Antonym: The Innovator’s Curse Edition

Google’s golden goose and – why we all feel behind the curve of AI

This week, we’re starting with a summary of the three key insights:

1. Everyone feels behind the curve in using AI to accelerate their business. Even Google.

2. AI capability is recursive. Machines and humans can improve their performance and their ability to improve their performance with AI.

3. Organised data yields massive and often unexpected value when AI can be connected to it.

The upshot of these three things: to compete in the AI race in your sector, invest in organising your data and increasing the AI literacy of your people. Wherever this revolution is headed it is these three things that will yield the greatest advantage.

---

Dear Reader

Most organisations feel like they are late to the AI race, and most leaders who are advocating for use of AI in their organisations are bemused as to why more people aren’t using the tech more often.

It may comfort us to note that these are challenges that also trouble a founder of the company which both foresaw and sparked the generative AI explosion: Google.

It reads like a fairy-tale.

The founders of Google lived in fear of the boomerang trajectory of The Innovator’s Dilemma, a warning and a prophecy made by Harvard Business School professor Clayton Christensen around the time they were building their business.

Google’s revolutionary search engine disrupted media, advertising and much besides. It rode an enormous wave of disruption, and created a Golden Goose called search engine advertising, which still lays golden eggs to the value of 80% of its $250 billion revenue every year.

The founders lived the fabulous life of princes of a disruptor company in Christensen’s fable. But even as they ordered a fleet of airliners-turned-into-party-buses, invested in asteroid mining companies and more prosaic billionaire foibles like super yachts, they remembered the circular nature of the prophecy: the disruptor always becomes the disrupted.

Remember, Christensen told his readers, it starts with laughing at things that are low-end and not worth their trouble. Nokia scoffed and the not-even-3G iPhone 1.0 with its puny camera, BlackBerry mocked the lack of a physical keyboard. General Motors executives who had never driven anything smaller than a Cadillac rolled around the floor of their boardroom guffawing at the first small, low-end Toyota Corolla.

The innovator’s curse

The Princes of Google did everything they could to avoid the moment when their golden Google was disrupted. It would be AI, they agreed. And they decided to invent it themselves. They bought upstart competitors and brilliant promising businesses like DeepMind. They hired and lavishly rewarded the best minds in computer science. After twenty years, they had it. The Transformer, the magic key that unlocked a new age of generative artificial intelligence (Gen A) was discovered and revealed to the world in a paper bearing not one but eight of the company’s computer scientists' names as authors.

But, like a fairy-tale curse, The Innovator’s Dilemma got them anyway.

Discovering something is not enough. You have to make into a business innovation. A product that people love. Google was beaten to market by OpenAI’s release of ChatGPT in November 2022.

One of the pair of Google princes, Sergey Brin came out of retirement and declared a “Code Red” threat to their core business. The company was reorganised, its huge computing, financial and talent resource configured to find a way to catch up and retake its lead in the AI race.

A Nobel prize is not enough

Two years on, the company has made massive progress – and y’know, won a couple of Nobel prizes – and some big product breakthroughs. At a scientific and invention level they are changing the world. But in terms of business – it is not enough.

At least 25% of code shipped by Google devs is AI-generated, but Brin still thinks people aren't using AI enough to boost their daily work.

Quoted in the New York Times, Brin…

⁠[…] highlighted the need for Google’s employees to use more of its A.I. for coding, saying that the A.I. improving itself would lead to A.G.I. [Artificial General Intelligence] He also called on employees working on Gemini to be “the most efficient coders and A.I. scientists in the world by using our own A.I.”

So many leaders I speak to say they are behind in terms of AI adoption across their organisation; they should take note: even if they had the tools and resources of Google, they would still feel behind the curve.

Another insight we have that Brin’s comment connects to: AI capability is recursive, both in terms of the technology (AI can improve AI) and human performance (the more you use AI the better you use AI).

