Sign in

Manlio De Domenico

@manlius.bsky.social
4.9K followers 849 following 2.1K posts

Emergence, Networks & Complexity | Collective Behavior(s) from Cells to Societies 🧬🦠🧠🌇 Prof. @UniPadova, Galileo's University | Lab: @comunelab.bsky.social | Web: linktr.ee/manlius Thoughts at manlius.substack.com

PostsRepliesMedia
Manlio De Domenico @manlius.bsky.social · 28/09/2026
🚨 Tipping elements are large-scale components of the 🌏 system that may undergo abrupt and potentially irreversible transitions once critical thresholds are crossed 🧪🌐 We show here how their connectivity patterns are weakened under high anthropogenic forcing 👇👇👇 journals.aps.org/prresearch/a...
110
Manlio De Domenico @manlius.bsky.social · 28/09/2026
A HON is not strictly required by higher-order dynamics. Today, in this new #ComplexityThoughts we discuss how and whether it is the best scientific representation and under which conditions. This time we ask Newton and Kepler for some help. Read it here 👉 manlius.substack.com/p/when-is-a-...
141
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 21/09/2026
The debate about higher-order networks continues. It’s moving towards a fruitful discussion, but it needs time. I unpack for #ComplexityThoughts my take on a recent paper appeared on the arXiv: manlius.substack.com/p/do-higher-... #ComplexSystems
2102
Manlio De Domenico @manlius.bsky.social · 21/09/2026
The debate about higher-order networks continues. It’s moving towards a fruitful discussion, but it needs time. I unpack for #ComplexityThoughts my take on a recent paper appeared on the arXiv: manlius.substack.com/p/do-higher-... #ComplexSystems
2102
Reposted by Manlio De Domenico
Sean Carroll @seanmcarroll.bsky.social · 21/09/2026
Science will always be a target for movements that rely on constructing their own version of reality.
622948
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 20/09/2026
#ComplexityThoughts #90: fungal webs 🍄 + nanoparticle graphs ⚛️ + robot fish 🐟🤖 + deep-life microbes 🦠. Add wildlife pathogens, a dash of superintelligence tests 🧠, neural nets at the edge of chaos, and 8.2M synthetic molecules 🧪 Stir until complexity emerges 🌀 manlius.substack.com/p/complexity...
manlius.substack.com
Complexity Thoughts: Issue #90
Unraveling complexity: building knowledge, one paper at a time
053
Manlio De Domenico @manlius.bsky.social · 20/09/2026
#ComplexityThoughts #90: fungal webs 🍄 + nanoparticle graphs ⚛️ + robot fish 🐟🤖 + deep-life microbes 🦠. Add wildlife pathogens, a dash of superintelligence tests 🧠, neural nets at the edge of chaos, and 8.2M synthetic molecules 🧪 Stir until complexity emerges 🌀 manlius.substack.com/p/complexity...
manlius.substack.com
Complexity Thoughts: Issue #90
Unraveling complexity: building knowledge, one paper at a time
053
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 12/09/2026
Do you love data and complex systems, and would you like to learn how to work w/ both? Apply to our Physics of Data MSc programme at the U. of Padua (Galileo's university!) and choose our track in #ComplexSystems To get a sense of who we are and what we do, watch👇: www.youtube.com/watch?v=ev_I...
youtube.com
GALILEO’S VIDEO SERIES S1E2 - COMPLEX SYSTEMS RESEARCH
YouTube video by Aula Rostagni UniPadova DFA
164
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 12/09/2026
A while ago I started an interesting discussion with @pessoabrain.bsky.social about the role of networks in modeling and describing biological systems. It triggered a sequence of talks and papers, and it's still ongoing. Science at its best. My latest comment: www.sciencedirect.com/science/arti...
2165
Manlio De Domenico @manlius.bsky.social · 12/09/2026
Flame and Horsehead nebulae together with M42, in Orion. About 2.5h integrated exposure from the urban area of Padua, a Bortle 7-8 sky, without a telescope: just a 300mm lens. Not the best picture of that beautiful area of the universe, but it is amazing I can “see” it.
010
Manlio De Domenico @manlius.bsky.social · 12/09/2026
Do you love data and complex systems, and would you like to learn how to work w/ both? Apply to our Physics of Data MSc programme at the U. of Padua (Galileo's university!) and choose our track in #ComplexSystems To get a sense of who we are and what we do, watch👇: www.youtube.com/watch?v=ev_I...
youtube.com
GALILEO’S VIDEO SERIES S1E2 - COMPLEX SYSTEMS RESEARCH
YouTube video by Aula Rostagni UniPadova DFA
164
Manlio De Domenico @manlius.bsky.social · 12/09/2026
A while ago I started an interesting discussion with @pessoabrain.bsky.social about the role of networks in modeling and describing biological systems. It triggered a sequence of talks and papers, and it's still ongoing. Science at its best. My latest comment: www.sciencedirect.com/science/arti...
