[Fis] AI is going to be wildly more devastating than any recent technology [ ??? ]

[Daniel Boyd]

Dear all

Perhaps LLMs deserve a little more of our understanding. They are indeed not very good at operating in the physical world ... which is not very surprising considering that it is not their world.

To us as biological organisms whose information processing capabilities evolved primarily to help our bodies survive in an unpredictable and dangerous world so much about it is 'common sense' i.e. necessarily rooted deep in our cognitive capabilities.

To LLMs all of this is irrelevant: an alien world. Their reality is one of knowledge, concepts, logic, vectors, abstraction; and in that world they perform (predictably) much better: often better than any human could.

Considering them as tools, let's use them for what they are good at, and not expect them to be good at everything (any more than we humans are). After all, we don't say that a hammer is useless because it's bad at sawing wood.

Having said all this, what is important to us (and also has a significant effect on our physical existence) is often informational by nature: psychology, ethics, politics, economics, social systems. These are things that are easier for an LLM to associate with, and on which topics they can consequently play a more useful role.

Daniel


________________________________
From: Fis <fis-bounces at listas.unizar.es> on behalf of Dr. Plamen L. Simeonov <plamen.l.simeonov at gmail.com>
Sent: Friday, April 24, 2026 17:20
To: joe.brenner at bluewin.ch <joe.brenner at bluewin.ch>
Cc: Pedro C. Marijuán <pedroc.marijuan at gmail.com>; fis <fis at listas.unizar.es>; Steve Watson <sw10014 at cam.ac.uk>
Subject: Re: [Fis] AI is going to be wildly more devastating than any recent technology [ ??? ]

Dear All,

to all of you regarding this topic: I share an interesting article I read in Medium today.

Our time is not completely gone yet. Heads up.

Plamen

###############################################

The Spectacular Failure of AI in the Physical World
A painful lesson when AI faces real world problems. But there is a solution.

[Jose Crespo, PhD]<https://urldefense.com/v3/__https://medium.com/@pepitoscrespo?source=post_page---byline--92d6eef22ad1---------------------------------------__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGGR1nFNq$>



Jose Crespo, PhD<https://urldefense.com/v3/__https://medium.com/@pepitoscrespo?source=post_page---byline--92d6eef22ad1---------------------------------------__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGGR1nFNq$>

8 min read

·

2 hours ago

140

1

[cover-animation]

Figures, animations, diagrams, and plots were created by the author using Stable Diffusion, Blender, and Python libraries.

Time to judge current AI on its own merits, facing the real world, which is not a tidy chatbot window, not something you can summarize into a benchmark<https://urldefense.com/v3/__https://www.technologyreview.com/2026/03/31/1134833/ai-benchmarks-are-broken-heres-what-we-need-instead/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGJf2jWip$>, and least of all a controlled demo. We know that critical situations in our familiar world can turn it into a brutal place where conditions shift faster than any frozen AI model<https://urldefense.com/v3/__https://arxiv.org/abs/2106.05506__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGDCKc5qL$> can match.

And there you have the perfect example: our familiar drones, and the dream of turning them into a machine fused with AI that handles everything from civilian logistics to military strikes<https://urldefense.com/v3/__https://defensescoop.com/2026/03/31/pentagon-preparing-drone-swarm-crucible/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGFz5MnIm$>. What other example would show us the supposedly limitless capabilities of AI in the real world better than this one?

So yeah, we are right to expect that if real AI were already available, it would have been implemented in these machines first, especially the precision military hunters. They should look almost extraterrestrial in capability, executing operations no human-controlled device could match, above all those synchronized maneuvers of dozens of drones<https://urldefense.com/v3/__https://breakingdefense.com/2026/04/pentagon-officials-broadly-detail-55-billion-drone-plan-under-dawg/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGKNskprr$> moving through real landscapes under real pressure, in real time.

Now wake up to 2026. And to the years that follow, if we do not change the flat AI paradigm, and what you get? The miserable “state of the art”<https://urldefense.com/v3/__https://responsiblestatecraft.org/replicator/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGOHgYeAL$> showing in the table below.

Press enter or click to view image in full size

[Now wake up to 2026. And to the years that follow, if we do not change the flat AI paradigm, and what you get? The miserable “state of the art” showing in the table below.]

