Predictions Archives - Futurist Speaker

The Asimov Manifesto

Thomas Frey — Tue, 26 May 2026 04:27:02 +0000

There is a photograph circulating the internet that should stop every one of us cold. It shows a robotic dog — sleek, mechanical, four-legged — with a rifle mounted on its back, trotting across a dusty field on autonomous legs, scanning for targets. No human hand on the trigger. No human eye behind the scope. Just a machine, doing what machines are being trained to do: find, aim, and kill.

Isaac Asimov saw this coming. He just hoped we were smarter than this.

In 1942, the visionary science fiction author embedded three simple laws into the fictional brain of every robot he ever wrote. They were elegant. They were obvious. And eighty years later, the engineers arming our machines have apparently never read them.

A robot may not injure a human being. Four words. Eighty years ignored.

This is our moment to change that. This is the Asimov Manifesto.

We Are Already Living in the World He Warned Us About

Let’s be precise about what is happening right now, because vague alarm is not enough. Quadruped robots originally designed for construction sites and disaster response have been fitted with weapons attachments by defense contractors. Unmanned ground combat vehicles armed with autocannons have been fielded in active conflict zones. The United States, China, South Korea, Turkey, and Israel are all racing to deploy lethal autonomous weapons systems — machines that can select and engage targets without meaningful human control.

Drone swarms equipped with explosive payloads have been documented in active combat across three continents. The threshold between “remote-controlled weapon” and “autonomous killing machine” is narrowing by the month. When a drone can identify a human face, calculate a flight path, and detonate — all without a human decision in the loop — we have crossed a line from which there is no easy return.

We are not building a safer world. We are building a more efficient killing machine and calling it progress.

We are not honoring Asimov’s First Law. We are dismantling it, contract by contract, prototype by prototype.

Efficiency Is Not a Virtue When the Goal Is Destruction

The military argument for autonomous weapons follows a seductive logic: fewer soldiers at risk, faster response times, emotionless decision-making, precision targeting. It sounds almost humanitarian — until you follow the logic all the way down.

A machine that kills more efficiently is not morally superior to a human who kills reluctantly.

The goal of warfare should be its cessation, not its optimization. When we build better killing machines, we are not building a safer world — we are building a world in which killing becomes cheaper, faster, and easier to authorize. Wars that cost too many human lives on both sides eventually end. Wars fought by machines, at scale, at minimal cost to the powerful, may never end at all.

Think about what happens when robot soldiers cost less than diplomacy. Think about what happens when a government can wage war without a single flag-draped coffin arriving home. Think about the wars that will be started precisely because the human cost — the moral weight of sending someone’s child into harm’s way — has been engineered out of the equation.

Remove the human cost of war and you remove the conscience that stops it.

This is the catastrophe hiding behind the word “innovation.”

The moment children fear the sky more than the dark, civilization has already crossed a line it may never fully return from.

The Child Who Grows Up Afraid of the Sky

There is a generation of children in conflict zones around the world who have grown up knowing the sound of a drone before they knew the sound of birdsong. They look up and do not see possibility. They see threat.

Now imagine that fear going global. Imagine it landing in your neighborhood.

Imagine a future where no crowd can gather without wondering whether an autonomous system overhead has flagged the assembly as a target. Imagine a future where authoritarian governments deploy robot enforcers in public squares, programmed to identify and subdue anyone the algorithm classifies as a dissenter. This is not science fiction. It is a procurement decision away from reality.

A society that lives in fear of its own machines has already lost something it cannot get back.

The greatest civilizational achievement we could hand to the next generation is a world in which no human being — anywhere, in any country, regardless of how they are classified by a government or a data set — has to live in fear of being harmed by a machine. That is a world worth building. That is a world Asimov imagined we were capable of choosing.

Morality Must Be Built In, Not Bolted On

Here is the insight that changes everything: we teach children morality before we teach them algebra. When they can behave well in a social situation, then we teach them language and complex reasoning. The sequence matters. Even the most sophisticated working animal is taught restraint before it is taught to act.

We have inverted this with robots. We have engineered speed, precision, payload, and target acquisition — and treated ethics as an afterthought. A feature to be added in a future software update. A press release consideration rather than a foundational design constraint.

You cannot retrofit a conscience. You have to build it in from the beginning.

If we are serious about coexisting with machines, morality cannot be optional. It must be the first requirement, not the last. Before a robot is taught to walk, it must be taught not to harm. Before it is taught to aim, it must understand that some things must never be aimed at. These are not restrictions on innovation. They are the preconditions for a future worth innovating toward.

I never met Isaac Asimov, but few minds have shaped my thinking about the future more profoundly than his.

The Five Principles We Must Enshrine

This is not a call for pacifism. This is not a call to disarm humanity. This is a call to draw one clear, permanent, non-negotiable line between the world we want and the world we are stumbling into.

Technology without ethics is not progress. It is a faster path to catastrophe.

One. No robotic or autonomous system shall be designed, manufactured, sold, or deployed with the primary or secondary function of injuring or killing a human being.
Two. Any robotic system capable of independent mobility in public or contested space must be incapable of lethal action without a verified, accountable, real-time human decision.
Three. The weaponization of commercial robotics platforms — robotic dogs, delivery drones, inspection systems — shall be treated as an international arms violation equivalent to the weaponization of civilian aircraft.
Four. Nations that develop, export, or deploy lethal autonomous weapons systems without meaningful human oversight shall face the same international censure as nations that deploy chemical or biological weapons.
Five. Asimov’s First Law shall be codified into binding international treaty as the foundational principle of the age of robotics: A robot may not injure a human being.

Five principles. One civilizational commitment. Eighty years overdue.

We are not just building robots. We are building the moral architecture of the future — and history will remember the choices we make now.

What We Build Next Defines Who We Are

Every technology is a choice. The printing press could spread knowledge or propaganda — and it did both. The internet could connect humanity or surveil it — and it does both. Robotics and artificial intelligence are the most powerful tools our species has ever held, and like every tool before them, they will reflect the intentions of the hands that shape them.

We do not get to build the future and then complain about who moved in.

We are at the hinge point. The decisions being made right now — in defense ministry budget meetings, on factory floors across three continents, in the corridors of the United Nations — will determine whether robotics becomes the greatest force for human liberation in history, or the most efficient instrument of human oppression ever built.

Isaac Asimov did not write his Three Laws because he was afraid of robots. He wrote them because he was afraid of us — afraid that we would build minds without wisdom, power without restraint, and capability without conscience.

He was right to be afraid. And we still have time to prove him wrong.

Sign the manifesto. Teach it. Demand it. Legislate it.

The robots are already here. The only question left is whether they serve humanity — or hunt it.

“The Three Laws of Robotics protect humans from robots, protect robots from humans, and force robots and humans to cooperate.” — Isaac Asimov. It is time we made them law.