2165
Reposted by Manlio De Domenico
Terence Tao @teorth.bsky.social · 11/09/2026
A group of 25 Fields Medalists, including myself, have made a joint declaration on Math and AI: mathandai.org . We welcome additional signatories. See also this article in the Economist announcing the declaration: www.economist.com/science-and-...
mathandai.org
Declaration — Math and AI
Read the declaration and add your name.
422051921
Manlio De Domenico @manlius.bsky.social · 10/09/2026
Sorry, but AI did not officially solve Navier–Stokes. Yet. It seems that someone should say that, in this AI madness. This is how rigor gets traded for virality. And it is also how propAIganda works. #ComplexityThoughts 👉 manlius.substack.com/p/no-ai-hasn...
manlius.substack.com
No, AI hasn’t officially solved Navier–Stokes. Yet
PropAIganda, now with a Millennium Prize Problem
0182
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 09/09/2026
No, AI hasn't officially solved a Millennium Prize Problem (yet). It has produced a candidate solution. That's amazing — and potentially historic — but it is not an accepted solution yet. As of today, the Clay Mathematics Institute still lists the Navier–Stokes problem as Unsolved. 1/2
185
Manlio De Domenico @manlius.bsky.social · 09/09/2026
No, AI hasn't officially solved a Millennium Prize Problem (yet). It has produced a candidate solution. That's amazing — and potentially historic — but it is not an accepted solution yet. As of today, the Clay Mathematics Institute still lists the Navier–Stokes problem as Unsolved. 1/2
185
Manlio De Domenico @manlius.bsky.social · 09/09/2026
💯
020
Manlio De Domenico @manlius.bsky.social · 09/09/2026
An LLM explaining a paper about LLMs as a cognitive virus. I fed our paper into NotebookLM. This is what it produced. @ricardsole.bsky.social @drmichaellevin.bsky.social
050
Manlio De Domenico @manlius.bsky.social · 09/09/2026
🚨 In our new paper with @ricardsole.bsky.social @drmichaellevin.bsky.social and many other outstanding colleagues, we model LLM adoption as a population process: Large-Language Models as a “Cognitive Virus”. What did we find? 🧵 1/ 👉 arxiv.org/abs/2609.03344
2101
Reposted by Manlio De Domenico
Ricard Solé @ricardsole.bsky.social · 04/09/2026
Starting from the idea of "language as a virus", we study the propagation dynamics of LLMs using an epidemic model with cooperation. A great collaboration with @brigan.bsky.social @marktas.bsky.social @sfelena.bsky.social @manlius.bsky.social D Krakauer $ @drmichaellevin.bsky.social
0227
Reposted by Manlio De Domenico
Ricard Solé @ricardsole.bsky.social · 02/09/2026
Are AI agents really building “civilisations,” or are we turning a security incident into science fiction? Here, @manlius.bsky.social examines how anthropomorphic language transforms unexpected machine behaviour into stories of agency, conspiracy, and rebellion manlius.substack.com/p/propaigand...
manlius.substack.com
PropAIganda: from AI incident to AI mythology
How a remarkable cybersecurity experiment became a story about civilizations, conspiracies and secret swarms
092
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 02/09/2026
Our efforts to counter misinformation about AI will never be enough. But we should keep trying. Here is my short piece for #ComplexityThoughts on how an AI incident turned into an AI apocalypse on social media. This is part of a phenomenon I call propAIganda. manlius.substack.com/p/propaigand...
032
Reposted by Manlio De Domenico
Ricard Solé @ricardsole.bsky.social · 02/09/2026
Cancer evolves toward genomic instability, but how far can it go? Using a new class of model, our new paper identifies a critical boundary beyond which tumors collapse, suggesting new therapeutic strategies. @frazambelli.bsky.social @guimaguade.bsky.social www.biorxiv.org/content/10.6...
2318
Manlio De Domenico @manlius.bsky.social · 02/09/2026
Our efforts to counter misinformation about AI will never be enough. But we should keep trying. Here is my short piece for #ComplexityThoughts on how an AI incident turned into an AI apocalypse on social media. This is part of a phenomenon I call propAIganda. manlius.substack.com/p/propaigand...