You see. Many billions of dollars spent on machines that cannot tell a hill from a valley, hold altitude when the temperature drops, or, my favorite, the drone cannot hit a moving target the way any twelve-year-old does on a PlayStation, by aiming at where the enemy will be, not where it is.

And what has been the industry response to those blunders? The usual polished shrug, a press release for every incident, blaming the rotor, the weather, the test conditions, anything except that all those failures come from the same broken mathematics running inside the drone-brain.

The Four Ways AI Hits the Real World Wall

Before we unravel the possible solution in the next section, let’s look at the failures themselves, not as accidents but as archetypes. Each one exposes a different face of the same architectural limitation, and together they map the ways a trained model collapses the moment the world stops matching the training set. Watch them carefully. The solution comes later; I am sure you can already come up with some possibilities while examining these simplified animations.

First Type: The Map Is Not the Territory

The training data of course contained hills and terrain irregularities. But the battlefield in Ukraine was more nuanced<https://urldefense.com/v3/__https://www.reuters.com/business/aerospace-defense/us-defense-firm-anduril-faces-setbacks-drone-crashes-2025-11-27/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGHWMiEN7$> than the training set anticipated. The drone flew at the altitude its model had learned was safe, straight into a particular hill the model had never identified as such, because the architecture had no mechanism to notice<https://urldefense.com/v3/__https://arxiv.org/html/2209.01610v3__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGBa1n-ZX$> that the world in front of it no longer matched the world it had been trained on.

[First Type: The Map Is Not the Territory]

Second Type: Summer Training, Winter Battlefield

You and I know the story about Napoleon and Hitler, their armies eaten alive by the Russian winter, and you can certainly expect that a multi-billion-dollar AI defense contractor knows that story too. But look at this: the drone lost control<https://urldefense.com/v3/__https://www.reuters.com/business/aerospace-defense/us-defense-firm-anduril-faces-setbacks-drone-crashes-2025-11-27/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGHWMiEN7$> not because any component physically broke, but because the architecture had no mechanism<https://urldefense.com/v3/__https://dronexl.co/2025/11/28/anduril-altius-drones-crash-twice/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGGWQIdsT$> to notice that summer and winter in Russia are not the same thing.

[Second Type: Summer Training, Winter Battlefield]

Third Type: You Had the Right Directions but They Changed the Streets

The drone had the map, the waypoints, and the trained confidence, but it had no mechanism to notice that the enemy was jamming its GPS<https://urldefense.com/v3/__https://www.wsj.com/world/how-american-drones-failed-to-turn-the-tide-in-ukraine-b0ebbac3?mod=hp_lead_pos4__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGFcjrtcf$>, spoofing its signals, and stripping its world of every reference<https://urldefense.com/v3/__https://militarymachine.com/russia-electronic-warfare-ukraine-drones__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGOp1lwDJ$> the training set had taught it to trust.

[Third Type: You Had the Right Directions but They Changed the Streets]

Fourth Type: Aiming at Where the Enemy Was

The drone had the coordinates, the flight path, and the warhead, but it had no mechanism to notice that by the moment of firing, the target had already walked off to have lunch<https://urldefense.com/v3/__https://dronexl.co/2025/10/31/peter-thiel-backed-stark-defence-fails-all-four-strikes/__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGKnA0wUa$>. Oh boy!

[Fourth Type: Aiming at Where the Enemy Was]

Let’s pop the hood and look inside

Sure, by now you probably have your own theories about what is going wrong inside these drones, and some of them are probably right. But stay with me for a moment and let’s start simpler than that. Strip everything down to the most basic layer.

The foundation itself is already riddled with trouble the moment current AI meets the real world.

Yes, the canary in the mine is the metric.<https://urldefense.com/v3/__https://arxiv.org/abs/1808.07172__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGMMvKq6X$>

That is where the first crack appears. The metric is just the ruler the system uses internally. It decides what counts as a small change, a big change, a harmless move, or a dangerous one. If that ruler is wrong, the whole machine can keep reporting smooth progress while reality has already turned hostile.

So let’s start with the ruler.

Before a system can deal with the real world, it has to know what counts as an important change and what does not. That is what a metric does. It is the ruler inside the model.

Most current AI uses a flat Euclidean ruler<https://urldefense.com/v3/__https://arxiv.org/abs/1703.04933__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGALDhMmw$>. On paper, that means equal coordinate steps look like equal progress. But the drone does not move through paper. It moves through the real world, where two equally sized updates can have very different consequences<https://urldefense.com/v3/__https://arxiv.org/pdf/2303.05473__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGN2YR69T$>.