IEEE Spectrum “Ban or No Ban, Hard Questions Remain on Autonomous Weapons” https://spectrum.ieee.org/ban-or-no-ban-hard-questions-remain-on-autonomous-weapons

IEEE Robotics and Automation Society “Robot Ethics: The Ethical Implications and Consequences of Robotic Technology” https://www.ieee-ras.org/robot-ethics/

Future of Life Institute “Autonomous Weapons Open Letter: AI and Robotics Researchers Call for a Ban” https://futureoflife.org/open-letter/open-letter-autonomous-weapons-ai-robotics/

The post The Asimov Manifesto appeared first on Futurist Speaker.

The Neutral Sky: Why Space May Be the Only Fair Ground for AI in the Developing World

Thomas Frey — Fri, 15 May 2026 00:43:35 +0000

By Futurist Thomas Frey and Futurist Teresa Grobecker

The geopolitics of AI infrastructure has left smaller nations with no seat at the table. A constellation of orbital edge computers may be the first genuinely neutral ground they have ever had.

Every conversation about AI sovereignty eventually runs into the same wall. The compute is owned by someone. The data center is in someone’s jurisdiction. The undersea cable lands on someone’s shore. The chip was fabbed in a facility dependent on someone’s export license. For the large nations — the United States, China, the European Union, and a handful of others — this dependency chain is manageable because they sit near enough to the top of it. For the 130-odd countries that do not, the emerging AI economy looks less like an opportunity and more like a new version of a very old arrangement: powerful nations own the infrastructure, smaller ones consume the output and generate the raw material, and the terms of that exchange are set by whoever holds the hardware.

There is, however, one domain that no single nation owns, where no single company holds the cable, and where the Outer Space Treaty of 1967 still theoretically guarantees freedom of access to all: the sky above the atmosphere. And a small but rapidly maturing cluster of companies, researchers, and space agencies are beginning to ask whether orbital infrastructure — specifically, AI compute deployed at the edge in low Earth orbit — might offer developing nations something the terrestrial internet never did: a place to process their own data on genuinely neutral ground.

What Orbital Edge Compute Actually Is

Let’s be precise about what we are and are not talking about, because the gap between the vision and the current reality matters enormously for honest assessment.

Edge compute in orbit means processing data aboard a satellite rather than transmitting raw data to a ground station and then on to a terrestrial data center for analysis. The satellite carries a processor — currently something in the range of a high-end embedded system or a compact GPU module, drawing between 10 and 100 watts of power — and runs inference models, image analysis, or sensor fusion directly on the hardware in space. The processed result, rather than the raw data stream, comes down to Earth. This is the model being pursued by companies including Loft Orbital, Unibap, D-Orbit, and a growing number of national space agencies equipping small satellites with AI accelerator chips.

The honest limitation is equally important to state. A satellite running 50 watts of AI compute is an edge node, not a training cluster. It can run a pre-trained model. It can perform inference — classifying an image, detecting an anomaly, flagging a pattern. It cannot train a large language model, cannot process petabytes of data, and cannot replace the industrial-scale compute infrastructure that foundation model development requires. Anyone claiming that orbital compute solves the AI sovereignty problem for developing nations wholesale is overstating a genuine but bounded capability.

What it can do, done well, is something narrower and potentially more immediately valuable: process locally generated data, locally, without routing it through infrastructure owned and monitored by foreign powers. That is not everything. But for a great many use cases relevant to developing nations, it may be exactly enough.

Orbital AI could flip the surveillance model—letting nations process their own territorial data in space instead of exporting it to foreign-controlled systems.

The Eye in the Sky, Reconceived

The phrase “eye in the sky” has historically carried surveillance connotations — the powerful watching the powerless from above. The orbital AI model being imagined here inverts that relationship in a way worth taking seriously.

Consider what a constellation of AI-equipped small satellites, operated under a neutral or collectively owned framework, could do for the nations currently most underserved by terrestrial AI infrastructure. Agricultural monitoring at a resolution and frequency no ground-based sensor network in a low-income country could afford — crop health, soil moisture, flood inundation, pest migration — processed aboard the satellite and delivered as actionable intelligence directly to a farmer’s phone. Deforestation detection in real time, not months after the fact when the logging trucks have already gone. Supply chain monitoring for commodity exports — cocoa, coffee, minerals — that lets producing nations verify independently what is being extracted from their territory and when. Disaster response coordination that does not depend on a functioning terrestrial internet that may itself be the casualty of the disaster.

Each of these applications shares a structural property: the raw data — the satellite imagery, the sensor readings, the spectral signatures — is generated by looking at the territory of the developing nation. Under the current model, that raw data is typically transmitted to ground stations in developed nations, processed in commercial cloud infrastructure, and sold back as a service. The developing nation is, once again, the source of the raw material and the consumer of the finished product, with no ownership stake in the processing layer that creates the value.

Orbital edge compute changes the geometry. If the processing happens aboard the satellite, the raw data never needs to leave the orbital pass over the country’s own territory. The intelligence comes down. The data stays up — or rather, never comes down at all. That is a meaningful shift in data sovereignty, even if it is not a complete one.

The Neutrality Problem, Honestly Examined

Here is where the argument requires the most honest examination, because the neutrality of space is more theoretical than operational in the current environment.

The satellites in low Earth orbit are not neutral. They are owned by companies incorporated in specific jurisdictions, launched on rockets manufactured and regulated by specific governments, and operating under spectrum licenses governed by the International Telecommunication Union in processes where large nations have disproportionate influence. Starlink is American infrastructure. OneWeb has British and Indian ownership. China’s planned LEO constellation is Chinese. The physical neutrality guaranteed by the Outer Space Treaty does not automatically translate into operational or political neutrality in the AI services running on orbital hardware.

What would genuine neutrality require? At minimum, it would require satellites operated under multilateral governance structures — perhaps through regional bodies like the African Union or ASEAN, perhaps through a new kind of orbital infrastructure cooperative modeled on the principles of shared sovereignty that have historically governed other global commons. It would require open-source AI models running on the orbital hardware, not proprietary systems with embedded data-reporting obligations to a foreign government or corporation. And it would require ground station infrastructure in the developing nations themselves, so that processed intelligence does not have to transit through foreign-controlled downlink facilities.

None of this exists at scale today. The International Space Station demonstrates that multilateral space infrastructure governance is possible, if difficult. But the ISS took decades and extraordinary political will to build. The urgency of the AI sovereignty question may not afford that timeline.

Orbital AI won’t end dependence overnight—but it may give smaller nations their first neutral layer for processing, protecting, and building intelligence on their own terms.

