What the Past Several Months on the Road Taught Me About Where AI Really Stands

What the Past Several Months on the Road Taught Me About Where AI Really Stands

Over the past several months, I’ve had the opportunity to travel across the country keynoting and speaking at conferences for business leaders, professionals, and organizations trying to make sense of artificial intelligence. During that stretch, I traveled to Philadelphia, St. Louis, Arizona (three times), Atlanta, Newark, Naples, and Palm Beach.

What has stood out to me most is not just how quickly AI is changing. It is how quickly the conversation around AI is changing.

A year ago, many people were still asking basic questions. What is ChatGPT? Is this just hype? Will AI really affect my business or industry?

Today, the questions are different. Leaders are asking how to implement AI safely. Employees are asking how to use it effectively. Organizations are trying to figure out where AI creates real value and where the risks need to be managed.

After speaking with people across different industries, here are the biggest things I’m seeing.

1. AI Is Moving Faster Than AI Governance

One of the clearest patterns I’ve noticed is that most businesses are interested in AI, but many still do not have clear rules for how employees should use it.

That is a problem.

In most industries, when I ask whether the organization has an AI usage policy, the answer from a significant majority of the room is still no. Many leaders know their employees are already using large language model tools, or LLMs, like ChatGPT, Copilot, Claude, Gemini, or Perplexity, but they have not clearly defined which tools are approved, what information should never be entered, or when AI-generated work needs to be reviewed by a person.

There are some exceptions. In fields like law, I am beginning to see more firms put structure around AI governance because confidentiality, ethics, and professional responsibility are already part of the daily conversation. But in industries like construction and manufacturing, when I ask about AI usage policies or governance, only a few hands go up.

This is especially important because AI is already being used, whether leadership has formally approved it or not.

The key point is simple: businesses do not need a complicated AI governance program to get started. But they do need basic guardrails.

At a minimum, organizations should start by deciding which AI tools are approved for work use. My recommendation is that companies begin with a very narrow list. Only allow tools where the organization is comfortable with employees entering confidential, proprietary, client, or company information.

That may sound restrictive, but it is a smart place to start.

Many generative AI tools use data in different ways, and most employees do not fully understand the privacy, security, confidentiality, and intellectual property risks involved. If a company has not reviewed and approved a tool, employees should not be entering business information into it.

In other words, do not bring your own AI tools to work.

This is the “training wheels” stage of AI adoption. Start with approved tools. Make it clear what can and cannot be entered. Explain which tools are allowed. Completely disallow unapproved generative AI tools for business use until the organization has reviewed them.

That gives employees a safe place to begin. It also gives leadership more confidence that AI is being used in a controlled and responsible way.

Over time, the organization can expand its approved tools, add more use cases, and build a more formal governance process. But at the beginning, clarity matters more than complexity. Start simple, start safe, and make sure everyone understands the rules.

2. AI Hype Is Outpacing Practical ROI

Another thing I’ve noticed is that expectations around AI are often ahead of the actual results businesses are seeing.

There is no question AI has evolved quickly. The tools are more powerful, easier to use, and increasingly built into the software people already rely on every day. But for the average business, the biggest return right now is not coming from AI replacing employees, running the company, or instantly creating brand-new revenue streams.

The biggest return is coming from efficiency.

The companies getting value from AI are using it to improve the work they already do. They are taking current workflows, current standard operating procedures, and routine tasks, then asking a practical question: can we do this faster, easier, or better with generative AI?

That is one of the best habits business owners, leaders, and employees can build right now.

Any time there is a repeatable task, a recurring process, or a workflow that involves creating, reviewing, summarizing, analyzing, documenting, researching, or communicating information, someone should be asking whether generative AI can improve it.

That might mean using AI to draft emails, summarize meetings, review documents, analyze data, create first drafts, improve client communication, prepare proposals, or streamline repetitive administrative work.

These examples may not sound as exciting as “AI agents running your business,” but this is where many organizations are seeing the most immediate value.

Business owners and leaders should make this expectation clear to their teams. The goal is not just to let employees use AI randomly. The goal is to educate employees to recognize where AI can disrupt current workflows and standard operating procedures in a productive way.