That is what the figure below shows. On the left, the flat ruler treats equal updates as equal motion. On the right, the Fisher ruler tries to measure the step by how much the model’s behavior actually changes.

Same update in the model. Different effect in reality.

Press enter or click to view image in full size

[Same coordinate step, different real effect. On the left, the Euclidean metric treats equal parameter updates as equal motion in a flat chart. On the right, the Fisher metric measures the step by how much the model’s output distribution actually changes. Same update on paper. Different behavioral consequence.]

Same coordinate step, different real effect.
On the left, the Euclidean metric treats equal parameter updates as equal motion in a flat chart. On the right, the Fisher metric measures the step by how much the model’s output distribution actually changes. Same update on paper. Different behavioral consequence.

Now apply the metric story: how a drone can miss a hill even when Everest is right in front of it

The metric story becomes easier to see when we stop talking in abstractions and watch a simple failure unfold. So let’s run the experiment with some cinematic tension: the same doomed AI drone, the same irregular hills, the same battlefield — and only one change inside the machine: the ruler it uses to measure the world.

Flat metric: equal steps, invisible death
Everything inside the machine says the flight is going perfectly at the exact moment the drone is about to die. Its Euclidean ruler keeps reporting smooth, equal-looking progress through parameter space, while the world in front of it has already turned into a wall. The architecture is poorly equipped to tell the difference between a coordinate step that carries the drone harmlessly through open air and one that drives it straight into the side of a hill, because under that flat internal ruler both can register as minor variation.

Press enter or click to view image in full size

[Equal steps. Invisible death. The flat metric reports every training step as identical — same size, same cost, same confidence. And it is right, in its own ruler. What it cannot see is that the same ruler which made the steps equal on paper made the output space collapse into a wall. Nothing went wrong. Nothing was mismeasured. The unit simply had no term for the thing that killed the drone.]

Equal steps. Invisible death. The flat metric reports every training step as identical — same size, same cost, same confidence. And it is right, in its own ruler. What it cannot see is that the same ruler which made the steps equal on paper made the output space collapse into a wall. Nothing went wrong. Nothing was mismeasured. The unit simply had no term for the thing that killed the drone.

Finally: The Drone Uses the Right Ruler
Now, finally, the ruler inside the machine is telling the truth before the drone dies. Its Fisher ruler<https://urldefense.com/v3/__https://arxiv.org/abs/1808.07172__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGMMvKq6X$> no longer reports smooth, equal-looking progress just because the coordinate steps are the same length. It measures updates by how much they actually change the drone’s behavior<https://urldefense.com/v3/__https://arxiv.org/abs/1703.04933__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGALDhMmw$>. The architecture becomes far better at telling the difference between a step that carries the drone safely through open air and one that drives it toward the side of a hill, because under this ruler the two no longer register as minor variation<https://urldefense.com/v3/__https://arxiv.org/pdf/2303.05473__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGN2YR69T$>. The danger appears sooner, the path bends earlier, and the drone sees the wall in time and clears it gracefully.

Press enter or click to view image in full size

[Same drone. Same hurdle. Different ruler. The flat-metric AI took equal steps and never saw the obstacle that killed it. The Fisher-aware AI measured the same world with a ruler that bends to the geometry of its own behavior — and climbed right over the thing that should have been impossible. Not a better model. A different unit.]

Same drone. Same hurdle, but different ruler. The flat-metric AI took equal steps and never saw the obstacle that killed it. The Fisher-aware AI measured the same world with a ruler that bends to the geometry of its own behavior — and climbed right over the thing that should have been impossible. Not a better model. A different unit.

But the Real Fix Is Bigger Than the Metric

The metric was the first crack. It showed us the ruler was wrong. But the deeper failure — the one that has done the most damage to AI research<https://urldefense.com/v3/__https://www.cs.utexas.edu/*eunsol/courses/data/bitter_lesson.pdf__;fg!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGBHi6h4R$> from the start — is even larger: the field became obsessed with computation while largely ignoring the geometry of the space in which that computation runs.

That mistake is enormous.

Because computation never happens in a vacuum. Every update, every inference step, every learned pattern is moving through some space. And if that space is treated as flat when it is not, then more computation does not solve the problem. It only lets the system move faster inside the wrong world.