The Honest Ceiling and the Real Floor

The tough question for advocates of orbital AI neutrality is the one that the technical specifications force: if space-based compute is edge compute — useful for inference, monitoring, and local data processing, but not for the training runs that determine which foundation models define the world’s AI capabilities — does it actually change the power dynamic, or does it just provide a more sophisticated version of the same dependent relationship?

The honest answer is: it changes it at the margin, significantly, for specific and important use cases, while leaving the deeper structural question of who trains the foundation models entirely unresolved. A Kenyan farmer with satellite-derived crop intelligence that was processed without her country’s data leaving sovereign-adjacent orbital space is genuinely better off than one dependent entirely on a subscription to an American agricultural AI platform. A developing nation with independent deforestation monitoring that it controls and interprets is in a meaningfully stronger negotiating position with international timber markets and carbon credit systems. These are real gains, not trivial ones.

But the nation that cannot train its own models, in its own languages, on its own cultural corpus, will remain dependent on models trained elsewhere for the highest-value AI applications — legal reasoning, medical diagnosis, financial risk assessment, policy analysis. Orbital edge compute does not close that gap. It provides a platform from which to begin closing it, by ensuring that locally generated data can be processed locally before it is harvested by the infrastructure of more powerful nations.

Think of it as the first genuinely neutral layer in a stack that still has many unfair layers above it. It doesn’t solve everything. But it might be the foundation that makes solving everything else possible — a place where smaller nations can stand while they build the rest of what they need.

The sky above the developing world has always been looked at from outside. The new question is whether the nations below it can finally use it to look back — and to think for themselves.

The post The Neutral Sky: Why Space May Be the Only Fair Ground for AI in the Developing World appeared first on Futurist Speaker.

The Token Revolution: How the Global South Becomes the Global Brain

Thomas Frey — Wed, 13 May 2026 23:45:00 +0000

By Futurist Thomas Frey and Futurist Teresa Grobecker

For two centuries, the developing world fed the machine with its land, its labor, and its people. The next economy runs on something different — and this time, the feedback loop runs in reverse.

Here is a prediction that should wake up every policy maker in Washington, every Silicon Valley executive, and every data center lobbyist who thinks America’s lead in artificial intelligence is structurally secured: it is not. In fact, the strategy currently being pursued — restricting data center development, gatekeeping compute resources, treating AI infrastructure like a national security vault — is a textbook example of what systems thinkers call a fixes-that-fail dynamic. A short-term intervention that appears to solve the problem while quietly guaranteeing a worse one downstream. And the nations once on their knees, mining copper, stitching garments, and growing crops for someone else’s table, are about to become the most powerful nodes in the most consequential network humanity has ever built.

Welcome to the age of tokens. The developing world has just been handed the keys — and this time, the system is designed to compound in their favor.

The AI economy mirrors colonial extraction: the world generates the data, a few centers capture the value. Structural shifts are beginning to challenge that imbalance.

The System That Was Always Running

To understand what is shifting, you first have to understand what has been running. The colonial economic model was not simply a political arrangement — it was a system architecture with a reinforcing feedback loop tilted entirely in one direction. Extracted resources flowed from periphery to center. Refining capacity concentrated at the center. Finished goods sold back to the periphery at premium. Profits reinvested in the center’s extractive capacity. Repeat. The loop ran for two centuries, compounding wealth in one direction with devastating efficiency.

Now look at the AI economy as it has operated from roughly 2015 to 2025. The same loop, wearing different clothes. Nigeria’s 220 million people generate stories, images, linguistic patterns, social behaviors — the raw material of machine intelligence. India’s 1.4 billion contribute data in hundreds of dialects and cultural registers that no model can afford to ignore. The favelas of São Paulo, the townships of Johannesburg, the markets of Dhaka — every interaction flows into training datasets owned and monetized by a handful of companies headquartered in a handful of zip codes in California. The developing world generates the stock. The tech economy controls the flow. In systems thinking, whoever controls the valve captures the value of the reservoir, regardless of who filled it. For thirty years, the Global South has been filling the bathtub. Silicon Valley has held the tap. Not one token of return has made it back to the communities that made it possible.

That is about to change — and the mechanism of change is not political. It is structural.

The Reinforcing Loop That Is About to Flip

The reason this moment is categorically different from previous inflection points in the developing world’s economic history is the behavior of the underlying system. The AI training loop — more data produces better models, better models attract more users, more users generate more data — is a classic reinforcing feedback loop. It compounds in whoever’s favor owns the nodes. The entire strategic question of the next decade is: who owns the nodes?

Until now, the nodes were owned by the platforms. The shift underway — through data sovereignty legislation, cooperative data trusts, and sovereign AI infrastructure — is a change in who owns the nodes the loop runs through. That is not a policy tweak. In systems terms, it is a change in the system’s goal, which the late systems theorist Donella Meadows identified as one of the highest-leverage interventions possible in any complex system. When you change who captures the return of a reinforcing loop, you don’t slow the loop. You redirect its entire compounding force.

The nations that were once paid pennies to mine the earth are now sitting on an inexhaustible deposit. And unlike copper, this one compounds every single day — if you own the loop.

What a Token Economy Actually Means in System Terms

When I say token generators, I mean something structurally precise. The next phase of AI development requires nations and communities to negotiate ownership of the data they produce — transforming their role from passive input supplier into active node in the value loop. This is not idealism. It is leverage-point identification.

We are already seeing the early architecture. Kenya, through the Africa Data Centres consortium, is building sovereign compute infrastructure — inserting a nationally owned node into a loop that previously bypassed the continent entirely. India’s homegrown AI models, trained on its own languages and cultural corpus, are a structural intervention: instead of exporting raw linguistic data, India is capturing the refining stage domestically. Brazil’s LGPD privacy framework is, in systems terms, a balancing feedback loop — a corrective mechanism inserted into a runaway extractive dynamic to restore equilibrium. These are not coincidences. They are early moves in a global repositioning, and balancing mechanisms always emerge in runaway systems. The only question is whether they are designed thoughtfully or arrive through disruption.

The transition looks like this: instead of a Nigerian click-farm worker earning two dollars an hour labeling AI training images for an American company, a Nigerian data cooperative earns licensing royalties from every model requiring access to West African linguistic patterns. Instead of Philippine call-center workers training voice AI for foreign firms, Filipino data trusts negotiate multi-year licensing agreements with global platforms that cannot function without them. The mechanism shifts from labor — a flow the market prices at its lowest feasible level — to ownership, a stock position that appreciates as the system scales. That is the whole game.

America is restricting chips while the real leverage shifts to data. Excluding nations from AI infrastructure may accelerate the rise of competing global ecosystems.

