AI for Professionals

Leaders who win will not be the ones who mastered every tool. They will be the ones who stayed in the cockpit while everyone else was under the hood. AI gives you the altitude to see patterns, compress intelligence, and make faster decisions with better information. I am talking about turning 40 pages of market data into a one-page strategic brief in minutes and actually acting on it. The bird's eye view has always separated great leaders from busy ones. AI just raised the ceiling on how high you can fly.
- Thea Ducrow, A Bird’s AI View

I think the answer to this is twofold: (1) It's to think “AI First” which means you need to think about how to get a job done with AI first. (2) It’s to understand the risks associated with AI. It’s not something that should be used without guardrails but should be used to augment human judgment.
- Will Wright, Health Rounds

I believe the most important skills are our mindset around AI – this includes being open-minded to how AI can enhance your role and the success of your team and organization. I also think adaptability and change-resilience are critical skills. Finally, it will be critical to protect your own original thinking, creativity, ideas, and decisions – so we need to make sure we aren’t outsourcing this to AI or we get further from our intuition and our original thinking.
- Vitale Buford Hardin, Hardin and Company

The biggest mindset shift is "AI can/can't do this" to "What do we have control over to get better results from AI?" We teach the ACTIVE framework at Caravan. The part of the acronym most relevant to you, the things you have the most control over, are Context, Tools, and Instructions — what does AI know about you/your business(memory), what tools can it work with (connectors), what instructions have you given it (skills)? Employees that make the most progress this year will learn how to pull these levers, and the best leaders will find ways to teach people this approach. A big challenge for everyone is that learning AI isn't like learning traditional software, it's more like learning to delegate, set clear expectations, provide the right resources, and give good feedback. Everyone is a manager now, so what will you do with your ten new (digital) employees?
- Austin Senseman, Caravan AI

The biggest mistake most organizations make is starting with the tools and/or buying tools that are popular, instead of starting with the business problem you’re trying to solve. If you just chase the latest technology or what’s hot, you’ll end up with a complicated, fragmented technology stack that nobody uses. Start with a high-value use case anchored in your organization’s real-world business problem. Where are your biggest inefficiencies, or where would decisions benefit from better insight? Then match the use case to the AI tool’s capabilities. For example, will you need Generative AI, Predictive AI or Automation +AI to solve the use care. This prevents overbuying tools that don’t actually solve your problem.
- Molly Wasko, Ph.D., university professor, Collat School of Business

I would try not to start with the tool. The better question is: what’s not working well today? Once the problem is clear, the options narrow quickly. Fit matters more than sophistication. A tool people will actually use, and that integrates smoothly, usually beats a more advanced one that creates friction. I also prefer small pilots - you learn very quickly what works in practice.
- Mohammad Firouz, Ph.D., associate professor, Collat School of Business

It depends on the use case of the organization. It is always best to start with a subscription to GPT, Claude, and Perplexity and then see which one you like more and which one can do more for you.
- Anurag Banerjee, Quilt AI

Start with the problem, not the tool. Evaluate each option on fit to workflow, total cost of ownership, vendor trustworthiness, and whether you can define success before you start. Always pilot with a real team before committing organizationally.
- Chris Edmonds, Ph.D., professor, Collat School of Business

The best way to cut through the hype is to put your hands on the keyboard and try to solve a meaningful problem. For individuals, download Claude Desktop, open up Claude Cowork and ask the following, "help me build a skill that solves a problem I work on every day, something that takes me an hour that we can cut down to 5 minutes. I'm tired of the hype, I want a practical result." For organizations, you can turn on logging in Claude for your organization, let it run for a week, and ask "who is solving practical problems using AI, what types of things are they working on, what can I do next week to encourage more practical, productive outcomes with AI?" The biggest unlock is this … Whatever question you have, that's the question to ask AI.
- Austin Senseman, Caravan AI

The hype is real, but it can be distracting. I’d shift the focus to: what specific problem are we solving? Start with a few meaningful, manageable use cases and do them well. Also, AI won’t fix weak processes or messy data - it tends to amplify them. Teams that ground their efforts in real operational needs see much better results.
- Mohammad Firouz, Ph.D., associate professor, Collat School of Business

Is there hype because it sounds exciting, but they don't understand how to get started or use it for their organization? Identify challenges, bottlenecks, or routine time-consuming tasks and ask AI for assistance. Jump in. Teams could explore and report back on their progress or solutions. It can move from hype to a fun and engaging exercise.
- Rush Neil, senior instructor, Collat School of Business

When I was Executive Director of the State Nursing Association, I did not ask for permission to build a $2M PPE distribution network. I built a small system, showed results in weeks, and the partners followed. AI is no different. Hesitant leaders do not need a TED Talk, they need a win they can see. Start small, solve a real problem they already care about, measure it, and share it publicly. Buy-in is not built in a boardroom. It is built one visible win at a time.
- Thea Ducrow, A Bird’s AI View

Consider this a pitch like any other. What are the concerns or reasons for resistance? Is it a comfort level with the technology, an ethical concern, or a lack of understanding what AI could offer the organization? Use a business case where the outcome is demonstrated, and the difference in performance/productivity is measurable and clear. If guardrails are required, that is an easy next step to begin integrating AI into the organization. Assume some may need training or hands-on instruction, depending on their comfort level with technological advancements. 
- Rush Neil, senior instructor, Collat School of Business

I believe strongly in showing versus… telling. Anyone trying to convey the power of AI to an antagonist is going to come across the wrong way. However, selecting an agreed upon problem (organization, data display, data analytics) and working collaboratively to show the power that AI can unlock— that’s transformational. The key there is the shared-problem framework. If everyone can agree that the use-case is indeed a problem worth tackling, then when AI shortens the time to solution doubters can move to advocates.
- Ben Meadows, Ph.D., associate professor, Collat School of Business

This is context dependent, but generally you’re looking to show improved output for a given input. So things like customer questions answered for a contact center chatbot, customers acquired per employee, or code written per engineer are all easily measurable pre/post AI. The best way to demonstrate business value is to test out the process against a control. If that test case shows value, then you can scale this up
- Will Wright, Health Rounds