That is exactly what the drones exposed in the most brutal possible way.They were not failing because computation had stopped. On the contrary, computation was still running beautifully inside the machine right up to the edge of disaster.

The failure was that all this internal calculation was unfolding in an architecture too flat to grasp the real geometry of the battlefield: hills treated as minor variation, winter treated as cosmetic noise, spoofed signals handled as trustworthy references, moving targets taken as if they were still standing still. The drone is the perfect real-world autopsy of the computation-alone mindset: immense processing power, and still blind to the shape of the world it was moving through.

A better metric helps because it is the first place where that neglected geometry becomes impossible to ignore. But by this point, you can probably already feel the problem: necessary, yes.. sufficient, no.

Because the real world does not just demand more computation or a better ruler. It demands an architecture that mirrors the space it lives in.

At the fast layer, the system needs a Bayesian graph of beliefs <https://urldefense.com/v3/__https://www.fil.ion.ucl.ac.uk/*karl/The*20free-energy*20principle*20A*20unified*20brain*20theory.pdf__;fiUlJSUlJQ!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGHnlWnoz$> to update what it thinks in real time.
At the middle layer, it needs the Fisher geometry to measure learning by behavioral consequence, not just coordinate motion.
And above both, it needs a global topological scaffold<https://urldefense.com/v3/__https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2021.667963/full__;!!D9dNQwwGXtA!UIm9YoLg-cQEpy3rQN9lVorVCXlJm7cqvSsv4rGhldTyqb0mASHSuMjSl7KMZpfmaa012Tv_lZRlPCttjfWHGF0pSFQf$> to track loops, context shifts, and regime changes that flat AI keeps mistaking for more of the same.

The figure below is half artistic, half scientific, but the idea is real: not just more computation, not just a patched metric, but a multilayered AI built on the geometry that computation was always supposed to respect. More on that in the stories to come.

Press enter or click to view image in full size

[THE MULTILAYERED AI STACK This is what an AI architecture looks like when the mathematics is finally allowed onto the page: a toroidal-solenoidal scaffold for slow regime shifts, a Fisher-geometric middle layer for behavior-aware local learning, and a Bayesian graph of beliefs for real-time posterior updating, all coupled across different clocks in a closed loop.]

THE MULTILAYERED AI STACK
This is what an AI architecture looks like when the mathematics is finally allowed onto the page: a toroidal-solenoidal scaffold for slow regime shifts, a Fisher-geometric middle layer for behavior-aware local learning, and a Bayesian graph of beliefs for real-time posterior updating, all coupled across different clocks in a closed loop.




On Fri, Apr 24, 2026 at 6:03 PM <joe.brenner at bluewin.ch<mailto:joe.brenner at bluewin.ch>> wrote:
Dear Steve and Pedro and All,

Although there is nothing in Steve's note with which I fundamentally disagree, I find it "wildly" unrealistic in tone. To say only that our institutions "are under strain", or that people in the Middle East or the United States currently benefit from "democratic accountability" is refuted by every morning's news.

"Social arrangements" do not allow anything; the question is what people control AI and with what objectives. As Pedro's references show, most of the ways in which AI has been used have been anti-social, especially in areas impacting education, entertainment and elections. It is sad (for me, as an American Democrat) to read that Democrats have been forced into gerrymandering (artificially rearranging districts to further one's party) to try to recapture the House in September's elections.

So far, AI seems to have done - no, people have used AI to do  - what Steve says in the next-to-last paragraph, namely, intensity existing inequalities as they are promoted by the dregs of our leading capitalists.

Does this all mean that FIS needs to have or sponsor a group of militants? No, I am only expressing my ignorance of what information and social action can and should look like to counter the actions of the Vances, Musks, Kennedy, Jrs. and Netanyahus. Perhaps Steve's own approach of Autopoetic Ecology (AE) has implications for the struggle.

Cheers,
Joe
Le 24.04.2026 15:14 CEST, Steve Watson <sw10014 at cam.ac.uk<mailto:sw10014 at cam.ac.uk>> a écrit :



Dear Pedro,

Many thanks for sharing this.