The American Miscalculation: Fixing the Wrong Variable

Here is where I have to say something uncomfortable, because the United States is committing a systems error that will be studied in policy schools for decades. The current US posture — restricting advanced chip exports, throttling foreign access to compute infrastructure, treating AI capacity like a weapons stockpile — is premised on an incorrect model of where the leverage point in this system actually sits. It assumes the bottleneck is hardware. It assumes that controlling GPUs means controlling AI outcomes. That logic was coherent in 2019. In 2026, it is what systems thinkers call intervening at the wrong leverage point — applying force to a variable that feels powerful but is not where the system’s behavior is actually determined.

The fixes-that-fail archetype describes interventions that relieve a symptom in the short term while creating side effects that eventually make the original problem worse. US chip export controls reduce adversary compute access today — a genuine short-term effect. The side effects are already compounding: accelerated domestic semiconductor investment in China, deepened AI partnerships between excluded nations and alternative providers, and the systematic erosion of US platforms’ data access in the markets that will generate the majority of the world’s training data for the next thirty years. The fix addresses hardware. The problem is data. The fix fails — and compounds.

What actually drives AI capability at the frontier is not hardware alone — it is the quality, diversity, and cultural breadth of training data. America is not the world’s most data-rich society. It is the society that has been most aggressive about harvesting everyone else’s data without compensation. Once the rest of the world gets organized — once Kenya, Vietnam, Brazil, and Indonesia recognize that their data is their GDP — the American advantage doesn’t erode gradually. It reaches a tipping point and tips.

More critically, by refusing to build data centers abroad and making it difficult for allied nations to access AI infrastructure, the US is triggering the emergence of alternatives — China’s sovereign AI initiative, the UAE’s Falcon program, Europe’s sovereign compute effort, and dozens of regional coalitions now in formation. In systems terms, every excluded node becomes a potential alternative attractor in the network. The US is not protecting a lead. It is distributing the conditions for its own displacement.

The Intellectual Service Economy Follows the Data

Every economy aspires to move up the value chain — from resource extraction to manufacturing, from manufacturing to services, from services to intellectual property. America built its twentieth-century dominance by occupying the top of that pyramid. AI is not just the next rung. It is a new ladder with a fundamentally different structure, one where the inputs are linguistic, cultural, cognitive, and experiential, and where the developing world holds an extraordinary natural endowment.

Consider the emergent properties that systems thinking predicts when distributed data ownership reaches critical mass. When Ethiopia, with over 80 distinct languages, builds a sovereign AI consortium to develop the first truly multilingual African large language model, it is not simply creating a product. It is inserting a new node into the global AI network with properties no existing platform possesses — and that every platform will eventually need. When the Philippines begins licensing its unmatched multilingual conversational corpus as a sovereign asset, it is not competing with American tech companies. It is becoming infrastructure for them, on its own terms. When Mexico, adjacent to the world’s largest AI consumer market with a 125-million-person bilingual population, becomes the world’s premier Spanish-language AI infrastructure hub, it captures a flow that currently exits its economy entirely.

Emergence in systems theory describes properties that arise from the interaction of components but cannot be predicted from any individual component in isolation. When distributed data-ownership networks reach sufficient scale and interconnection, they will generate capabilities — linguistic depth, cultural nuance, behavioral diversity — that no centralized platform, however well-resourced, can replicate. The emergent property of a truly global, sovereign data network is not just more data. It is qualitatively different intelligence. That is the prize no hardware restriction can protect against.

The post The Token Revolution: How the Global South Becomes the Global Brain appeared first on Futurist Speaker.

The Data Centers That Will Float

Thomas Frey — Wed, 06 May 2026 21:32:54 +0000

Why the Most Radical Solution to the AI Energy Crisis Is Already at Sea

By Futurist Thomas Frey

The Ocean Has Been Waiting for This Conversation

There is a moment in every infrastructure crisis when the most obvious solution turns out to be the one nobody was willing to consider. We’ve been having an increasingly urgent conversation about where to put AI’s insatiable appetite for power — and the answer, it turns out, may be covering 71% of the planet’s surface.

Floating data centers. Not as a curiosity. Not as a science experiment. As a genuine, scalable, commercially viable response to the single biggest constraint on the AI revolution.

Peter Thiel apparently agrees. He is leading a $140 million funding round into a company called Panthalassa — named, fittingly, for the ancient superocean that once covered the Earth — which is building floating data centers powered by wave energy. When one of the most consequential technology investors of the last two decades puts $140 million behind an idea, it’s worth understanding exactly what he sees that others don’t.

What a Floating Data Center Actually Is

Strip away the novelty and a floating data center is solving a straightforward engineering problem with a remarkably elegant solution. You need computing power. Computing power generates heat. Heat requires cooling. Cooling requires enormous amounts of energy and water. Land is expensive, permitted, taxed, and increasingly constrained. The grid is aging and overwhelmed.

Now look at the ocean. It is cold. It is vast. It is largely ungoverned. It is already full of the water you need to cool your servers. And in the case of wave energy systems like Panthalassa’s, it is generating mechanical energy twenty-four hours a day, driven by forces that will never send you a bill.

The basic architecture involves a vessel or platform — either a purpose-built structure or a converted ship — housing server racks in sealed, climate-controlled modules. Seawater is circulated as a coolant, either directly or through heat exchangers, replacing the massive air-conditioning infrastructure that typically accounts for 30 to 40 percent of a conventional data center’s energy consumption. Power comes from wave energy converters: devices that capture the kinetic energy of ocean swells and translate it into electricity through linear generators, hydraulic systems, or oscillating water columns.

The result is a facility with no grid dependency, dramatically lower cooling costs, and a power source that is genuinely continuous — not intermittent like solar or wind, but rhythmic and relentless, like the ocean itself.

Microsoft proved underwater data centers worked. The world ignored it—until AI-driven energy demand turned a fascinating experiment into an urgent infrastructure solution.

Microsoft Proved the Concept. Nobody Scaled It.

Here is where the story gets interesting — and where the tough questions begin.

Microsoft ran Project Natick from 2015 to 2022. They submerged a server-packed cylinder off the coast of Scotland in 2018, left it on the seafloor for two years, retrieved it, and found that the hardware failure rate was one-eighth that of comparable land-based systems. One-eighth. The hypothesis was that the stable temperature, lack of human interference, and nitrogen-filled interior produced a dramatically gentler operating environment than a conventional data center. The results were compelling enough that Microsoft published extensive research.

And then… nothing. Microsoft did not build a fleet of underwater data centers. The experiment sat in the archive. Other companies did not rush in to capitalize on the demonstrated proof of concept. Why?

The honest answer involves a combination of factors that felt insurmountable at the time and look increasingly surmountable now. Maintenance is the first: replacing a failed component in a facility under 117 feet of seawater is categorically different from calling a technician. Connectivity is the second: subsea fiber optic cables are expensive, and latency considerations limit how far offshore you can reasonably push compute infrastructure. Regulatory complexity is the third: maritime law, environmental permitting, and jurisdictional ambiguity across international waters create a legal labyrinth that corporate lawyers at large companies are institutionally allergic to.