When employees are encouraged to look for these opportunities, the organization is much more likely to capture the efficiency gains that generative AI can create today.

My belief is that AI will eventually touch almost every workflow and standard operating procedure, especially when a computer is involved. If your team creates, reviews, summarizes, analyzes, communicates, documents, researches, or reports information, AI will likely change that process.

The opportunity is not to chase every new AI headline or let FOMO pull you into the latest AI tool. The opportunity is to look at the work your organization already performs and consistently ask: can this be done faster, easier, or better with AI?

3. AI Is Enhancing People, Not Replacing Them

For businesses that are integrating AI well, one thing is becoming clear: generative AI is not replacing people as much as it is enhancing people.

There is still a lot of fear that AI will eliminate jobs overnight. But in most organizations, the more immediate reality is different. AI is helping employees write faster, summarize information, prepare for meetings, analyze documents, improve communication, and reduce the time spent on repetitive work.

In other words, AI is not replacing the person. It is enhancing the person.

That is why the saying is so important: AI will not replace you, the people using it will.

I would take that one step further. If you become the person using AI, you increase your value.

The employees who learn how to use generative AI well will be able to produce better work, move faster, and contribute at a higher level. They will be able to take work that used to take hours and reduce it to minutes. They will be able to organize their thinking, improve their drafts, and spend more time on judgment, strategy, client service, and decision-making.

For leaders, the message is also clear. Do not view AI as a way to cut headcount. View it as a way to increase the capacity and effectiveness of the people you already have.

The companies that win with AI will likely be the ones that train their employees to use it well, not the ones that simply hope the technology will replace their employees and solve everything on its own.

4. AI Literacy Is Moving From Optional to Mandatory

We are leaving the stage where generative AI is simply “nice to know.”

We are entering the stage where it is becoming a core professional skill.

To be clear, most businesses are not yet requiring employees to know how to use AI. We are still early. Many organizations are still figuring out their own policies, tools, and training. But the direction is obvious.

Generative AI is becoming part of how work gets done.

At some point, saying “I don’t use AI” may sound a lot like someone walking into a job interview with a typewriter and saying they do not use computers. That may sound extreme today, but it is not hard to see where this is going.

AI is already affecting how people write, research, analyze, communicate, prepare, summarize, document, and make decisions. It is showing up in Microsoft, Google, Adobe, Salesforce, accounting platforms, legal research tools, customer service systems, marketing tools, and almost every major software category.

That means AI literacy will not remain optional forever.

The challenge is that these tools are evolving so quickly that it can feel harder for beginners to know where to start. New features are being released constantly. Voice, images, video, agents, custom chatbots, deep research, integrations, automation, and multimodal tools are all advancing at the same time.

That can feel overwhelming.

But the answer is not to wait. The answer is to start now and build the habit.

You do not need to master everything. Start by using AI in your daily work. Ask it to help draft, summarize, organize, analyze, prepare, and improve. Learn what it does well. Learn where it makes mistakes. Learn how to review and refine the output.

The people who start building these skills now will have a much easier time keeping up as the technology continues to evolve.

5. Do Not Overestimate Your AI Skill Level

Another interesting thing I noticed during my presentations is that many people tend to overestimate where they are in their AI journey.

At many conferences, I would ask the audience to rank themselves as beginner, intermediate, or advanced users of generative AI. What I found is that people often rated themselves higher than their actual skill level.

That is easy to understand. If you have used large language models for a while, it can feel like you know what you are doing. And compared to someone who has never used these tools, you probably do.

But this technology is moving so quickly that I think we need to be careful about calling ourselves advanced users.

There is simply too much changing too fast. New tools are being released constantly. Existing tools are adding new features. Voice, images, video, agents, deep research, custom assistants, automation, and integrations are all evolving at the same time.

For most people, it is almost impossible to keep up with all of it.

That is why humility is important. Do not overestimate your skill level, and do not assume a tool is not useful just because you tried it once and did not get the result you wanted. With AI, a tool that felt limited three months ago may be significantly better today.

