Artificial intelligence has changed remarkably quickly.

The conversation is no longer simply about ChatGPT or whether AI can write an email. Today's AI tools can help research information, analyze documents, work with images, write computer code and carry out increasingly complicated, multi-step tasks.

And while much of the AI conversation can sound as though it's happening somewhere far away in Silicon Valley, China or an imaginary place called "the cloud," some of the infrastructure making the AI boom possible is being built right here in Round Rock.

But first, there's a new AI term you're probably going to hear more often.

What in the world is "open-weight AI"?

It sounds more complicated than it is.

Most familiar AI services—such as ChatGPT, Claude and Gemini—are systems you access through a company's service.

With an open-weight AI model, the trained model itself is made available so developers and organizations can download it and, depending on its license, run or customize it themselves.

Think of it this way:

Instead of only being allowed to eat at the restaurant, you're being given access to much more of the kitchen.

You still need the equipment and know-how to use it, and "open-weight" doesn't necessarily mean that everything about the AI is open or free.

Why should an everyday AI user care?

More choices.

Open-weight models can give organizations more control over customization, privacy and where their AI operates. They also increase competition among AI developers.

Chinese companies have become particularly important in this part of the AI industry. Developers including Moonshot AI, DeepSeek, Alibaba's Qwen and Z.ai are competing in a rapidly developing market. Reuters reported this month that Chinese AI developers have remained relatively close to leading U.S. labs in model sophistication while competing heavily on efficiency, pricing and downloadable open-weight models.

That means some of the technology influencing the next generation of AI tools isn't coming only from the familiar American technology companies.

So what does this have to do with Round Rock?

Quite a bit.

Round Rock is home to Dell Technologies, and Dell is increasingly involved in building the computing infrastructure required for AI.

In July, the Texas A&M Engineering Experiment Station selected Dell to design and build IGNITE — the Innovative Growth in Next Generation AI Technology Ecosystem. The platform will support AI and high-performance computing for engineering, scientific, national-security and other research.

And the infrastructure story gets even closer to home.

According to the City of Round Rock, the city currently has two major data-center developments.

Sabey Data Centers has redeveloped the former Sears call-center property on Louis Henna Boulevard. Its campus is planned to eventually reach approximately 430,000 square feet and will host the Texas Advanced Computing Center's Horizon supercomputer.

Switch is also developing a large data-center campus on property purchased from Dell Technologies.

Horizon is particularly interesting. The Texas Advanced Computing Center selected Round Rock as its home, and the system is expected to become the largest academic supercomputer in the National Science Foundation's portfolio when operational.

AI doesn't really live "in the cloud"

That's where the local connection becomes easier to understand.

When we ask an AI system to analyze a document, create an image or answer a complicated question, the work has to happen on real computers in real buildings somewhere.

Those computers require electricity, cooling, networking equipment and enormous data-center capacity.

And that growth is creating some very real questions for Texas.

On Sept. 21, Gov. Greg Abbott directed the Texas Commission on Environmental Quality to halt permits sought by data centers until ERCOT completes an audit. The state is examining data-center effects on electric-grid reliability and water resources, among other requirements.

So the AI boom isn't just a technology story anymore.

It's also becoming an energy, water, infrastructure, economic-development and community-planning story—including here in Central Texas.

What does all of this mean for the rest of us?

You don't need to build a supercomputer—or understand how one works—to benefit from AI.

For a Round Rock small-business owner, AI might mean answering routine customer questions, analyzing information, creating marketing materials, improving customer service or automating repetitive office work.

For the rest of us, it might help plan meals, understand a complicated document, research a purchase, organize a trip, brainstorm a project or tackle something we've been putting off.

And we're gaining choices—not only between familiar names such as ChatGPT, Claude and Gemini, but between different kinds of AI designed for different jobs.

One might be particularly useful for research. Another for writing. Another for analyzing documents. And some models can be operated privately inside an organization's own systems.

The more useful question may soon become:

Not "Should I use AI?" but "Which AI should I use for this particular job?"

That's what we'll continue exploring in AI in Everyday Life—what's changing, why it matters and, most importantly, what you can actually do with it.