AI Inflection Point

Artificial intelligence… are you tired of hearing about it? You might be skeptical, thinking it’s just another buzzword that will fade into the background, or maybe you acknowledge its potential but can’t quite gauge its impact.

Either perspective is valid, especially when you consider the numbers. Take the Q2 earnings season, for instance. An analysis of the quarterly calls from companies in the S&P 500 revealed a staggering 7,358 mentions of AI-related terms. That’s a 4.6x increase from Q1 and more than triple the mentions in Q4 2022.

Interestingly, investors seem to be riding the AI wave, favoring stocks of companies that talk up AI over those that don’t. Call it irrational exuberance if you will.

Or perhaps, it’s something else.

We’re at the cusp of an inflection point—a technological sea change. Some might label it a revolution; others might see it as a mere transition, but either way, it's reshaping our past, challenging our longstanding beliefs, and redefining the future.

History reminds us that inflection points have always heralded major shifts, whether it was the advent of fire, electricity, the automobile, or more recent disruptors like the internet and social media. And now, AI joins that list.

Change is inherently uncomfortable. It tells us that our familiar view of the world is outdated, that our deep-seated beliefs are up for debate. Our instinct is to resist, to cling to the known rather than embrace the unknown.

Biologically, we seek homeostasis—a stable internal state amidst external fluctuations. When it's disrupted, we react—sweating to cool down, shivering to warm up. Inflection points, by their very nature, disturb our equilibrium, prompting reactions of avoidance or defiance.

Yet, eventually, we adapt, finding a new balance, just as we always do with technological upheavals—given time, acceptance, and a tangible improvement in our lives.

For a technology to truly hit an inflection point, it must tick three boxes: power, data, and intelligence.

Power is fundamentally about the computing might, the raw processing power that breathes life into technological aspirations. At the heart of this power are processors, the workhorses of computing. The processor is a marvel of modern engineering, an intricate tapestry woven with millions of transistors—tiny electronic switches that regulate the flow of electricity and thus the flow of digital information.

The evolution of these processors is a testament to human ingenuity. If we journey back to the 1970s, we find the early microprocessors carrying a few thousand transistors. Today, we have embarked on an era where processors are embedded with billions of transistors. It's not just about the sheer quantity; these transistors have shrunk to the nanometer scale, allowing more to fit on a chip, thereby increasing their performance and energy efficiency.

Take for instance, the first 4004 processor introduced by Intel, which was a 4-bit microprocessor containing merely 2,300 transistors. Now, we stand in an era where processors like Apple's M1 chip boast 16 billion transistors. This monumental growth in transistor count, known as Moore's Law, has been a guiding star for the industry, predicting a doubling of transistors on a chip approximately every two years, leading to continuous advancements in processing power.

This is also the reason we have moved from graphics like Tecmo Super Bowl on Super Nintendo to realistic VR graphics like this one seen between Mark Zuckerberg and Lex Fridman. (LINK )

The cost of these transistors, and hence the cost of processing power, has witnessed an inverse trajectory. The average cost of a transistor in the 1970s was around a dollar, a price that has plummeted over the decades to fractions of a penny today. This drastic reduction in cost is akin to a form of economic reverse inflation, making powerful processors increasingly accessible.

It’s this interplay between the increasing number of transistors and the decreasing cost that has powered the computational leaps we’ve witnessed. The speed at which these processors operate has soared, and the storage capacity they support has expanded exponentially. This allows for the complex computations required for deep learning and other AI processes to take place efficiently and at scale.

Merging power with data and intelligence, we see that AI isn't new—it’s been around since the 1950s, evolving from simple rule-based operations to learning from vast swathes of data, culminating in AI systems like IBM’s Watson and the AI that defeated the world Go champion in 2016.

But power, data, and intelligence aren't enough. A technology also needs accessibility, applicability, and affordability to truly revolutionize our everyday lives. This is where AI shines today.

Consider the evolution of GPS technology. It was a military tool long before it became a civilian staple. It wasn't until smartphones integrated GPS that it became ubiquitous, thanks to improved accessibility, applicability, and a price point that made it practical for the average person.

AI mirrors this trajectory.?

Accessibility: AI is at our fingertips. ChatGPT converses with us through a chat interface, accessible via an app on our phone—simplifying access to artificial intelligence.

Applicability: AI is now versatile, aiding in everything from culinary creativity based on your pantry's contents to crafting job descriptions for potential hires.

Cost: Unlike its expensive predecessors, AI platforms like ChatGPT are free, democratizing access to this powerful technology.

I feel this image that was used in a recent Andressen Horowitz article portrays how AI has met each of the above 6 parameters.?

Martin was the author of the article and he showed what it would cost to create a Pixar character of himself pre-AI and during AI. What you can see is that the cost is cheaper and the speed is faster.

There will always be dissenters during times of change, loud voices urging a return to the old ways—our collective societal shiver or sweat as we strive for homeostasis. But like our body's involuntary responses, these are signs that change is already upon us.

As we navigate this AI revolution, it’s essential to remember that AI is a skill, not a substitute for employment. It’s an opportunity to enhance our capabilities, and the choice is ours: evolve with the technology or risk obsolescence.

The inflection point is here, and the path we choose will define our future.