But the fourth factor — and the most important one — was simply that the energy crisis hadn’t arrived yet. In 2020, nobody was staring down the prospect of data centers consuming 12 percent of U.S. electricity by 2030. The grid seemed adequate. Land seemed available. The problem that floating data centers solve most dramatically hadn’t become urgent enough to justify the complexity.

It has now.

Why Wave Energy Changes the Calculation

Previous floating data center concepts — including Project Natick — still relied on grid power delivered via undersea cable. They solved the cooling problem brilliantly but left the energy dependency intact. Panthalassa’s approach, pairing the floating platform with on-site wave energy generation, closes that loop entirely.

Wave energy has been the perpetually almost-arrived technology of the renewable energy sector. Unlike solar and wind, which suffer from obvious intermittency, ocean waves are driven by wind patterns that operate continuously and predictably. A wave energy converter off the coast of Cornwall generates power at 2 a.m. in January the same as it does at noon in July. For AI infrastructure that cannot tolerate gaps in power delivery, this matters enormously.

The efficiency numbers have historically been the problem. Early wave energy devices were mechanically fragile, expensive to maintain in corrosive salt water, and produced electricity at costs that couldn’t compete with shore-based alternatives. But materials science has advanced significantly in the last decade. Polymer composites, advanced coatings, and better understanding of resonant frequency matching have improved device durability dramatically. And crucially, when your wave energy converter is already sitting next to the thing it powers — eliminating transmission losses entirely — the economic equation shifts.

Think of it this way: a land-based data center in Virginia pays for electricity generated in Pennsylvania, transmitted through aging infrastructure, stepped down through substations, and delivered with 6 to 8 percent transmission losses baked in. A Panthalassa platform generates power ten feet from the servers consuming it. That eliminates an entire layer of cost, inefficiency, and dependency.

The Questions That Deserve Direct Answers

Let’s not pretend this is without complications. Several hard questions sit at the center of this concept, and anyone serious about evaluating it needs to ask them directly.

Can you actually maintain these systems affordably at sea? The Microsoft data suggests hardware runs more reliably in a stable, sealed marine environment. But when something does fail, the economics of marine maintenance — specialized vessels, divers or ROVs, weather windows — need to work at scale. Panthalassa’s $140 million will need to answer this question with real operational data, not just engineering projections.

What does ocean-based computing do to the marine environment? Thermal pollution from heat exchange systems, noise from mechanical wave energy devices, electromagnetic fields from power transmission, and physical obstruction of marine ecosystems are all legitimate concerns. The regulatory frameworks governing these impacts are nascent at best. Unlike land-based data centers, which operate in well-established permitting environments, ocean platforms are entering genuinely ambiguous territory.

Who governs a data center in international waters? This question is simultaneously a legal headache and, for some operators and some data types, potentially a feature rather than a bug. A server rack twelve miles offshore sits in a very different jurisdictional space than one in Northern Virginia. The implications for privacy law, national security review, and data sovereignty are not yet worked out.

And perhaps most pointedly: if this is such an obviously good idea, why did it take until 2026 for serious capital to arrive?

The Infrastructure Inversion

The most interesting thing about floating data centers isn’t the technology. It’s what they represent conceptually: a complete inversion of how we think about the relationship between computing infrastructure and the physical world.

For thirty years, we built data centers the way we built everything else — find land, connect to the grid, manage the heat as best you can. We designed computing infrastructure around the constraints of terrestrial civilization. Floating data centers, particularly wave-powered ones, say something different: take the infrastructure to where the resources are. Cold water is not a resource you bring to the data center. It is a resource you bring the data center to.

That inversion has happened before in other industries. Offshore oil platforms took extraction to where the oil was. Container ships took manufacturing to where labor was cheapest. The logic is the same: when the cost of moving your infrastructure is lower than the cost of moving the resource, you move the infrastructure.

The ocean has been waiting for this conversation for a long time. The AI energy crisis may be exactly the forcing function that finally makes it happen.

IEEE Spectrum — Microsoft’s Underwater Data Center Resurfaces After Two Years https://spectrum.ieee.org/microsoft-underwater-data-center-project-natick

MIT Technology Review — Wave Power Is About to Have Its Moment https://www.technologyreview.com/wave-energy-ocean-power-data-centers

International Energy Agency — Data Centre Electricity Use Surged in 2025, Driving a Scramble for Solutions https://www.iea.org/news/data-centre-electricity-use-surged-in-2025-even-with-tightening-bottlenecks-driving-a-scramble-for-solutions

The post The Data Centers That Will Float appeared first on Futurist Speaker.

The Cyberattack That Could Shake the World Is No Longer a Thought Experiment

Thomas Frey — Sun, 12 Apr 2026 19:31:03 +0000

When Altman warns of a world-shaking cyberattack, it’s not hype—it’s a signal. The capability curve is outrunning preparedness, and the gap is widening fast.

By Futurist Thomas Frey

Sam Altman doesn’t rattle easily.

The man has spent years at the center of the most consequential technological development in human history, fielding questions about existential risk with the calm of someone who has thought about it longer and harder than most of his critics. So when he sits down for an Axios interview in early April 2026 and says that a “world-shaking cyberattack” this year is “totally possible,” it’s worth putting down whatever you’re doing and paying attention.

This isn’t hype. It isn’t positioning. It’s a warning from someone who sees the capability curve up close — and who understands that the gap between what these systems can do and what the world is prepared for is widening faster than most people realize.

What Changed, and When

For most of the history of cybersecurity, large-scale attacks required one of two things: a nation-state with the resources to field an elite hacking team, or a criminal organization with years of accumulated expertise and operational infrastructure. Both existed. Both caused significant damage. But they were constrained by the fundamental bottleneck of human skill — finding the right vulnerabilities, writing the right exploit code, coordinating the right campaign required people who had spent years developing rare capabilities.

AI has just removed that bottleneck.

What once required an elite team can now be automated or AI-assisted: vulnerability discovery, exploit generation, reconnaissance, highly personalized phishing in any language, malware that iterates to evade detection, and full attack chains that connect multiple exploits into a coordinated campaign. According to Red Canary, adversaries are already using large language models for 80 to 90 percent of tactical operations in espionage campaigns. IBM reported a 44 percent spike in public-facing application exploits in 2026, driven in significant part by AI-assisted attacks. Trend Micro has called this year “the AI-fication of cyberthreats.”

This is not a future threat. It is the current situation, and it is accelerating.

The Anthropic Model Nobody Gets to Use

The detail that sharpens all of this from interesting to genuinely alarming came from Anthropic just days ago.

