrepreneurs. It wouldn’t be a very compelling case study.”

He started with “small asks” like flexible attendance, arguing that it was more important for him to run his company than to participate in group projects. Then, he applied for every grant and pitch competition he could find at the school, winning “$40,000 in a couple of months.”

From there, he reframed himself as both a case study and a vendor. Miami University became a client. Barr’s agency turned the school into “the most-followed public university on TikTok in America,” a result he argues paid back any support many times over.

“For every $1 they spent, whether it was in contracts with us or grants for the business, I’m sure they made at least $10 back in tuition from students who heard of Miami and were drawn to the school,” he says. “So it was probably a good deal for everyone.”

Building Flashpass

Barr’s latest venture, Flashpass, looks very different from a TikTok agency. At its core, Flashpass is his answer to a looming question: What happens to workers if AI replaces 25% to 50% of jobs?

“If we could actually build a way for these 25% to 50% of people who might lose their jobs to be able to quickly get certified online and go find a new job in 30 days, that would be a very valuable service to government as well as to individual users,” Barr explains.

Flashpass is an online platform built around “micro credentials.” Users can learn a new skill in 30 days or less and then be matched with jobs in industries that need talent.

“We have things like natural energy, like oil and gas careers. We have things like medical billing and coding,” he says. “These are all industries where they have a lot of job openings, and they can’t find enough people, and the average pay is over $80,000 a year.”

How Flashpass makes money

The platform doesn’t charge individual users or employers. Instead, Flashpass sells its services to state governments. 

“Typically what we do is we’ll partner with a school, and the government will pay the school, and we will split the revenue with the school,” he says. 

The school helps build curriculum and recruit candidates; the government treats Flashpass as one more education and workforce tool.

“If we could take this Flashpass idea and actually give it to the government and make it free for everyone who loses their jobs as a result of AI, we could build a very valuable business,” Barr says. 

The bet appears to be paying off. Flashpass began with a $4 million, two-year pilot contract in Ohio, with roughly $2 million in annual revenue. Barr says he invested about $75,000 of his own money to build a demo, then used it to land that pilot. Since then, the company has added contracts in Louisiana (about $1 million a year) and Delaware ($2.3 million a year), and has proposals out in 17 states.

“This year, just based on the existing contract volume, we’re set to do at least $8 million, and that’s a four-fold increase over last year,” he says.

Work-life imbalance

Today, Barr estimates his net worth is around $35 million, up from $25 million when he spoke with Business Insider in December. He’s open about the personal cost of becoming a millionaire. In college, his schedule was packed to the minute. He took college classes from 8 a.m. to the early afternoon, then conducted back-to-back calls until 7 p.m. and went to networking dinners. He did his “true work” on the business until 4 a.m., finishing his day with three hours of sleep. 

“I gained 80 pounds,” he says. “I lived off of Red Bull…four or five cans of Red Bull each day.” He skipped holidays and ignored invitations to go out.

He’s since lost 30 pounds and hired a trainer who comes to his house twice a day. The hardest part, he says, is realizing “it’s three times harder to undo the damage than to do the damage initially.”

He also has a personal chef, a home assistant and a driver. Barr still works 19-hour days, but he says he’s calmer and more measured as a leader with the extra help.

One lesson he wishes he’d learned earlier: Don’t spend your 18- to 20-hour days chasing small goals.

“It takes the same amount of effort to do something big as it does to do something small,” he says.

A service business owner usually knows how the work should be done: what a good client interaction sounds like, when an account needs attention, which problems require escalation and where margins can disappear.

That clarity makes the early stages feel manageable. The owner can catch problems, answer questions and keep clients satisfied. But as the business grows, the habits that once held everything together begin to strain.

Ten employees become 30, one market becomes three, and the owner can no longer know what’s happening everywhere. Decisions that once took a quick conversation now require others to have the context and authority to act.

That’s when a surprisingly common problem surfaces: the business has grown, but its systems have not.

I see versions of this throughout franchising. One of the most revealing questions I can ask an operator is not about revenue or customer acquisition. It’s this: “What happens here when you’re not available?”

The answer shows whether the company has translated the owner’s instincts into operating rhythms others can follow, or whether it still depends on informal knowledge passed along one interruption at a time.

If routine decisions stop without the owner, the company may have strong demand and talented people, but it does not yet have a business that can operate independently of the person who built it.

Growth exposes what the owner has been carrying

Small service businesses can run well on institutional knowledge because a few people carry the details that matter.

Someone knows which client needs a call before a schedule change, which employee can handle a difficult assignment, and which account needs an extra quality check.

Scale changes that. More clients, employees, and locations create a greater distance between the person who knows the answer and the person who needs it.

The systems gap appears when knowledge that once lived comfortably inside a few people’s heads needs to become repeatable across an organization.

Documentation is not the same as a system

Owners often respond by creating more procedures, as if a larger binder or longer checklist will automatically create consistency.

A functioning business system should help someone make the right choice, even when the usual decision-maker isn’t standing beside them.

But employees need more than tasks. They need the desired outcome, decision boundaries, and clear escalation points.

I’ve seen this distinction become especially important in commercial cleaning because the work happens across client locations, often outside traditional business hours. A manager cannot physically supervise every team at every facility.

In that environment, the system must operate independently. Ongoing education, quality controls, communication protocols, and accountability must create consistency even when management is miles away.

Technology cannot repair a broken process

Artificial intelligence and automation make this issue more urgent because service businesses now have more tools to speed up scheduling, communication, reporting, and performance management.

Those tools are valuable, but they can tempt leaders to automate before they’ve defined the process they want to improve.

A bad process does not become a good system because software executes it faster. If decisions are unclear, technology simply moves confusion more efficiently.

The sequence matters: clarify the desired outcome, identify inconsistent decisions, and understand why employees improvise. Then technology can reinforce a process that already works.

The CEO test is whether the business needs the CEO

That can be uncomfortable for founders because being needed often feels productive, especially in the early years.

Answering questions feels like leadership. Solving problems feels like service. Stepping in feels like proof the owner is still close to the business.

Over time, however, those strengths can become constraints when every answer still must pass through the same person.

As a CEO, I’ve learned my job is not to answer every operational question. It’s to build an organization that can reach the right answer without me.

Closing the systems gap

Service businesses often chase growth by adding more clients, employees, markets, and technology.

Sometimes the next stage requires something less visible: examining how decisions get made when the owner isn’t in the room.

Every recurring decision that requires the owner may reveal an opportunity to strengthen the system and make growth more durable.

The work often starts by naming the decisions that repeat each week, then deciding who should own them, what information they need, and when they should escalate. That simple discipline turns experience into guidance and gives managers confidence to act before small issues become larger problems.

Revenue and headcount show size. They don’t show whether the business can function without constant direction.

That may be the better test of scale: not just whether the business can grow, but whether it can keep making sound decisions after growth has stretched the founder’s reach.

A service business owner usually knows how the work should be done: what a good client interaction sounds like, when an account needs attention, which problems require escalation and where margins can disappear.

That clarity makes the early stages feel manageable. The owner can catch problems, answer questions and keep clients satisfied. But as the business grows, the habits that once held everything together begin to strain.

Ten employees become 30, one market becomes three, and the owner can no longer know what’s happening everywhere. Decisions that once took a quick conversation now require others to have the context and authority to act.