AI in Business 2025: Real-World Impact, Deepfake Risks, and Life-Saving Innovation

Dr. Ryan Ries here. I’ve been in this industry for over 20 years, and I am still amazed by how quickly technology has moved in this space over the last 5 years. Particularly when we look back to where we were just a year ago.

For example, I saw this video on Reddit showing how much AI-generated video has improved over the last two years. Remember the trend of Will Smith eating spaghetti?

That progression from laughable to convincing happened fast. But this week, we saw what happens when AI video quality crosses from "impressive demo" to "prime-time television that nobody questions" and “geospatial analysis saves lives at scale.” 

When AI Stunt Journalism Gets Too Real

U.K.'s Channel 4 just pulled off something that should make all of us pay attention. They aired an entire hour-long documentary about workplace automation called "Will AI Take My Job?" 

But here’s the twist… The presenter was completely AI-generated, and nobody knew until the final reveal.

Here was the presenter's mic drop: "AI is going to touch everybody's lives in the next few years. And for some, it will take their jobs. Call center workers? Customer service agents? Maybe even TV presenters like me. Because I'm not real."

Channel 4's head of news explicitly stated they won't make this a habit because "AI is not capable" of the "premium, fact-checked, duly impartial and trusted journalism" they deliver. The news station said they tried this stunt to show just how indistinguishable AI has become from the real thing. They also wanted people to realize that they should question what they’re seeing.

Why This Matters in Business

Think about your business for a second. If an AI can host an hour of prime-time television convincingly enough that viewers don't notice, what does that mean for your customer-facing roles? Your training videos? Your internal communications?

The interesting business question isn't whether AI can do these jobs. Channel 4 just proved it can. The question becomes where human judgment, creativity, and accountability create value that AI genuinely cannot replicate.

Google Earth AI: From Interesting to Indispensable

Google announced Earth AI this week, and it’s very impressive. They're taking decades of geospatial modeling work and making it actionable through Gemini-powered reasoning.

Here’s their value prop, which quickly explains what Earth AI can do: "Breakthroughs in understanding the Earth that previously required complex analytics and years of iteration are now made possible in a matter of minutes."

Earth AI’s Impact

To get more specific, they have use cases on their landing page that explain the massive impact this technology has. 

Google's flood forecasting models now cover more than two billion people with life-saving predictions before significant river floods. Their MetNet nowcasting model predicts precipitation with high accuracy using satellite data, filling gaps in radar coverage. 

For organizations working on infrastructure planning, environmental monitoring, disaster response, or supply chain logistics, this shifts what's possible. With natural language processing you can ask questions like the examples they gave:

• "where are all the storm drains within 50 meters of public schools in Red Hook, Brooklyn?" 
• "identify buildings in flood-risk zones across this region"

This week, I’m sure by now you’ve seen the devastating impact of Hurricane Melissa in Jamaica. Technology like Earth AI gives me genuine hope that we're building better tools for emergency response and preparedness.

What This Means for Your AI Strategy

These stories share a common thread. AI capabilities are moving from "impressive in demos" to "deployed at scale and creating real impact." 

The strategic question for your organization isn't whether AI will impact your business. That's settled. The question is whether you're actively shaping how that happens or just reacting as competitive pressure forces your hand.

If you want to dig into what this means for your specific situation, let's talk. I'm seeing patterns across industries that might save you from expensive mistakes.

Before we go, here’s quick plug for our webinar tomorrow: MCP vs. RAG. Jonathan (Mission’s CTO), and Casey Bleeker from SurePath AI are explaining how to choose the right architecture for your business use case. Make sure to sign up here.

Until next time,
Ryan

Now, time for this week’s AI-generated image and the prompt I used to create it:

I want you to turn me into a news reporter. Use this background in the image I am attaching. I have also attached a picture of what I look like. I want me as a news reporter to look exactly like me.

I decided to try this ChatGPT vs. Gemini. 

Here’s ChatGPT’s:

Then I complained that it aged me 10 years and told it to try again but it just made me look even older!

Thank you to Gemini for saving the day: