In recent months, multiple clients, co-workers, and friends have come to me with their concerns about artificial intelligence. We are living in a new world that is becoming increasingly dependent on it.

Think about how many times you have consulted ChatGPT. Too many? Do not worry. You are not alone. Millions of people are turning to AI powered chatbots to solve complex business challenges and even deeply personal questions. You know something is in demand when people are willing to pay a subscription for it.

Among consultants and entrepreneurs, especially those in the services space, I have noticed a mixture of curiosity, uncertainty, and sometimes fear. Below are some of the most common questions I have encountered in my consulting and coaching practice, along with the responses that many have found helpful.

How can AI support my business?

About a year ago, this was one of the most common questions I received. Today, AI has integrated into nearly every digital platform we use, so I hear it less often. Still, it is important to understand its influence.

Much of what people are using today falls under what is called generative AI. It creates content. It is especially strong at written communication through AI assistants and chatbot platforms like ChatGPT. Through advanced language models, AI can communicate in nearly any language, including software code.

Beyond text, AI can now generate images, audio, and other media. Many smartphones use AI to enhance photos. One feature I personally appreciate is the ability to remove an object from a picture and have the background fill in naturally.

For client facing roles in business, AI can generate pitch decks in seconds. With clear and specific prompts, you can build organized charts and tables that once required hours of formatting and revision.

That said, many of my engineering clients ask about using AI for mathematical computations. It is important to understand the fundamental nature of generative AI. It is designed to recognize and reproduce patterns in language. It is not inherently built to follow strict mathematical rules the way a symbolic math engine or calculator does. Because of this, AI can produce incorrect numerical results. These are commonly referred to as hallucinations. For that reason, calculations should always be verified.

To improve reliability, modern AI systems are increasingly connected to external calculation tools, code execution environments, and specialized agents. Instead of guessing a result based on patterns, these systems can call a mathematical engine, run structured code, and then return a computed answer. This significantly reduces error rates. Even so, for safety critical engineering work, careful human review is still essential.

What is the business impact of AI?

This question challenges leaders to reconsider how they define productivity and efficiency.

For example, in sales, AI is not closing deals. People are. However, AI can dramatically reduce the time spent gathering information, preparing summaries, drafting proposals, and organizing communications. That saved time allows people to focus on conversations and decision making.

If a company does not measure time saved or improved efficiency, the value of AI may be difficult to quantify. Measuring AI often comes down to tracking hours saved. For salaried employees, that can be sensitive without creating a feeling of being micromanaged. Ideally, AI reduces repetitive work. However, if prompts are unclear and require multiple revisions, time can quickly be lost in the process.

How will AI impact jobs?

The full impact of AI on employment is still unfolding. From a workforce perspective, we can look at this through two lenses: jobs reduced and jobs created.

Much of the public concern focuses on job loss. In the past year, several major technology companies have announced significant workforce reductions. Amazon has laid off 30,000 employees since October 2025. Intel reduced its workforce by 25,000 employees in 2025. Microsoft laid off 15,000 employees in 2025. Headlines often attribute a large portion of these reductions to AI. At the same time, it is difficult to separate the influence of AI from broader economic and political factors.

There is another side to the conversation. AI systems require advanced microchips, significant computing power, and large amounts of energy. According to the World Economic Forum, a single AI query can use substantially more electricity than a traditional internet search. One ChatGPT query uses enough to energy a lightbulb for a few minutes (1). Increased energy demand means expanded power generation. Large data centers require land, infrastructure, and long-term investment. Companies investing in AI are purchasing property in rural areas to support these facilities. In that sense, AI is not only reshaping digital work but also physical infrastructure.

What can be trusted with AI?

AI continues to improve as it is trained on larger datasets and refined by engineers. Even so, responsibility always rests with the person applying its output in the real world. When thinking about trust, I find it helpful to take a risk-based approach.

Low risk activities include grammar checks in emails, drafting social media content, or creating marketing materials. If an AI generated flyer contains an error, it may cost some leads, but it can be corrected. A quick review is typically sufficient.

High risk activities include engineering calculations involving margins of safety or medical diagnoses. These decisions directly affect human life and require careful validation. Many high risk tasks rely on precise mathematical rules. Generative AI is built to create patterns, not to invent new mathematical truths. There is only one correct value when squaring a number. In one consulting engagement, I tested AI by asking it to calculate the load that would cause an object to buckle. I provided inputs in both SI and US units. The output was off by several orders of magnitude. When I requested step by step calculations, I noticed it had computed ten to the fourth power incorrectly. Only after asking it to verify that specific step did it recognize the error and correct the result.

That experience reinforced something simple. AI can support analysis, but it should not replace professional judgment in safety sensitive situations.

Will AI replace me?

We are in a transitional period. There are more unknowns than knowns, and uncertainty naturally creates discomfort. If you are asking yourself this question, you are not alone.

At its core, this is a question about identity.

Who were you three years ago before AI became a daily topic of conversation? Who do you want to become five years from now?

A strengths-based perspective can help. Ask yourself how AI might support you in becoming the person you want to be. If you feel uncertain, reflect on how you have handled uncertainty before. The same adaptability applies here. We live in a society that depends heavily on technology. Still, we remain individuals with judgment, creativity, relationships, and character. Our value extends far beyond the tools we use.

When you truly understand who you are and what you uniquely contribute, you may find that you are far more irreplaceable than you originally thought.

Reference

1. World Economic Forum, www.weforum.org/videos/energy-use-chatgpt/