Our Age of Machine Intelligence
Hi there, and welcome! It’s good to have you here. On this page, I’d like to tell you a little bit about myself, so that you’ll know what to expect on these pages.
About me
In my day job, I’m a professor at EPFL, one of the two Swiss Federal Institutes of Technology (think MIT, but surrounded by Swiss Alps). There, I run the Digital Epidemiology Lab. If you wonder what that means, have a look at a book I’ve written, freely accessible online: www.digitalepibook.com. By the way, if you’re interested in digital epidemiology and digital health, I write another Substack dedicated to that topic.
I’ve been using machine learning in my work since 2009 when I started recognizing social media as a promising data source of epidemiology. Over time, I realized that the speed of avdvancements in machine learning was so high that we’d eventually all use this technology in our work. In 2016, together with some colleagues at EPFL, I co-founded AMLD (Applied Machine Learning Days), which is now the largest ML conference in the region, with thousands of participants each year. I now also co-direct the EPFL AI Center.
In addition, together with one of my PhD students, I co-founded AIcrowd, an AI challenge platform used by companies like Google, OpenAI, Amazon, and others, as well as by many universities.
Why I created this newsletter
So, why this topic, and why this platform?
I’ve been active on social media, especially Twitter, for a long time. While I’m now mostly using LinkedIn and Mastodon, I very much love the format of Substack and its economic model behind it, which is not driven by ads, but by subscriptions. I much prefer that model.
On social media platforms and elsewhere, I was quite active in conversations about machine learning and what it means for our world. When ChatGPT came online at the end of November, I was both thrilled and shocked. I realized instantly that the world had changed overnight. Ever since then, I have not spent a single day without feeling this way. The speed of developments is mind-boggling. The end of 2022 marked the end of the beginning: The machine intelligence age has finally begun.
But what does that mean? And how can you and I benefit from it? As I started embracing this new technology, I felt that the best way to share what I found, in near real-time, was a newsletter. So this is what you will find here - my findings, both practical and philosophical, about our current age of machine intelligence.
Why I have two subscription types
I’ll get right to the point: I encourage readers to subscribe to the paid subscription.
Why do that? Why not just go all free? I’m relatively old-fashioned in my belief that a serious relationship between a reader and a writer involves an exchange of value, which goes both ways. I pay for my books and magazines, and I gladly do so when I get value out of it. The fact that the internet has made many of us believe that this can be a one-sided value transaction has created the ad-based model, which is at the core of what is broken about the internet.
Starting in 2024, I will donate all financial support to the EPFL AI Center.
What I will write about
As you can see from the archive, this newsletter is about our interactions, both individually and socially, with the latest AI models. My goal is to include as many practical aspects as possible. Some posts will be “bigger picture”
reflections on aspects of AI, and I hope you find them insightful, too.
My general proposition is that I will share with you what I’m learning as someone who is engaged with this technology on a daily basis. In my role as a researcher at the intersection of science and engineering, I may see things relatively early, and you may find those observations interesting. But mostly, like many people, I am trying to understand how this can help me become better at what I do. Thus, most of my posts will contain very concrete advice and examples.
Once again, welcome to this newsletter, and I hope you will subscribe, enjoy, and engage!
CODA
The icon was created with DALL-E 3.