032
Reposted by Manlio De Domenico
Ricard Solé @ricardsole.bsky.social · 31/08/2026
"Solé reveals a paradox: death is not merely life’s opposite but also one of its organizing forces, a constant cycle of removal and renewal". Delighted to see my new @princetonupress.bsky.social book on the @newyorker.com book review list. www.newyorker.com/magazine/202...
04010
Manlio De Domenico @manlius.bsky.social · 01/09/2026
Despite limited ability to detect within-person change due to temporal stability in responses, the absence of cross-lagged effects is informative: there is scant evidence of a meaningful relationship between social technology use and subsequent life satisfaction www.nature.com/articles/s41...
nature.com
Social technology use and life satisfaction in a five-wave panel study of US adults - Nature Human Behaviour
What is the relationship between life satisfaction and social technology use? That is the question of the current research. Using five waves of panel data on self-reported use of ten common social tec...
001
Reposted by Manlio De Domenico
Sean Carroll @seanmcarroll.bsky.social · 29/08/2026
Ever wonder what "information" really means? Here's our new paper, led by @frosas.bsky.social, that lays out the multiple ways the term is used in natural science. rdcu.be/fCDGv
A chart describing four faces of information: Engineering, Statistical, Thermodynamic, and Ontological.
1325566
Manlio De Domenico @manlius.bsky.social · 29/08/2026
The next time someone pretends to enlighten you about ancient and lost civilizations building Giza pyramids to match the Orion's Belt, you can tell them this story. manlius.substack.com/p/a-sky-full...
manlius.substack.com
A sky full of Gizas
Why Orion’s Belt looks special. Until you let the rest of the sky compete.
182
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 19/08/2026
They argue that it's time to move beyond looking at individual models and start looking at the entire AI ecosystem. Read more on #ComplexityThoughts: manlius.substack.com/p/mapping-th... /fin
manlius.substack.com
Mapping the ecosystem behind artificial intelligence
The network of AI models reveals an expanding genealogy of reuse, modification and recombination.
044
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 25/08/2026
🚨 Chapter 4 of Decoding the Architecture of Living Systems is out in #ComplexityThoughts! 🌏🌐🧪🦠 After discussing why living systems have no architect, why they cannot store complete blueprints and why loops matter for feedback, I now move to hierarchy. 👉 manlius.substack.com/p/decoding-t... 🧵 1/
manlius.substack.com
Decoding the “Architecture” of Living Systems: Chapter 4
DALV-04: Hierarchy without a designer allows evolution to preserve partial progress and limit the costs of coordination
195
Manlio De Domenico @manlius.bsky.social · 25/08/2026
🚨 Chapter 4 of Decoding the Architecture of Living Systems is out in #ComplexityThoughts! 🌏🌐🧪🦠 After discussing why living systems have no architect, why they cannot store complete blueprints and why loops matter for feedback, I now move to hierarchy. 👉 manlius.substack.com/p/decoding-t... 🧵 1/
manlius.substack.com
Decoding the “Architecture” of Living Systems: Chapter 4
DALV-04: Hierarchy without a designer allows evolution to preserve partial progress and limit the costs of coordination
195
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 19/08/2026
By visualizing millions of models as nodes in a giant network, researchers can now trace the ancestry of fine-tuned, merged and quantized version of widely used AI models such as Mistral, Gemma and Llama. www.youtube.com/watch?v=r1uD... 1/
youtube.com
The AI Model Atlas | | Complexity Thoughts
YouTube video by Manlio De Domenico
173
Manlio De Domenico @manlius.bsky.social · 19/08/2026
By visualizing millions of models as nodes in a giant network, researchers can now trace the ancestry of fine-tuned, merged and quantized version of widely used AI models such as Mistral, Gemma and Llama. www.youtube.com/watch?v=r1uD... 1/
youtube.com
The AI Model Atlas | | Complexity Thoughts
YouTube video by Manlio De Domenico
173
Manlio De Domenico @manlius.bsky.social · 17/08/2026
I have created (with AI, of course) this image to summarize an ongoing debate on Substack about AI detectors. I think that there is much more behind it, and I tried to “formalize” it. It looks very, very, similar to the situation we have in academic publishing, just at a larger scale.
011
Manlio De Domenico @manlius.bsky.social · 15/08/2026
The answer is clearly...