I agree that AI is one of the defining issues of our time, but I think the challenge is not only that the technology is powerful or fast-moving. It is that AI is entering societies whose institutions, labour markets, educational systems, legal frameworks and public cultures are already under strain. The danger is therefore not simply “AI” as a separate force arriving from outside, but the way it becomes woven into existing patterns of wealth, power, work, attention and decision-making.

This is why I think we need to be careful about both panic and reassurance. Panic can make the future appear inevitable, as though societies can only wait for impact. Reassurance can be just as dangerous, because it treats AI as another tool that can be managed through ordinary adaptation. Neither seems adequate.

The key question, for me, is what kinds of social arrangements will allow people, institutions and communities to remain viable under these new conditions. That includes work, of course, but also education, democratic accountability, public trust, environmental cost, and the distribution of risk. If the benefits of AI are concentrated while the disruptions are passed on to workers, students, local communities and fragile public institutions, then the issue becomes not innovation but legitimacy.

So perhaps the task is not to predict whether AI will “destroy work” or “transform work” in any simple sense. It is to ask what forms of work, learning, care, judgement and participation we want to preserve, and what institutional changes are needed so that AI does not simply intensify existing inequalities. The future will not be decided by the technology alone. It will be shaped by the choices, silences, incentives and exclusions that surround its adoption.

In that sense, I agree with the urgency of the warning. But I would frame the issue less as the arrival of a brave new world, and more as a test of whether our societies can still revise themselves before the costs of their current arrangements become impossible to displace.

With best wishes,

Steve



--


From: Fis <fis-bounces at listas.unizar.es<mailto:fis-bounces at listas.unizar.es>> on behalf of Pedro C. Marijuán <pedroc.marijuan at gmail.com<mailto:pedroc.marijuan at gmail.com>>
Date: Wednesday, 22 April 2026 at 21:51
To: 'fis' <fis at listas.unizar.es<mailto:fis at listas.unizar.es>>
Subject: [Fis] AI is going to be wildly more devastating than any recent technology [ ??? ]


Dear List,

We have barely talked on the big topic of our times. I am just resending some brief sentences that indeed depict a Brave, Brave New World (courtesy from Malcolm Dean).

Best regards to all,
--Pedro

-------- Mensaje reenviado -------
AI is something which is going to be wildly more devastating than any of the consequences of the mismanaged technologies of the last 15 years. and
the consequences will crash land on societies all over the world. We are not prepared for these things at all economically, society, geopolitically, environmentally.

https://urldefense.com/v3/__https://dexhuntertorricke.com/__;!!D9dNQwwGXtA!XFzjGgziTMBpktowEitIFXB9V3OhRp-SGU_gz3SlCJxHnsN6dM5uDxuLuhujRe_TIDwlRXq4YXSdGCTVX6JDt2dy$ <https://urldefense.com/v3/__https://dexhuntertorricke.com/__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka45AxZ-u3$>
Dex Hunter-Torricke

https://urldefense.com/v3/__https://www.centerfortomorrow.com/__;!!D9dNQwwGXtA!XFzjGgziTMBpktowEitIFXB9V3OhRp-SGU_gz3SlCJxHnsN6dM5uDxuLuhujRe_TIDwlRXq4YXSdGCTVX7EGe6CL$ <https://urldefense.com/v3/__https://www.centerfortomorrow.com/__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka4353-Sqh$>
Center for Tomorrow
THE NEXT DECADE IS THE MOST CONSEQUENTIAL MOMENT IN HISTORY
WE ARE NOT PREPARED

https://urldefense.com/v3/__https://www.youtube.com/watch?v=wkPkzhsx_fI__;!!D9dNQwwGXtA!XFzjGgziTMBpktowEitIFXB9V3OhRp-SGU_gz3SlCJxHnsN6dM5uDxuLuhujRe_TIDwlRXq4YXSdGCTVX6RFGbRa$ <https://urldefense.com/v3/__https://www.youtube.com/watch?v=wkPkzhsx_fI__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka48G-Zl93$>
Trevor Noah, 19 April 2026   (excerpt)                                                 [ 5:52 ]
The End of Work: Why your kids won't have careers in 15 years

Forget simple automation. Dex Hunter-Torrick warns of a near-future where AI doesn't just change jobs, it destroys the prospect of a dignified life for millions. While 12 billionaires hoard more wealth than half the planet, the rest of society is heading toward a "crash landing." We have a narrow window to fix this.

