Director of Centre for Digital Habits (https://xitter.cf/t.co/M1ZVfL75Y3), Research Fellow at University of Oxford, and serves on Denmark's Media Council for Children & Youth

Oxford, England
Joined March 2014
Thrilled to share the official version of my DPhil thesis "Ulysses in Cyberspace: Examining the Effectiveness of Design Patterns for Digital Self-Control"! 😁🎉 I explore how digital devices’ design can help us use them in line with longer-term goals. 🧵 ulyngs.github.io/phd-thesis/
1
5
35
Ulrik Lyngs retweeted
The best way to pace the frontier is to hold the labs fully liable for the behavior of their models.
930
1,944
434
21,405
1,305,024
*Sigh*. I SOOOO want to digital minimalism companies like @punkt to succeed. But they're hard to take seriously when they seem to not understand the importance of the services they provide. Example: the @punkt MP02 dumbphone.
2
2
161
How we spend our time and attention is how we spend our lives. So I get so sad when glossily-designed 'solution' companies with great PR teams don't take their responsibility seriously by dedicating the technical expertise to their task that it requires.
1
13
Claude wants to RUN YOUR LIFE now!! All dystopian vibes aside (do we want closed-source AI companies to know and run the daily lives of hundreds of millions of people...?), I think a key implication is that it CHANGES THE INCENTIVES of social platforms like Meta's. The business model for platforms like Facebook has been: 1) attract consumers and business to a useful platform, 2) make it impossible to leave the platform, 3) change the design to maximise revenue with ads and data extraction. So far, regulators have been unable to change the market incentives to break that cycle. But if hundreds hundreds of millions of people experience how insanely useful it is that their AI can read their own data on those platforms, IT CHANGES THE MARKET INCENTIVES: it creates intense pressure on Meta to make their platforms open, so AIs can tell if there's something important in a WhatsApp or Instagram message. Convenience is a powerful driver of mass adoption of tech. And the allure of the convenience that AI can triage your personal information may TRUMP NETWORK EFFECTS and make people move personal communication to open platforms...
100
Ulrik Lyngs retweeted
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter. AI will transform every industry, power every company, and be built by every country. Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty. The world needs both frontier closed models and frontier open models. images.nvidia.com/pdf/Open-W…
15,986
29,290
10,949
171,557
66,487,276
Ulrik Lyngs retweeted
Once we have completed our review for security vulnerabilities, we will make the entire codebase of 𝕏 open source, with no exceptions. Moreover, we will invite third party reviewers to examine the system that is running to confirm that the open source code is what is running. Trust through total transparency is the only thing that should be believed.
5,781
11,789
2,346
108,486
7,354,104
Ulrik Lyngs retweeted
An important piece from Demis. We need more of this kind of thinking. A good reminder that the goal is a frontier ecosystem that promotes innovation and choice, while avoiding any one model drop that breaks the world!
413
964
66
7,324
1,839,229
Do you work at a US university or school with digital distraction problems? Join us! 🇺🇲 The debate about smartphones and social media is stuck in a state of worry: "apps are addictive" and now "AI will make things even worse". Since 2019 we've iterated on a practical solution: helping people redesign their digital environment to get back control of their time and attention. It's based on our 10+ years of peer-reviewed research at Oxford, and tested at 30+ institutions across the UK, Denmark, Germany and Thailand. Now, we're bringing it to the US for the first time. In September, at Santa Clara University in the heart of Silicon Valley (thanks to @klukoff !) we're running our first-ever trainer course for US universities and schools. Interested? Join us on July 7th for an online demonstration of our Reduce Digital Distraction workshop ahead of the course. It'll teach you 10 years of insights from our digital distraction research in 90 minutes. Sign-up: reddfocus.org/scu
1
36
Ulrik Lyngs retweeted
How will your country's demographics change this century? At Our World in Data we built a tool to let you see for yourself. The UN's assumptions are the starting point. But then you can adjust the three drivers of change — births, deaths, and migration — to what you expect.
12
76
8
294
58,309
Ulrik Lyngs retweeted
If thought were independent of language, we wouldn’t have any use for language, because the whole point of language is to convey ideas. So it’s not independent, but it’s not the same thing as language. We know it’s not the same thing for many reasons. One of them is: how would children learn language to begin with if they were incapable of thinking until they had language? The way that language acquisition has to work is that children have some understanding of the world, of the objects around them, of the intentions of the people that they interact with, and they map them onto the sequences of sounds and gestures that they perceive. Also, we know that there are many forms of thought that are not linguistic, such as visual imagery, spatial cognition, and mental maps, all of which are specialties of Professor Tversky, who I’m sure will enlighten us about those. Even when it comes to the content of language, what we take away after the words have faded is the gist of language, not the words themselves. Within a few seconds after I stop speaking, or a few minutes after you’ve read something, you probably could not reproduce a single sentence verbatim, but hopefully you would come away with the content, the meaning, the gist. And one of the most robust findings in our field, cognitive psychology, is that long-term memory for gist is far more robust than memory for text itself. We know from studies of the brain that there are networks of language in the brain that are not the same as the parts of the brain that light up when people are thinking visually, thinking spatially, or thinking in terms of motion. Again, to go back to one of Professor Tversky’s specializations, we can imagine tying a shoe; we can imagine diving off a board. We have auditory imagery—a song can run through our head—and there are probably other forms of thought that we don’t even have names for. We also know that we sometimes struggle to put our thoughts into words. We write down a sentence, and we realize, no, that’s not what I meant, and rewrite it until the sentence comes out right. Full @IAI_TV video: iai.tv/video/lost-in-thought
172
27
895
110,732
Fortunately if we teach kids how the web works, we can give them simple & effective ways to protect themselves from nefarious designs -- e.g. with a few clicks you can get rid of anything on the web thanks to browser extensions (easy to build github.com/ulyngs/reddfocus-…) Let's teach kids how to hack the digital world to work for them instead of telling them they're doomed
In The Anxious Generation, I underestimated the harm from the phone-based childhood because I focused on the mental health outcomes, which is where we had the best data while I was writing the book. I now believe that the widespread diminishment of the human capacity to pay attention is an even larger harm, affecting the majority of children, and even many adults. Diminished focus, executive function, and book-reading means diminished life chances.
1
44