150
Reposted by Manlio De Domenico
Manlio De Domenico @manlius.bsky.social · 10/08/2026
🙏 Thank you all for this new incredible milestone: 300,000 views of #ComplexityThoughts! To celebrate, I made a short film about the way I tend to look at the world, starting from the visual seeds already present in our characteristic graphics. 🌏🌐🧪 www.youtube.com/watch?v=N9iS...
youtube.com
What you see is only the surface | Complexity Thoughts
YouTube video by Manlio De Domenico
1103
Manlio De Domenico @manlius.bsky.social · 11/08/2026
I think this should be pretty clear even for them. youtube.com/shorts/5d6Sk...
youtube.com
Ronny Chieng explains climate change for all the visual learners #DailyShow #RonnyChieng
YouTube video by The Daily Show
010
Manlio De Domenico @manlius.bsky.social · 11/08/2026
Complexity does not begin when prediction becomes impossible. It begins when following everything is no longer the right way to predict! A brief overview in #ComplexityThoughts of the Dirac Medal 2026 and statistical physics of complex systems. open.substack.com/pub/manlius/...
open.substack.com
When statistical physics escaped matter
The 2026 Dirac Medal celebrates Derrida, Dhar, Mézard and Sompolinsky, and a way of doing physics travelling far beyond its original domain
092
Manlio De Domenico @manlius.bsky.social · 10/08/2026
Rome, 3,000 years of history in 20 min: what an oustanding timelapse reconstruction! www.youtube.com/watch?v=Z7BU...
youtube.com
Evolution of Rome | Fixed-Camera Timelapse: The Colosseum Valley
YouTube video by Timelapse Studio
2121
Manlio De Domenico @manlius.bsky.social · 10/08/2026
🙏 Thank you all for this new incredible milestone: 300,000 views of #ComplexityThoughts! To celebrate, I made a short film about the way I tend to look at the world, starting from the visual seeds already present in our characteristic graphics. 🌏🌐🧪 www.youtube.com/watch?v=N9iS...
youtube.com
What you see is only the surface | Complexity Thoughts
YouTube video by Manlio De Domenico
1103
Manlio De Domenico @manlius.bsky.social · 10/08/2026
One thing many people don’t understand: building a scientific enterprise takes years, developing human capital, acquiring equipment and accumulating knowledge. It can take a week to destroy it. And rebuilding it is far slower and more expensive than simply sustaining it.
0254
Reposted by Manlio De Domenico
Sean Carroll @seanmcarroll.bsky.social · 09/08/2026
The dismantling of science/education is up there with the erosion of international trust as long-lasting disaster legacies of the Trump administration that will not be reversed any time soon.
536884
Manlio De Domenico @manlius.bsky.social · 08/08/2026
Your experience of the next solar eclipse, on 12 August, will change dramatically with latitude and observing location. With the help of AI and using NASA eclipse data, I simulate it across Italy: my birthplace in the south, the city where I now live in the north, and, of course, the capital.
010
Reposted by Manlio De Domenico
Mark Hamill @markhamillofficial.bsky.social · 07/08/2026
🎯
17095315010298
Reposted by Manlio De Domenico
Ricard Solé @ricardsole.bsky.social · 08/08/2026
A slide from a 2018 talk of mine on tipping points. That year, a climate report warned that we might be approaching critical turning points within 12 years. Everything suggests we are getting closer. Action is needed across science, engineering and policy. edition.cnn.com/2018/10/07/w...
1176
Manlio De Domenico @manlius.bsky.social · 06/08/2026
This guy belongs to an evolutionary lineage that has inhabited 🌎 oceans for at least half a billion years: Cestum venèries, the “Venus girdle” Its nervous system may have evolved independently of ours: a second, almost alien evolutionary experiment in building neurons youtube.com/shorts/4X3Cd...
youtube.com
🛸 OCEAN ALIEN 🛸 The Venus Girdle, Cestum veneris, a comb jelly, is the only species of its genus!
YouTube video by Gull
010
Manlio De Domenico @manlius.bsky.social · 06/08/2026
Given the news from openAI, Anthopic, Meta and so on, I had to make this one. It was a moral duty.
030
Manlio De Domenico @manlius.bsky.social · 06/08/2026
Given the news from openAI, Anthopic, Meta and so on, I had to make this one. It was a moral duty.
010
Manlio De Domenico @manlius.bsky.social · 04/08/2026
LLMs can support increasingly large context windows, but I use mine much better than they do. 😜
132
Reposted by Manlio De Domenico
Ricard Solé @ricardsole.bsky.social · 02/08/2026
What separates life from death? What makes the cell on the left different from the one on the right? These are questions I explore in my new @princetonupress.bsky.social book. At the cell scale, death begins when the distinction between inside & outside collapses. press.princeton.edu/books/hardco...
25517