The AI Revolution: DIY Edition

Author

Cat Valverde, MBA
March 06, 2025

Hi There, AI Adventurers!

We knew this day would come—building your own AI assistant is no longer just for enterprises with million-dollar budgets. Meanwhile, Google is flipping the script on search, testing AI-only results. And in the background, a big question looms: How much energy will all this AI actually consume?

Let’s dive in.

Enterprise AI Solutions // Created with Midjourney

How to Build Your Own AI Assistant

Creating your own AI Assistant is easier than you’d think. Today both businesses and individuals can create their own AI assistants without needing a PhD in machine learning. Here’s the rundown:

Step

What to Do

How to Do It

1. Define the Purpose

Decide what your AI assistant will do.

- Will it answer FAQs, schedule meetings, help with coding, or manage tasks?
- Ex: “I want an AI assistant to help me summarize emails.”

2. Choose a Platform

Pick a tool or framework to build it on.

- No coding? Use ChatGPT Custom GPTs, Google Bard, or Microsoft Copilot.
- Some coding? Use OpenAI API, Rasa, or LangChain.

3. Set Up a Development Environment (If Coding)

Get the right tools to build.

- Install Python (Anaconda, VS Code) or use a cloud service like Google Colab.
- If using APIs, get an API key from OpenAI or other AI providers.

4. Train Your AI (Optional)

Customize the AI to your needs.

- Upload files or text it should learn from.
- Use embedding models to improve context retention.

5. Build a Chat Interface

Decide how users will interact with it.

- For personal use: Keep it in ChatGPT.
- For a website/app: Use a chatbot framework like Dialogflow, Rasa, or BotPress.

6. Connect to Other Apps

Make the AI assistant more useful.

- Use Zapier or APIs to connect with Google Calendar, Slack, or Notion.

7. Test and Improve

See how well it works and refine.

- Try different questions, fix incorrect answers, and update its data.

8. Deploy and Use

Put it to work!

The verdict? Building tailored AI assistants that enhance productivity is getting easier. But without a clear plan, you’ll end up with an over-engineered chatbot that no one wants to use.

You Knew It Was Coming: Google’s AI-Only Search

Google just announced testing for AI-only search results, meaning no traditional blue links—just AI-generated answers. This is massive.

Why it matters:

  • SEO shake-up – If this sticks, websites that rely on search traffic (aka, most of the internet) are about to experience turbulence.

  • Accuracy concerns – Google’s AI-generated answers are known to be hit-or-miss. What happens when there are no links to fact-check?

  • Advertising impact – If fewer users click search results, Google’s ad revenue model might need a serious rethink.

For now, this AI-only mode is just a test, but if Google goes all-in, it could change how we access information online.

How Much Energy Will AI Really Consume?

AI isn’t just eating up computing power—it’s consuming massive amounts of electricity. Researchers are calling for greater transparency in how much energy AI models actually use.

The good: AI is optimizing energy grids, reducing waste, and improving efficiency in areas like logistics and manufacturing.

The bad: AI training and inference require data centers that suck up power. Estimates suggest that AI-related electricity demand could match that of small countries within the next decade.

The unknown: Companies like OpenAI, Google, and Microsoft aren’t exactly forthcoming about the full environmental impact. Without transparency, it’s hard to know how sustainable AI’s growth really is.

Expect more regulatory scrutiny as this conversation heats up.

Are AI-generated search results better than traditional search?

Login or Subscribe to participate in polls.

Enterprise AI Solutions // Created with Midjourney

AI is Changing How People Plan Travel

Booking flights and hotels is getting a serious AI upgrade. According to NBC, travelers are using AI for:

  • Itinerary generation: AI scrapes through reviews, maps, and recommendations to build custom travel plans.

  • Flight and hotel deals: AI monitors price trends and suggests the best time to book.

  • Local insights: Chatbots can now act as on-the-go tour guides.

The downside? Personalization isn’t perfect yet. AI might recommend an off-season trip to the Amalfi Coast… where everything is closed.

TL;DR

  • Build your own AI assistant – It’s easier than ever, but plan ahead.

  • Google’s AI-only search test – The SEO world is watching closely.

  • AI’s energy demands – We need more transparency.

  • Travel AI is here – Expect smarter bookings, but with quirks.

Closing Thoughts

Between DIY AI assistants and AI-driven search, we’re shifting into a world where AI doesn’t just suggest things—it decides them. The question is, do we trust it?

Stay sharp,

Cat Valverde
Founder, Enterprise AI Solutions
Navigating Tomorrow’s Tech Landscape Together