2023: Three Predictions
Prediction #1: Crypto sinks
2022 showed us that crypto was overhyped, and many projects were shady (heard about the Logan Paul CryptoZoo project?). Essays like the one from Moxie Marlinspike published a year ago started to expose that web3 was everything but decentralized. Even the most famous investors who put millions of dollars into web3 struggled to explain the underlying benefits.
2022 saw many stories unfold, the most famous being the recent FTX debacle. Many web3 projects failed to show any value. Still, many engineers tried to ride the hype wave, promising a new decentralized world free of FAANG oppression, without looking at the fundamentals: nobody gives a fuck about a decentralized internet.
The only place where people want a decentralized internet is in Silicon Valley.
And it’s a product of fiction.
I am not claiming there is no value to web3 or crypto. I am just saying that web3 technology is not as valuable or important as people claimed it to be. Compare how many millions of dollars were traded for bored apes a year ago, where it is today, and the utility it offers today to society.
Prediction #2: AI rises
Unlike crypto, AI will rise in the years to come. It took years for the technology to mature (we needed better software and hardware capabilities to train AI models), and we now see more applications daily. In the last two years, we have seen the following:
the release of GPT3
availability of GitHub Co-Pilot (GPT3 trained on GitHub repositories)
release of DALL-E for generating pictures
ChatGPT gained millions of users in days
The rise of great AI-based startups like Hugging Face
The main problem now is to integrate this technology into products. There are already a few good products (e.g., Jasper.ai, an AI-based platform to write blog posts), but many other products will change or redefine many industries.
The majority of ML-based applications is still to come.
Still, very few engineers know how machine-learning models work, and even fewer understand how to integrate machine-learning models into an application. In other words, actual knowledge of ML is very scarce, and good engineers will be hired at a premium price.
Investments in AI will accelerate in the coming years: OpenAI will release GPT-4 this year. Microsoft plans to integrate ChatGPT into Bing (which may potentially turn bing into the most popular search engine). Existing companies need to adopt AI in their products (like John Deere, which is working on an autonomous tractor) or newcomers will outcompete them.
In other words:
If you start a new tech company, make sure you have a ML component or your competitor will do it for you
Invest in ML companies that may revolutionize existing industries
Side Note: if you are a software engineer, start learning how ML works: it will differentiate you from the mass and help you get a high-paying job. This may be critical in our current environment.
Prediction #3: the startup world will be a warzone
We had a tech bull run in the last ten years. Clear winners emerged and unicorns popped like acne pimples on teenagers.
The next few years are going to be way harder for existing companies:
Companies that raised rounds in 2019-2022 and had inflated valuations are unlikely to raise again unless they (a) have solid metrics or (b) accept a down round - meaning the new valuation is lower than the previous one.
The economic situation is unlikely to improve soon, which means that
sales will go down (e.g., companies are going to save money, not spend it)
raising will be harder than before
David Sacks summarizes well how startups should approach 2023 in less than 140 characters.
Still, I believe the next few years will be a great time to start new companies, especially if you focus on emerging technologies (especially AI-based companies). Starting in 2023 means (1) being lean, (2) making a great MVP that attracts few customers, and (3) allows you to raise a pre-seed round.
In summary, 2023 will be the year when many existing, inflated startups will die, and some newcomers will thrive.
Books I am excited to read
I generally read between 10 and 20 books per year. 2022 has been a year focused on building Codiga; therefore, I read less than usual.
The 2022 books I enjoyed the most are:
Who moved my cheese?: my best book in 2022. “Who moved my cheese” is like a Miyazaki movie. It’s simple. Anyone can understand it. Still, no matter how many times you read it, you will get value from it.
Good to Great: if there is one business book I recommended in 2022, this is the one. It will explain why some companies are more successful than others and why tech companies like Tesla or SpaceX may be good but not great (hint: how do you see these companies once their superstar founder is gone?).
In 2023, I am looking forward to reading:
iGen: while I do not have kids, I want to understand the next generation of adults and how our culture and society evolve.
Thomas Piketty: I find it interesting that the so-called elite is praising socialism, a system that repeatedly showed how toxic it is for society. I want to understand why people praise such ideas considering their (potential) educational level.
Public Opinion: I read many good reviews about this book and want to dig further into the topic.
Neuromancer: reading this book has been an obsession since I finished Cyberpunk2077.
If you are interested, you can follow my progress on Goodreads.
If you are not a subscriber to this newsletter, subscribe to be notified of new posts.