AI in Plain English: Atlanta’s Real-World Impact

Artificial intelligence is rapidly transforming how businesses operate, and understanding its fundamentals is no longer optional. But where do you even start? Is mastering AI a pipe dream, or can anyone get a handle on this complex technology?

## 1. Understanding the Different Types of AI

First, let’s clarify what we mean by “AI.” The term is thrown around a lot, but it encompasses several distinct categories. It’s not all sentient robots taking over the world (at least, not yet). We generally talk about three main types:

• Narrow AI (or Weak AI): This type of AI is designed for a specific task. Think of your spam filter, a recommendation engine on a streaming service, or even the AI that helps diagnose medical images at Emory University Hospital. It excels at its defined purpose but can’t do much else.
• General AI (or Strong AI): This is the kind of AI you see in movies – an AI that can understand, learn, and apply knowledge across a wide range of tasks, just like a human. We haven’t achieved this yet, though many researchers are working on it.
• Super AI: This is hypothetical AI that surpasses human intelligence in all aspects. It’s the stuff of science fiction and raises significant ethical questions.

For now, most of what we encounter is narrow AI. And that’s where we’ll focus our attention.

## 2. Exploring AI’s Impact in Atlanta

AI is already making waves in Atlanta, and understanding its local applications can make the technology feel more tangible.

• Healthcare: As mentioned, hospitals like Emory are using AI to analyze medical images, helping doctors detect diseases earlier and more accurately.
• Logistics: With Hartsfield-Jackson Atlanta International Airport being one of the busiest in the world, AI is used to optimize flight schedules, predict delays, and manage baggage handling. The Fulton County Airport, while smaller, also uses AI for similar optimization tasks.
• Finance: Several fintech companies based in Atlanta are using AI for fraud detection, risk assessment, and algorithmic trading.

Seeing AI in action locally helps to demystify the technology and understand its real-world potential.

## 3. Building Your First AI Model with Teachable Machine

Ready to get your hands dirty? One of the easiest ways to understand AI is to build a simple model yourself. We’ll use Teachable Machine, a free, web-based tool from Google that requires no coding.

Step 1: Access Teachable Machine.

Open your web browser and go to the Teachable Machine website. Click “Get Started.”

Step 2: Create a New Project.

Choose “Image Project.” You’ll have the option of “Standard image model” or “Embedded image model.” Select “Standard image model.” The embedded model is useful for integrating with specific hardware, but for this exercise, the standard model is perfect.

Step 3: Define Your Classes.

Think of “classes” as the categories you want your AI to recognize. For example, let’s say you want to create a model that can distinguish between pictures of apples and oranges. You’ll need two classes: “Apple” and “Orange.” Click on “Class 1” and rename it “Apple.” Do the same for “Class 2” and rename it “Orange.”

Step 4: Gather Your Data.

This is where you feed Teachable Machine examples of each class. You can either upload images from your computer or use your webcam to take pictures directly.

• Uploading Images: Click the “Upload” button under the “Apple” class. Select several images of apples from your computer. Aim for at least 20-30 images per class for better accuracy. Repeat this process for the “Orange” class.
• Using Your Webcam: Click the “Webcam” button under the “Apple” class. Hold up an apple to your webcam and click “Hold to Record.” Record for several seconds, rotating the apple to capture it from different angles. Repeat this process for the “Orange” class, using an orange.

Pro Tip: The quality and variety of your data are crucial. Use images with different lighting, angles, and backgrounds. The more diverse your data, the better your model will perform.

Step 5: Train Your Model.

Once you’ve uploaded your data, click the “Train Model” button. Teachable Machine will now analyze your images and learn to distinguish between apples and oranges. This process may take a few minutes.

Step 6: Test Your Model.

After training is complete, you’ll see a preview window where you can test your model. You can either upload an image or use your webcam. Hold an apple or an orange in front of your webcam, and the model will predict which class it belongs to. The results will be displayed as probabilities for each class (e.g., Apple: 95%, Orange: 5%).

Common Mistake: If your model is inaccurate, it’s likely due to insufficient or poor-quality data. Go back and add more images, ensuring they are diverse and representative of each class.

Step 7: Export Your Model (Optional).

Teachable Machine allows you to export your trained model for use in other applications. You can export it as a TensorFlow.js model, a TensorFlow Lite model, or a standard TensorFlow model. This is useful if you want to integrate your AI model into a website, mobile app, or other software.

I had a client last year who was skeptical about AI. After walking him through this exact Teachable Machine exercise, he was amazed at how easy it was to create a working AI model. He ended up using a similar approach to build a model that could identify different types of manufacturing defects in his factory, saving him thousands of dollars in manual inspection costs. For more ideas, see how startups break the tech bottleneck for legacy firms.

## 4. Ethical Considerations

As AI becomes more prevalent, it’s essential to consider the ethical implications. AI systems can perpetuate biases present in the data they are trained on, leading to unfair or discriminatory outcomes. For example, facial recognition software has been shown to be less accurate for people of color, raising concerns about its use in law enforcement. A 2024 study by the National Institute of Standards and Technology (NIST) highlighted the importance of diverse datasets in mitigating bias in AI systems.

We ran into this exact issue at my previous firm. We were developing an AI-powered loan application system, and we discovered that the model was unfairly denying loans to applicants from certain zip codes. After digging deeper, we realized that the training data was biased towards wealthier neighborhoods. We had to retrain the model with a more representative dataset to ensure fair outcomes. This highlights why it’s critical to avoid these costly tech mistakes.

Here’s what nobody tells you: AI isn’t magic. It’s a tool, and like any tool, it can be used for good or for bad. It’s our responsibility to ensure that AI is developed and used in a way that is ethical, fair, and beneficial to all.

## 5. The Future of AI

While predicting the future is always tricky, several trends suggest where AI is headed. We’ll likely see:

• More accessible AI tools: Platforms like Teachable Machine are making AI development more accessible to non-technical users. Expect to see more no-code and low-code AI platforms emerge.
• AI embedded in everyday devices: AI will become increasingly integrated into our everyday lives, from smart home devices to autonomous vehicles.
• Increased focus on explainable AI (XAI): As AI systems become more complex, it’s crucial to understand how they make decisions. XAI aims to make AI models more transparent and interpretable.

According to Gartner’s 2026 report on emerging technologies (Gartner), XAI will be a critical differentiator for businesses looking to adopt AI responsibly.

What about Georgia? Look for the state government to expand its AI initiatives. The Georgia Technology Authority (GTA) already has a roadmap for AI adoption across state agencies. Expect to see more AI-powered services in areas like transportation, education, and public safety. In fact, Atlanta’s AI paralysis might soon be a thing of the past.

AI isn’t some distant future; it’s happening now. By understanding the basics and experimenting with simple tools, you can position yourself to take advantage of this transformative technology.

AI is more than just a buzzword; it’s a tool that can empower you to solve problems, automate tasks, and gain new insights. Your first step? Build a simple AI model using Teachable Machine today. If you are still skeptical, read about AI myths that have been debunked.