This is one of the biggest mindset shifts people need to make. With traditional software, if a tool did not work well, you might ignore it for years. With generative AI, that approach can cause you to miss major improvements.

The better approach is to stay curious. Revisit tools. Test new features. Keep practicing. Ask better questions. Learn what the tools are good at, where they still fall short, and how they are changing.

In AI, confidence is helpful, but overconfidence is dangerous. This field is moving too quickly for any of us to assume we have it mastered.

6. Do Not Fall in Love With Any One AI Tool

Another important lesson is that people should not fall in love with any one generative AI tool.

That can be hard to avoid. Once someone gets comfortable with one large language model, they often start to assume that is “their tool.” They build habits around it. They trust it. They stop testing alternatives.

But in generative AI, that can be a mistake.

These tools are evolving too quickly. One tool may feel like the clear leader today, and a few weeks later, another tool may release an upgrade that changes the conversation. Over the past few years, I’ve seen tools that seemed to be falling behind suddenly come back with major improvements. I’ve also seen tools get praised as the obvious best choice, only for another platform to leapfrog them shortly after.

That is why professionals and organizations need to remain large language model agnostic.

The goal is not to be loyal to one platform. The goal is to use the best tool for the job. Sometimes that may be one large language model. Sometimes it may be another tool entirely.

If you “marry” one AI tool too early, you may miss other tools that are better suited for your work, your team, your industry, or a specific task.

This is especially important for business leaders. Do not let your organization’s AI strategy become overly dependent on the assumption that one tool will always be the best. The market is moving too quickly for that.

The better mindset is to stay flexible. Test multiple tools. Revisit tools you dismissed. Watch for major upgrades. Compare results. Encourage your team to think in terms of capabilities, not just brand names.

In AI, tool loyalty can become a blind spot. Stay curious, stay flexible, and stay focused on outcomes.

7. We Are Still Early, and You Can Still Become a Leader

Even though ChatGPT was released more than three years ago, we are still early in the generative AI journey.

That may sound surprising because AI has been part of the business conversation for several years now. But when you look at how most organizations are actually using AI, it becomes clear that we are still at the beginning.

Many companies still do not have an AI usage policy. Many employees are experimenting on their own. Many leaders are still unsure where to start. In many industries, formal training, governance, workflow integration, and AI strategy are still limited.

That should be encouraging, not discouraging.

For anyone who feels like they are behind, the reality is that there is still plenty of time to catch up. In many cases, you can do more than catch up. You can become one of the AI leaders in your industry.

You do not need to compete with an AI researcher, software engineer, or full-time AI specialist. For most professionals, the opportunity is much more practical than that. You simply need to become more capable with AI than the average person in your field.

That is an achievable goal.

This is especially important for people who are newer in their careers or still have many years ahead of them. If you enjoy learning about generative AI, experimenting with tools, and finding ways to use AI in real work, you have an opportunity to build a valuable personal brand.

Become known (brand yourself) as the person in your organization and industry who understands AI.

Not in a theoretical way. In a practical way.

Be the person who knows how to use AI to improve workflows, speed up repetitive tasks, draft better communication, analyze information, prepare for meetings, improve client service, and help teams work more efficiently.

Most industries are still just getting started. If you start learning seriously now, you can still become one of the people others look to for guidance.

8. Get Comfortable Talking to Your AI

Another major shift I’m seeing is how quickly AI is moving from something we type into to something we talk to.

Most people are still used to typing prompts into large language models and other AI tools. But that is not where this is heading. These tools are becoming more conversational, more mobile, and more integrated into the apps and devices we already use.

That matters because the way we interact with AI is going to change.

Typing prompts will still be useful, especially for detailed work. But more everyday AI use will happen through voice. We will ask our AI assistants to summarize information, draft responses, organize ideas, prepare for meetings, review documents, and help us think through decisions while we are working, driving, walking, or sitting at our desk.

So one simple recommendation I’ve been making is this: start getting comfortable speaking to your AI.

Use the dictate button in the prompt box. Use voice mode in the mobile app. Try talking through an idea instead of typing it. Ask AI to help you organize your thoughts after you speak them out loud.