The company has developed a frontier AI model — internally designated Claude Mythos Preview — that can autonomously identify and exploit thousands of high-severity vulnerabilities across every major operating system, every major web browser, and key enterprise software systems. Including zero-days: previously unknown vulnerabilities that no patch exists for, that defenders have no warning about, that an attacker armed with this capability could use before anyone knows they’re there.

Anthropic is not releasing this model publicly. They know exactly what it represents. Instead, they’re sharing limited access with cybersecurity firms through a program called Project Glasswing — a race against time to use the model’s offensive capability defensively, patching the vulnerabilities it finds before a bad actor with similar capability finds them independently.

Read that again. The AI company that built the model decided the responsible thing to do was not release it, and is instead running a controlled program to use its attack capability for defense. That’s a remarkable level of institutional seriousness about what this technology can do. It’s also a signal about where the capability frontier actually sits right now — not where people imagine it will be in five years, but where it is today.

AI removes limits on cyberattacks—scale, speed, and reach explode. Against aging, vulnerable infrastructure, the risk isn’t theoretical anymore. It’s already within reach.

The Scale Problem

Here’s what makes this different from every previous wave of cybersecurity concern.

Past attacks, even sophisticated ones, were constrained by human bandwidth. A team of hackers, however skilled, could only run so many campaigns simultaneously. They had to choose targets, allocate resources, manage operations. The attack surface they could cover at any given time was finite.

AI removes that constraint. A sufficiently capable model can scan massive codebases simultaneously, run parallel campaigns against multiple targets, generate exploit variants faster than detection systems can update their signatures, and do all of this continuously without fatigue. The attack surface that a nation-state or well-resourced criminal organization can cover with AI assistance is orders of magnitude larger than what was possible before.

Altman’s specific concern — a coordinated disruption of critical infrastructure, finance, or supply chains — is the scenario that keeps defense experts up at night. Not because it requires some theoretical future capability, but because the capability to attempt it exists right now, and the systems it would target were largely not designed to withstand this kind of assault.

Defense expert John Arquilla, responding to Altman’s warning, called the risks “certainly real” and pointed to something that doesn’t get enough attention: our baseline cybersecurity is already poor. Most of the infrastructure that runs critical systems — power grids, water treatment, financial networks, healthcare systems — runs on software that is old, under-maintained, and riddled with vulnerabilities that haven’t been patched because the organizations running these systems don’t have the resources or the urgency to patch them. Add AI-assisted offensive capability to that landscape and the arithmetic gets uncomfortable very quickly.

The Arms Race Is Already On

The one genuinely encouraging part of this picture is that defenders are using AI too.

Anomaly detection that would have taken human analysts days to surface is now happening in near real time. Automated patching systems are closing vulnerabilities faster than before. The same capability that makes offensive AI powerful also makes defensive AI more capable — scanning environments for weaknesses, identifying unusual patterns, responding to incidents faster than any human team could.

But here’s the honest assessment: right now, the offense has the advantage. Attacking is inherently easier than defending. An attacker needs to find one way in; a defender needs to close every way in. AI amplifies that asymmetry. The attacker’s AI is scanning your entire surface looking for one opening. Your defensive AI is trying to monitor the entire surface at once. In a resource-constrained environment — which most organizations are — offense wins more often.

That gap will close. The tools are improving on both sides. But the window we’re in right now, before defensive AI catches up to offensive AI at scale, is the window Altman is worried about. It’s the window Anthropic is running Project Glasswing to address. It’s the window that cybersecurity reports from IBM, Red Canary, PwC, Trend Micro, and Health-ISAC are all, independently, identifying as the highest-risk period in the history of digital infrastructure.

AI threats aren’t inevitable—they’re manageable. Basic security now carries real weight. What was best practice yesterday is mission-critical today. The difference is urgency, not possibility.

What This Actually Means

There is a version of this conversation that slides into fatalism — the technology is too powerful, the surface is too large, the bad actors are too motivated, nothing can be done. That version is wrong, and it’s counterproductive.

What can be done at the individual and organizational level is real and meaningful. Strong multi-factor authentication. Network segmentation that limits the blast radius of any single breach. AI-aware monitoring that looks for the behavioral signatures of AI-assisted attacks, which are different from the signatures of human-operated ones. Vulnerability management programs that treat patching as a continuous function rather than a periodic maintenance task. Tabletop exercises that game out the specific scenarios — coordinated infrastructure attack, supply chain compromise, simultaneous multi-vector campaign — that AI capability makes more plausible.

None of that is new advice. What’s new is the urgency. The same recommendations that were good practice last year are now load-bearing. The organizations that treated basic cybersecurity hygiene as optional or aspirational are carrying real and growing risk.

At the policy level, the conversation about AI governance, vulnerability disclosure, and international norms around AI-enabled offensive capability needs to move faster than it has been. Altman is pushing for exactly this. The Anthropic approach with Project Glasswing — coordinated defensive disclosure before offensive capability spreads — is one model. It won’t be sufficient at scale, but it’s a serious attempt to use the technology responsibly in a moment when responsible use is genuinely difficult to define.

The Bottom Line

Sam Altman said a world-shaking cyberattack is totally possible this year. Anthropic built a model capable of finding vulnerabilities across every major operating system and decided not to release it. IBM, Red Canary, and Trend Micro are all saying the same thing from the outside that the AI labs are saying from the inside.

The window is open. The capability exists. The baseline defenses are insufficient.

That’s not a reason to panic. It’s a reason to move. The organizations and governments that treat this as a high-priority operational reality right now — not a planning exercise, not a future scenario — are the ones that will be in a defensible position when the window either closes or something comes through it.

The threat is real. The preparation is optional.

For now.

Breaking: The Company Using Biology to Eat the Plastic Crisis

Thomas Frey — Wed, 08 Apr 2026 20:05:36 +0000

Five billion tons of plastic already surrounds us—and growing.
This isn’t waste; it’s accumulation without end. The real breakthrough will be how we undo it.

By Futurist Thomas Frey

There is a number that should stop you cold.

Five thousand million tons.

That’s how much plastic is currently sitting in landfills, floating in oceans, and embedded in ecosystems around the world. Not the amount produced since plastic was invented — the amount that’s already out there, already dispersed, already working its way into the food chain and the water supply and the bodies of every living creature on Earth. Scientists have found plastic particles in Antarctic sea ice, in the deepest ocean trenches, and in human blood. A liter of bottled water contains, on average, nearly a quarter of a million nanoplastic fragments.

And every year, we add 390 million more tons to the pile.

The recycling system that was supposed to manage this — the one with the little arrows on the bottom of every container — handles roughly 9% of what gets produced. The rest is incinerated, buried, or abandoned. Incineration releases toxic gases. Burial means the plastic sits there for centuries. A plastic fishing line, left alone, takes 600 years to break down. A dental floss container, 80 years. A paintbrush, up to a thousand.

This is the problem that Breaking was built to solve. And the way they’re going about it is unlike anything that’s been tried before.

