AI is having a Model T moment

AI will never be this disappointing again. At the same time, it is amazing.

When we live through paradigm shifts, it’s easy to forget that the shift is incomplete and there is always more to come.

The Ford Model T was the first generation of mass-produced car. 100 years of automobile innovation later, we can see how basic they were when they rolled off the assembly line.  

Today, we are witnessing the first generation of mass market Generative AI. It is far from complete and every day we discover what good looks like and what best can be.

When there are major paradigm shifts, the power of the prototype drives accelerated adoption. People confuse this with something being complete and fully baked. That’s often a byproduct of outsized impact even while development occurs rapidly in the background.

Why is this important?

If you buy an answer to a question about how AI can solve your problems without understanding what you are buying, you will ultimately buy somebody else’s answer. The difference is they know it is not the answer and simply part of their learning journey. If you are implementing new technology, participation in the discovery of its capabilities and potential is required, otherwise you are funding somebody’s else benefit—not your own.

You will run into two types of people when attempting to purchase an AI answer:

• Those selling you answers in the hope you don’t know what you don’t know as they are paid to learn in your environment and define winning somewhere else.
• Those not selling to you since they are too busy learning for their own benefit.

Gartner predicts there will be 40% failure rate of current AI projects by 2027.

I have no idea how they settled on that number, unless they think there is a 40% failure rate in everything done. It is important to understand, though, that the failure will only be with the stakeholder paying, not with the one learning. The latter will receive 100% of the value. And at some point, they will not fail, and then they will win 100% of the time.

Is this always true?

No, it is only true when there are paradigm shifts. In my professional lifespan this has happened twice before:

1. The internet and mobile revolution starting ~1995
2. The smartphone, cloud and app revolution starting ~2007

We know AI is a real paradigm shift since it is already changing our behaviors at a massive scale.

During paradigm shifts, progress comes from two fronts: new research driving breakthroughs and engineering improving what already exists. We’re seeing this with AI, where novel model designs emerge weekly, while existing architectures are reworked for better scale, performance and energy efficiency.

This is the telling sign for true change versus ones pretending (e.g., web 3.0, metaverse, etc.). AI is already shifting behavior from searching for links to expecting answers directly. And the smartphone-based typing/screen model is being displaced with natural language and multi-modal input, where the bandwidth between human and machine is exponentially faster and more efficient.

To say this changes everything is a gross understatement. On an appearance of the Lex Fridman Podcast, Google DeepMind co-founder and CEO  Demis Hassabis, noted the change is 10x bigger than the industrial revolution and 10x faster. It is going to change our social understanding in ways that will challenge us.

What to do?

Stop buying, start learning.

It is fine to buy, but do not buy and not learn. You will be destined for the 40%, contributing only to the learning of others.

AI will change the fundamentals of your future potential as an individual, and the future of any company where you are employed and are making decisions. You cannot afford to not understand.

The question shouldn’t be: “What is your company doing about AI?” but rather,  What are you doing? What are you learning? What are you experimenting with? What don’t you understand yet and how are you closing that gap?

In a moment where the rules are still being written, that’s the only leverage that matters.

TL; DR:

1. Anyone selling you an answer is looking for funding to learn.
2. The technology is at the Ford Model T stage of maturity; understand it.
3. Invest in your own learning and that of your company.
4. Actively partner with any supplier, participate, LEARN!

In short, start being an active participant in the design of your own future. Otherwise, somebody else will do it for you. And that will not be what you want, even if you are not in the first wave of Gartner’s predicted doomed 40%.

You’ll know you are heading in the right direction when you stop asking for answers and start discovering better questions.

Catch Geoff Hollingworth on Zero-Touch Live next week as we dive deeper into this discussion. In the meantime, mention him in the comments to start a conversation.