The more comfortable you get with this now, the easier it will be to adapt as these tools become more conversational and more embedded into daily work.

This does not mean every prompt should be spoken. There will still be times when typing is better, especially when precision matters. But for brainstorming, summarizing, planning, role-playing, preparing for a conversation, or getting a first draft started, voice can be faster and more natural.

The future of AI will be less about learning how to type perfect prompts and more about learning how to communicate clearly with a digital assistant that is always available.

9. Today’s AI Pricing May Not Be Tomorrow’s AI Pricing

Another issue business leaders need to think about is the cost of AI.

Right now, many of us are using powerful large language models at prices that are still relatively low compared to the value they can create. But those prices may not reflect the true cost of delivering these tools.

Generative AI is extremely expensive to build and operate. These companies need massive computing power, data centers, chips, energy, engineering talent, and ongoing model development. Many of the companies building the leading AI models are losing billions of dollars or spending billions more than the current economics appear to justify. Even the profitable parent companies behind some tools are making enormous AI infrastructure investments.

In practical terms, users are benefiting from subsidized compute.

That matters because those subsidies may not last forever. As these companies face pressure from investors, public markets, and financial realities, they will eventually need to show that their AI products can become profitable. That could mean higher subscription prices, more expensive premium tiers, usage-based pricing, API increases, or limits on what is included in today’s plans.

That creates a practical warning.

If you are building a process, product, or entire business model that depends heavily on today’s AI pricing, you need to be careful. The economics could change. A workflow that makes sense at today’s cost may look very different if the price increases several times over.

This does not mean businesses should avoid AI. It means they should use AI thoughtfully.

The best place to start is still with efficiency. Use AI to improve existing workflows, reduce repetitive work, speed up standard operating procedures, and help your team produce better work in less time. Those gains are easier to measure and less risky than building an entire strategy around the assumption that AI tools will always be inexpensive.

AI can create tremendous value, but it is not free infrastructure. Businesses should treat AI cost as part of their planning, just like software, labor, marketing, or any other operating expense.

The companies that win with AI will not just be the ones using it. They will be the ones using it in a way that still makes financial sense if the price goes up.

10. As AI Content Increases, Human Connection Becomes More Valuable

The final thing I keep thinking about is that as generative AI gets better, it is also getting harder to know what is real and what is not.

We are seeing more AI-generated content, more deepfake videos, more digital twins, and more generic writing that clearly sounds like it came straight from a prompt box. There is a lot of AI slop out there.

People are using these tools to create content quickly, but they are not always editing it, personalizing it, or making sure it still sounds like them.

And people can tell.

Those who use generative AI regularly can often recognize when something was copied directly from a large language model without enough human judgment behind it. The language feels too polished, too generic, or disconnected from the person who supposedly wrote it.

But it does not have to be that way.

AI should help you communicate more clearly, not erase your voice. It should help you think, draft, organize, and improve, but the final product should still sound like you.

Ironically, as AI-generated content becomes more common, real human interaction may become more valuable than ever.

Face-to-face conversations, trust, empathy, listening, judgment, and relationship-building are not going away. If anything, they are becoming more important. When people are unsure what is real online, personal connection becomes a differentiator.

That is why professionals should not abandon their human skills while learning AI. The real advantage will come from combining both.

If you understand how to use AI and you are also strong at communicating, listening, leading, building relationships, and connecting with people, you will be incredibly valuable.

Do not let AI replace your voice or your relationships. Use it to enhance your work, but keep sharpening the human skills that make people trust you.

Final Thought

After the past several months of speaking across the country, my biggest takeaway is that we are still early, but we are moving fast.

AI is no longer just a technology trend. It is becoming a business issue, a leadership issue, a workforce issue, and a personal development issue.

The organizations that succeed will not be the ones that chase every headline. They will be the ones that create practical guardrails, train their people, improve their workflows, and use AI in ways that make real business sense.

And the individuals who succeed will not be the ones who wait until AI becomes mandatory. They will be the ones who start building the skill now.

The opportunity is still wide open. But it will not stay that way forever.