A Microbe That Eats Plastic for Breakfast

In 2022, researchers at the Wyss Institute for Biologically Inspired Engineering at Harvard University discovered something extraordinary in their lab. A microorganism — not engineered, just found — that could break down plastic by eating it. Not one type of plastic. Multiple types. Including polyolefins, which are the toughest plastics in common use, the ones that have historically resisted every biological degradation attempt on record.

The microbe was catalogued as X-32. And what it does is genuinely remarkable. It breaks down the hydrocarbon chains inside plastic polymers — the chemical bonds that make plastic so durable and so persistent — using those plastics as its primary food source. The byproducts are carbon dioxide, water, and biomass. No toxic residue. No microplastic fragments. Just the basic building blocks of organic chemistry, which the environment already knows how to handle.

In lab tests, X-32 started breaking down paintbrush bristles, fishing wire, and dental floss within five days. At scale, it has demonstrated the ability to degrade up to 90% of certain polyesters and polyolefins in under 22 months. In plastic terms, that is essentially instantaneous.

Breaking, the company that was spun out of Colossal Biosciences in April 2024, launched with $10.5 million in seed funding specifically to develop X-32 into a commercial product. The founding team reads like a who’s-who of synthetic biology: George Church from Harvard, Donald Ingber who founded the Wyss Institute, and CEO Sukanya Punthambaker, a career synthetic biologist who has spent decades working toward exactly this kind of breakthrough.

Ben Lamm co-founded Breaking and serves on its board. Kent Wakeford, who you’ll remember as the co-CEO of Form Bio, is the executive chairman.

The pattern is the same. A tool built inside Colossal’s orbit, spun out when it became clear the problem it was solving was bigger than Colossal’s mission alone.

You can’t restore life in a plastic-filled world. Cleanup isn’t separate from revival—it’s prerequisite. Fix the environment first, or nothing else we bring back will survive.

Why This Connects to Everything Else

Lamm has been direct about why a de-extinction company is in the plastic business. You cannot restore an ecosystem if the ecosystem is full of plastic. The northern white rhino, the woolly mammoth, the Tasmanian tiger — none of them can thrive in an environment saturated with synthetic polymers that their biology has no way to process. Ecosystem restoration and plastic remediation are not two separate goals. They’re the same goal looked at from different angles.

That framing matters because it explains why Breaking isn’t just an environmental startup that happened to spin out of a biotech company. It’s a mission-critical piece of Colossal’s larger puzzle — the piece that has to work before the rest of the restoration agenda can fully work.

The first commercial applications are targeted at the food waste and composting industry, which turns out to be a surprisingly concrete entry point. Food waste in American landfills costs taxpayers $16 billion per year. The reason so much of it goes to landfills rather than compost is that it’s contaminated with plastic packaging that composting facilities can’t process. If X-32 can remove that plastic contamination efficiently and cheaply, it unlocks a massive and largely untapped composting infrastructure — with direct benefits for greenhouse gas emissions, landfill reduction, and soil health.

From there, the roadmap extends to wastewater treatment, marine bioreactors for ocean microplastic cleanup, and industrial waste management. Each application uses the same core technology, scaled and adapted for a different environment.

The Hard Question

There is an obvious question that every thinking person asks when they hear about a microbe that eats plastic: what happens when you release a plastic-eating organism into the environment?

It’s a fair question. Breaking takes it seriously. Lamm has been consistent that X-32 has no known negative environmental ramifications, that it produces only harmless byproducts, and that the team is focused carefully on all regulatory and safety requirements before any open-environment deployment. The initial applications — food waste facilities, industrial wastewater systems, controlled bioreactors — are contained environments where behavior is observable and risks are manageable.

The broader question of deploying engineered organisms in open ecosystems is one that the regulatory frameworks are still catching up to. This is not unique to Breaking. It’s the central challenge of the entire synthetic biology field. The science is moving faster than the governance. That gap is not an argument against the science — it’s an argument for building the governance faster.

What sets Breaking apart from most of the solutions that have been proposed to the plastic crisis is that it actually works on polyolefins. Polyethylene. Polypropylene. The most common plastics in the world, present in virtually every form of packaging, textile, and consumer product. Every previous microbial approach has stumbled on polyolefins because the carbon bonds are simply too strong for most biological systems to break. X-32 breaks them.

From genomes to software to cleanup—this is a coordinated system for rewriting biology itself. The tools are finally matching the scale of the problems we created.

The Bigger Picture

Each company in this series has shown us a different face of the same underlying strategy. Colossal builds the biological tools. Form Bio builds the software to manage the data those tools generate. Breaking takes the synthetic biology capability developed in Colossal’s labs and turns it toward one of the most urgent environmental problems on the planet.

Together, they form something that starts to look less like a collection of companies and more like a coordinated system — one designed to read the living world, understand it, and intervene in it at the level where the real damage is being done.

Plastic is one of the defining problems of the last century. The tools to solve it are, for the first time, starting to look adequate to the scale of the challenge.

Five thousand million tons is a big number. X-32 is a very small organism. But so is every microbe that has ever changed the world.

Up Next: Astromech — the stealth AI startup that just surfaced with a $2 billion valuation and a goal that might be the most ambitious thing Ben Lamm has ever tried: predicting biological change before it happens.

Form Bio: The Operating System for Science

Thomas Frey — Wed, 08 Apr 2026 16:39:37 +0000

Ben Lamm (left) and George Church (right) pose in front of a woolly mammoth.

By Futurist Thomas Frey

When Colossal Biosciences launched in 2021, one of the first things Ben Lamm did was sit down with his team and map out all the software they would need to actually do the work. The list came to 55 different applications and algorithms. Fifty-five separate tools, each handling a different piece of the research pipeline, none of them talking to each other particularly well, none of them designed for the kind of work Colossal was trying to do.

There was no single platform that could take a scientist from a raw idea all the way through data analysis, workflow management, result visualization, and collaboration with researchers at other institutions. Not for this kind of biology. Not at this scale. Not with the complexity that de-extinction research demands.

So they built one.

And then — almost by accident — they realized they’d built something the entire life sciences industry had been waiting for.

The Problem Nobody Had Solved

Here’s what biological research actually looks like inside a modern lab, away from the glamour of the headlines. A scientist has a dataset — maybe a genome sequence, maybe the results of a CRISPR editing experiment, maybe microarray analysis from a gene therapy trial. That dataset is enormous. It connects to other datasets. It needs to be analyzed using computational models, cross-referenced with other results, validated through additional experiments, and eventually shared with collaborators at other universities or companies who are using completely different software systems.

In most labs today, that process is held together with institutional knowledge, personal preference, and a lot of custom code that one specific researcher wrote and that no one else fully understands. When that researcher leaves, a piece of the lab’s institutional memory walks out the door with them.