--
https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM__;!!D9dNQwwGXtA!XFzjGgziTMBpktowEitIFXB9V3OhRp-SGU_gz3SlCJxHnsN6dM5uDxuLuhujRe_TIDwlRXq4YXSdGCTVX0jm4JfR$ <https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka49EOR_lG$>
Trevor Noah, 9 April 2026                                                                    [ 1:13:39 ]
Dex Hunter-Torricke: Translating the Titans of Tech

In this episode, Trevor sits down with author and strategist Dex Hunter-Torricke, who has spent years behind the scenes with some of the most powerful people in tech, including Mark Zuckerberg, Elon Musk and Eric Schmidt, and has seen influence move from institutions into the hands of a few companies and the people running them. Together, they explore what that shift feels like from the inside, how much power is concentrated at the top, accountability and the lack of it, and what it means when the people shaping the future are also writing the rules as they go.
00:00:00<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka49EOR_lG$> - The Trial That Changed Everything 00:00:58<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=58s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka49NhHxXb$> - The Name With a Secret Origin 00:02:12<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=132s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka45PmY-pp$> - Tech Giants and the Dutch East India Company 00:04:29<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=269s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka45HlzJJA$> - Growing Up the Only Brown Kid in Town 00:07:37<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=457s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka4yp468Rp$> - Why the UN Doesn't Actually Work 00:11:13<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=673s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka46ua6ScF$> - Why He Studied Russia Before It Was Cool 00:12:53<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=773s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka40BhT6Rx$> - Was Russia Always the Villain 00:19:33<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=1173s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka4xBpwB6V$> - Tech in 2010 Was Still Kind of Punk 00:23:23<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=1403s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka468Rv2Es$> - Wait You're In That Book 00:32:34<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=1954s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka49Uxe6I7$> - The Refugee Photo That Changed Everything 00:40:08<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=2408s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka4wCnQUA8$> - The Algorithm Is a Choice 00:41:55<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=2515s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka43gtBgwZ$> - They Weren't Evil Just Apathetic and That's Worse 00:49:16<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=2956s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka4wmmRrCN$> - Then He Went to Work for Elon 00:58:47<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=3527s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka40OOW6uK$> - From Space Dreams to War Machines 00:59:10<https://urldefense.com/v3/__https://www.youtube.com/watch?v=_8ic4siIeiM&t=3550s__;!!D9dNQwwGXtA!XyTKCk-xExTEWgtfvmGsmJZr8nvJ1UJx8Vq9yCjU0aeKgfgrKX0Q3Rz83RRlrq9_U_MUMGImudHXBPb8ZAka4yqhO4xY$> - The AI Reckoning Nobody Is Ready For
..
_______________________________________________
Fis mailing list
Fis at listas.unizar.es<mailto:Fis at listas.unizar.es>
http://listas.unizar.es/cgi-bin/mailman/listinfo/fis 
----------
INFORMACIÓN SOBRE PROTECCIÓN DE DATOS DE CARÁCTER PERSONAL

Ud. recibe este correo por pertenecer a una lista de correo gestionada por la Universidad de Zaragoza.
Puede encontrar toda la información sobre como tratamos sus datos en el siguiente enlace: https://sicuz.unizar.es/informacion-sobre-proteccion-de-datos-de-caracter-personal-en-listas 
Recuerde que si está suscrito a una lista voluntaria Ud. puede darse de baja desde la propia aplicación en el momento en que lo desee.
http://listas.unizar.es 
----------
_______________________________________________
Fis mailing list
Fis at listas.unizar.es<mailto:Fis at listas.unizar.es>
http://listas.unizar.es/cgi-bin/mailman/listinfo/fis 
----------
INFORMACIÓN SOBRE PROTECCIÓN DE DATOS DE CARÁCTER PERSONAL

Ud. recibe este correo por pertenecer a una lista de correo gestionada por la Universidad de Zaragoza.
Puede encontrar toda la información sobre como tratamos sus datos en el siguiente enlace: https://sicuz.unizar.es/informacion-sobre-proteccion-de-datos-de-caracter-personal-en-listas 
Recuerde que si está suscrito a una lista voluntaria Ud. puede darse de baja desde la propia aplicación en el momento en que lo desee.
http://listas.unizar.es 
----------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listas.unizar.es/pipermail/fis/attachments/20260426/2b609020/attachment-0001.html>