NotebookLM is a glimpse of Google’s power

Since Brin’s return, Google has restructured around artificial intelligence, launching some often glitchy and disappointing rival chatbots to ChatGPT. But the company’s power is beginning to shine through in unexpected places. As we discussed in The 150 Million Word Limit Edition, NotebookLM continues to be a stealth hit as an app and a powerhouse of an AI tool.

Many people were wowed by its Audio Preview feature, but few have treated it as a serious tool for work. I’ll be uncharacteristically hyperbolic and unreserved about it: NotebookLM is amazing. At Brilliant Noise, we are using AI to build all sorts of fabulous apps and solutions for clients, things beyond our wildest dreams just a year ago, like chatbots that can media plan, or do a week’s work of web design in an hour, using tools like Lovable, StackBlitz, Cursor and V0 to turn us into coders and software designers with access to data analysis that would have been unaffordable until now. But with NotebookLM a few PDFs, a well-put together Google Doc and some instructions can create breakthrough innovations in how we work.

It. Is. Incredible.

Readwise + AI = 🤌

Readwise however is a shockingly brilliant new tool. Well, actually Readwise is a well established tool that has been transformed with the inclusion of AI. I’ve been using it since it launched. It sucks in all of your Kindle highlights, and you can also add other notes and even use its Reader app to keep adding highlights and notes.

Up until now it has been a valuable store of my reading notes. Searchable for themes and connecting ideas of things I’ve read over almost 20 years. Here’s my Readwise stats:

A few hundred books, a few thousand articles and notes, and 28,000+ individual highlights.

So now we have to stretch our new AI-era frames of reference, from the idea that we can have a conversation with a document to we can have a conversation with a library made up of everything we’ve ever found fascinating.

It works like a dream. A really good dream.

For instance, keen readers will have heard me bang on about the three phases of the AI revolution we will go through: 1. We do the things we do already better, 2. We will find better ways to do things, 3. We will do new things we haven’t even dreamt of yet.

This week someone told me the shape of that thought and attributed it to a book called Power and Prediction by Joshua Gans, Avi Goldfarb and Ajay Agarwal. It’s a zeitgeisty framework, a logical one, which connects back to Power and Prediction, but did I read that before?

Not according to my Readwise library. No, I’d likely formed it from essays by J P Castlin, fei-Fei Li and – inevitably (because I adore his work) – Ethan Mollick.

Amazon. Well, I’ve now bought Power and Prediction and, since the three phase model is so central to my work at the moment, I will go back and read those three sources that have left their tracks across my mind.

If you’d like to read them, they are:

• Fei-Fei Li: “Fei-Fei Li Says Understanding How the World Works Is the Next Step for AI” (The Economist The World In 2025.

• JP Castlin: “The One About Strategy” (his newsletter - with thanks to Jason Ryan for his insistent recommendation).

• Ethan Mollick: “AI in Organisations: Some Tactics” (One Useful Thing, newsletter)

Watch Readwise’s video about this wonderful thing here:

One more thing: the latent value of well organised data

The value of my thousands of highlights has cubed because of this new feature. It’s a reminder of a principle for AI everywhere. The more data you have, the better organised it is, the more value you will be able to quickly create as the technology and discovery and innovation advance.

AI literacy is the best investment to put alongside organising data. It takes AI literacy to spot the obvious opportunities that will suddenly appear, at least they will be obvious to those who have learned to think with machines at their side.

That’s all for this week

Thank you for reading. If you liked please like it by hitting the 🤍to make it ❤️.

Antony

---

#Ad

The AI-B-C Power Hour

Feeling Behind the AI Curve? Jump start your learning.

In just 60 minutes together, we'll tackle your most pressing business challenge using AI tools that deliver tangible outcomes. No theoretical discussions—just practical solutions and transferable skills you can immediately apply. Plus, a bespoke report with everything you learned and recommendations to improve your personal AI skills.

One focused hour. £300 + VAT until the end of March. Apply here.