In this issue of The AI Edge
🔥 AI Economy Boom, Bubble or Breakthrough? — AI now accounts for 92% of economic growth in early 2025, driving trillions in data-center investment and reshaping the global economy. But are we in an AI bubble—or building the infrastructure of the next industrial revolution?
🎯 AI in the Wild: Digital Ports, Real Impact — At the Port of Corpus Christi, AI isn’t just analyzing data—it’s simulating the future. Thousands of daily “what-if” scenarios predict everything from tanker collisions to cyberattacks, cutting costs and transforming operations.
💬 The AI Takeaway: The Signal vs. The Slop — Half of all online content is now AI-generated. As attention spans shrink, deep focus and critical thinking become the new superpowers. The future belongs to humans who can cut through the noise—and create meaning no algorithm can replicate.
🔥 Signal, Not Noise
AI is eating the world. AI accounting for 92% of economic growth during the first half of 2025. It is creating it's own economic slice of the pie between tech companies and chip companies.
The growth isn't just in the digital world either. Data centers are being built at a staggering pace, and it is forecasted that an additional $2 trillion in investment will be needed by 2030 to build all of them out.
Of course the conclusion is now that we're in an AI bubble. Or so the news would have you think. But are we really? That is tough question to answer. There are frequent references to the dotcom era, where companies that weren't profitable had staggering levels of investment. Do we have that today? Absolutely. What is different this time around is the usage of LLMs and other forms of AI, and how they are being applied in companies.
The value proposition is simple: if I can make an investment in AI that will give me positive ROI, I'm going to do it. It is becoming less cost prohibitive to make these investments. A decade ago, if you weren't a very larger company, there was no way you were going to scale your AI capabilities. You wouldn't have enough capital to do it and there was a level of career risk that you were taking. Not today. Getting a ChatGPT paid license for $20 changes this dynamic.
I also don't think AI is in a bubble because of the amount of physical investment, particularly in data center investment. The economics of data centers are very different than microchips. You don't make an investment in a data center and expect high returns in 3-5 years. In many cases, you're looking at a 15-20 year investment in power arbitrage, real estate and infrastructure. Additionally, the labor, support and maintenance of keeping it running. No executive team worth their salt is going to be making multi billion dollar investments in data centers to lose money.
The demand is picking up as well. You can see it in microchip purchases and investments, along with electricity costs. Data centers require power companies to make infrastructure investments to support them, and some of this cost is passed down to the consumer. In an ideal world, increased demand would also lead to more efficient energy product (ie, nuclear).
Yet I could be completely wrong. A lot of people were saying "this time is different" in the dotcom era, and then it wasn't. What emerged were battle tested companies like Amazon and Google. And other companies like Microsoft transformed their business models significantly. Even if there isn't a bubble, there will be a downturn due to the standard business cycle. Yet what will emerge is more trillion dollar technology and AI companies that will power the future.
📌 Quick Hits
AI Gets a Body — 1X building humanoid robots for the home, blending large scale AI with real world mobility. Their NEO robot doesn't just respond. It walks, lifts, and learns like a physical AI assistant. The era of "AI you talk to" is shifting to "AI that lives with you." Read more →
AI Megafactory Launch — Samsung Electronics and NVIDIA are teaming up to build a sprawling AI factory powered by 50,000 NVIDIA GPUs, aimed at making chip manufacturing, robotics and digital twins fully agentic. The move signals that the next frontier isn't just AI in software, it's AI building the hardware itself. Read more →
Sora Ignites a Firestorm - OpenAI's new video generation app Sora is breaking download records while sparking major copyright debates. Its hyper-realistic clips are blurring lines between creator and machine, and leaving artists and studios scrambling to define ownership in an AI-driven world. Creativity just got a new disruptor. Read more →
🧰 Prompt of the Week
We're going to level up this weekly slightly with this prompt. One of the most frequent questions I get asked is "how do I transform a manual process into an AI-driven one?". A tough question to answer, but one where an LLM can be a great guide. It won't do all the work for you, but will accelerate your ability to improve your current process. Here is the prompt below:
I want to transform a manual business workflow into an AI-driven, automated, and scalable one.
Context about the process:
[Paste the steps, tools, roles, and pain points here]
Your tasks:
Rewrite the workflow for an AI-first operating model
Identify steps to eliminate, automate, or augment
Suggest AI agents + human checkpoints
Recommend tools, integrations, and data dependencies
Provide a 1-week execution plan to pilot it
Output formats:
Executive workflow map
RACI-style AI+Human responsibility table
KPIs to measure impact
Optimize for speed, low effort, and measurable business value. Assume I am a [insert role here] driving value through AI adoption.
🎯 AI in the Wild
The Port of Corpus Christi, the largest energy export gateway in the U.S., is using AI to literally see the future. Its new OPTICS platform creates a dynamic digital twin of the entire port, combining satellite feeds, sensor data, and logistics records into a live simulation of ship traffic, weather conditions, and hazardous cargo movement.
The system isn’t just for monitoring, it also makes predictions. AI models run thousands of simulated “what-if” scenarios every day: tanker collisions, pipeline leaks, cyberattacks, or extreme weather disruptions. Each simulation helps identify weak points in real-world operations and trains emergency crews before a crisis hits.
Beyond safety, the economic implications are enormous. By optimizing shipping routes, predicting maintenance needs, and reducing downtime, the port expects to cut annual operational costs by double digits. This is what real digital transformation looks like, not another dashboard, but an intelligent operating system for the physical world.
💬 The AI Takeaway
AI generated content is outpacing human created content:

You hear the term "AI slop" and now 50% of the time that is true. Straight copy and paste from ChatGPT, with all the emojis and perfectly formatted hashtags. Or even completely ripped off from an original source.
It changes the way we process and understand information. Before the internet age, knowledge and information was synonymous with power. Having an asymmetrical advantage with hard to find information was a differentiator. The problem today is that information doesn't equal knowledge. We are bombarded by lots of information with a very low level of signal.
This creates a new opportunity for differentiation. The doomscrollers are literally rotting their brain because of information overload, and it is shortening attention spans each year. Successfully adapting to this new future requires you to choose where you focus your attention wisely. If you're consistently a slave to notifications and emails, you're not going to get much done. You'll also find it more challenging to have a deep level of focus. Yet it is that deep level of focus where the magic happens. Great ideas don't get formulated in 30 second windows.
The other important skill will be critical thinking. It is safe to say today that when you see a Facebook post or X tweet, assume that it is false or misleading. Only after verifying it should you believe it. The algorithms for any social media platform are optimized for engagement, and this usually means posts that elicit extreme emotional reactions in their users.
The other opportunity this new world creates is the premium that will be placed on well written, well though out human content and experiences. Counterintuitively, those experiences that cannot be scaled easily will become the most valuable. Think of great artistic masterpieces. They are essentially a 1/1, where one individual can own the original. It is possible to do this with digital experiences as well, though slightly more challenging. Expect true hyper-personalization at a 1/1 level.
Lastly, it will make all of us question what it means to be human. What are the truly human experiences that AI cannot replicate or replace? This answer is different person to person, and only in your individual exploration will you find the answers.
-Ylan

