Understanding AI's Role in Business Tasks

People often ask me what I think about the current state of AI: Do I have reservations about the technology? Do I think it’s even better at “working” than most people think? Or perhaps I think it’s worse? Is it safe? Is it unsafe?

I could go on and on, but I’ll stop here and say my answer to all of these questions is: “Yes.”

If that sounds contradictory, it’s because it is contradictory. Because AI is contradictory in many ways: technologically, ethically, sustainability-wise, etc.

This makes adoption of the technology difficult for most people. I mean, if you were to ask someone (in the moments before you cannonballed into a lake in Florida) if there were alligators in the water, and their answer was “Well…that’s kind of complicated,” would you still dive in?

I wouldn’t.

At Core Concepts, we're not here to make you pro-AI or anti-AI. Instead, we want to show you the nuanced space between these extremes. Let's explore what that means for your business.

What AI Does Surprisingly Well

AI excels at pattern recognition - it's like having an assistant with superhuman perception. While we might miss subtle connections in mountains of data, AI's "eyes" can see hidden trends, its "ears" catch faint repetitions, and its "brain" processes these clues faster than we ever could. Imagine trying to find a specific thread in an enormous web - while humans might get lost, AI effortlessly traces every strand.

Here's what this looks like in your office:

When your sales team needs to understand customer behavior across thousands of interactions, AI spots patterns about which industries are more likely to renew subscriptions. When your legal team faces hundreds of contracts, AI efficiently finds specific clauses and compliance issues. During team meetings, AI can transform meandering discussions into clear action items and deadlines.

Where AI Falls Flat

Despite its strengths, AI has significant limitations. It struggles with context and ambiguity, and real creative thought is still beyond its reach. Take away its data, and you're left with... not much.

This shows up in your workplace when:

You need someone to mediate a conflict between employees about workload - AI can't grasp the emotions and cultural nuances at play. When your product team needs truly innovative ideas, AI can only remix existing concepts. During an unprecedented system crisis, AI falters because it can't think outside its training data.

A Simple Way to Decide When to Use AI

Before diving into any AI project, ask yourself:

1. Is this task repetitive and time-consuming?

2. Are we dealing with loads of data?

3. Would doing this faster actually matter?

4. Will humans still need to oversee the final decision?

If you answered "yes" to most of these, you might have found a good AI opportunity.

Not-so-shameless-plug: This is something we’ll dive into in our AI Essentials Workshop Series with the University of New Hampshire. You can see the schedule here: https://training.unh.edu/programs/certificates/ai-essentials-certificate-program

Key Concept of the Week: Automation vs. AI

Let's clarify an important distinction: while automation and AI often work together, they're not the same. Automation follows pre-defined rules to complete tasks (like automatically sending welcome emails to new subscribers). AI, however, learns from patterns and can adapt its responses (like analyzing customer service emails and generating personalized responses based on tone and content).

Generative AI Tool of the Week: Gamma

Gamma AI is a design tool that allows you to create presentations and documents in minutes. In fact, Gamma AI has shown a 40% boost in audience participation compared to traditional presentation methods.

It’s a tool we use a lot at North Light AI, not so much to generative content as to structure it, both visually and informationally.

Slide decks are highly pattern-driven and ideal for AI automation because:

Structural Patterns

• Standard flow: overview → main points → conclusion

• Often there are consistent slide layouts for different content types

• Predictable content hierarchy within slides

Design Rules

• Brand guidelines dictate colors, fonts, logos

• Visual design principles (rule of thirds, contrast ratios)

• Standardized spacing and alignment

Content Patterns

• Bullet point progression and grammar

• Common headline formulas

• Industry-specific terminology

• Data presentation conventions

Creation Process

• Repeatable steps: outline → template → content → formatting

• Context-specific requirements (sales vs. training)

• Brand compliance checking

This systematic nature means AI can learn from millions of presentations to generate contextually appropriate slides while following established design and content patterns.

Random Fun Fact of the Week

The standard size and capacity of CDs was determined by Beethoven's Ninth Symphony.

In the 1980s, during CD development at Sony, vice president Norio Ohga (a trained conductor) insisted that a CD must fit the entire symphony without interruption. The longest known recording was 74 minutes, so this became the CD's capacity specification…roughly 650 megabytes of data. This classical music requirement ended up shaping not just audio CDs, but also CD-ROMs, influencing early computer storage, software distribution, and video games.

I should say there are some who debate the accuracy of this anecdote. To those people we say: But we got this from AI! How could it be wrong?

Contact us at NorthLightAI.com to learn how we can help you build a stronger data foundation for your AI future.