The situation at Harvard, where Colossal’s co-founder George Church runs one of the world’s most advanced genetics labs, was typical. Fifty-five different data systems in active use. Researchers from Colossal and Harvard trying to collaborate, but with no common infrastructure for sharing experiments, workflows, or results in a way that was consistent and reproducible.

“There was no cohesive solution,” said Kent Wakeford, who became co-CEO of Form Bio when it spun out. “So we developed one.”

Science is becoming software. When biology runs on integrated platforms, discovery accelerates, collaboration scales, and the real breakthrough isn’t the experiment—it’s the infrastructure powering it.

What Form Bio Actually Does

The simplest way to describe Form Bio is this: it’s what happens when you apply software product thinking to the workflow of science.

Scientists aren’t typically software engineers. The tools they use were mostly built by other scientists or small academic teams, optimized for specific tasks, and never designed to work together as a system. Form Bio replaces that patchwork with a single integrated platform — one place where a researcher can design an experiment, run computational analysis using AI and machine learning models, visualize the results, and share everything with collaborators anywhere in the world, with proper permissions and data security built in.

George Church, who has spent decades running one of the most productive genetics labs on the planet, put it plainly: the platform is “critical to pave the way” for the kind of science that’s now becoming possible. When one of the architects of modern genomics says your software is necessary infrastructure, that’s not a testimonial. That’s a signal.

The use cases stretch well beyond de-extinction. Drug discovery. Gene therapy development — specifically the design of AAV vectors, which are the delivery vehicles used to get gene-editing tools into human cells. Biomanufacturing. Agricultural biotech. Academic research across every field that generates large biological datasets, which is most of them now. The CIA’s venture arm, In-Q-Tel, invested in Colossal specifically — by their own admission — not because of the mammoths, but because of the underlying capability. The computational biology infrastructure is what interested them.

The Pattern Behind the Spinout

Form Bio was spun out of Colossal in September 2022 with a $30 million Series A that was oversubscribed — meaning investors wanted in faster than the round could close. It launched as an independent company with its own leadership team, its own staff, and its own capital structure, while maintaining a close relationship with Colossal as both a customer and a co-development partner.

This is a pattern worth paying attention to, because it’s not an accident. Lamm has been explicit that Colossal’s long-term strategy involves spinning out the technologies built in the process of doing the research — letting each tool become its own company, raise its own capital, and pursue its own market, rather than trying to run everything under one roof.

It’s the same instinct that made NASA’s technology transfer program one of the most productive sources of commercial innovation in American history. When you’re solving genuinely hard problems at the frontier of what’s possible, you generate tools that have value far beyond the original problem. The question is whether you’re organized to capture that value. Lamm is organized to capture it.

Form Bio is the first example. Breaking — the plastic degradation company built on Colossal’s synthetic biology infrastructure — is the second. Astromech, the predictive biology AI that surfaced publicly just last week, is the third. Each one started as internal tooling built to solve a specific problem inside Colossal. Each one turned out to be a product.

Form Bio was born trying to bring back a woolly mammoth. Where it ends up may be considerably larger than that.

Why This Matters Beyond Biology

There’s a larger story here about what happens when software thinking meets science.

Academic research has always moved slowly, and part of the reason is structural. Scientists work in relative isolation, each lab developing its own methods, its own tools, its own ways of doing things. Reproducibility — the ability for another lab to run the same experiment and get the same result — is one of the most persistent problems in modern science, and a lot of it comes down to the fact that the computational infrastructure for sharing and standardizing workflows simply hasn’t existed.

Form Bio is building that infrastructure. The comparison its co-CEO reached for was GitHub — the platform that transformed software development by giving programmers a shared environment for building, testing, and collaborating on code. What GitHub did for software, Form Bio wants to do for biology. Create a common layer. Make the workflows reproducible. Let researchers spend their time on science instead of on data wrangling.

That’s not a small ambition. Biology is becoming the defining technology of this century in the same way that computing defined the last one. The platform that becomes the operating system for biological research — the place where scientists from Cambridge to Tokyo to Dallas all run their experiments and share their discoveries — will be one of the most consequential pieces of software infrastructure ever built.

Up Next: Breaking — how Colossal’s synthetic biology toolbox turned into a potential solution for 5,000 million tons of plastic.

Colossal Biosciences: The Moonshot That Became a Business

Thomas Frey — Wed, 08 Apr 2026 02:56:02 +0000

Ben Lamm (right) co-founded Colossal Biosciences with Harvard geneticist George Church (center) in 2021

By Futurist Thomas Frey

In April 2025, three wolf pups were born that shouldn’t exist.

Their names were Romulus, Remus, and Khaleesi. They were healthy, they were photographed, and they made the cover of Time magazine. They were also, genetically speaking, something the world hadn’t seen in 13,000 years. The dire wolf — the apex predator of the Ice Age, the creature that once ruled North America alongside the woolly mammoth and the saber-toothed cat — had been gone since before recorded human history.

Then Ben Lamm brought it back.

Peter Diamandis, who has spent his career finding the most audacious things happening in science and technology, called it the scientific miracle of the decade. He’s not given to overstatement.

But here’s the thing most people missed in all the excitement: the wolf wasn’t the goal. It was the proof.

What They’re Actually Building

Ben Lamm co-founded Colossal Biosciences with Harvard geneticist George Church in 2021. When he announced that the company intended to bring back the woolly mammoth, most people assumed it was a publicity stunt — a flashy headline wrapped around a research project that would never really get anywhere.

They were wrong about that.

What Lamm understood from the beginning is that you don’t build a company around an animal. You build it around the tools you need to get to the animal. And those tools — new ways to read ancient DNA, new methods for editing genes with surgical precision, new technology for growing embryos outside a living body — those turn out to be useful for a lot more than bringing back one extinct species.

Think of it this way. When NASA developed materials that could survive the heat of reentry from space, those materials didn’t stay in rockets. They ended up in everything from firefighting gear to running shoes. The technology built for one extreme purpose found its way into everyday life because the underlying science was genuinely new and genuinely powerful.

That’s what Colossal is doing with biology. The de-extinction projects are the extreme purpose. The tools being built to achieve them are the real product.

Extinction isn’t permanent anymore. With $10B backing and gene editing breakthroughs, life itself is becoming programmable—revived, redesigned, and repurposed at a scale we’re just beginning to grasp.

The Numbers Tell the Story

By January 2025, Colossal had raised $435 million in total funding. A $200 million Series C round valued the company at $10.2 billion, making it Texas’s first decacorn — a startup worth more than ten billion dollars. The investor list includes venture capitalists, institutional funds, and more than a few celebrities whose names you’d recognize. That’s not an accident. Lamm has always understood that a company trying to do something this large needs cultural momentum, not just scientific credibility.

But the science is what earns the valuation. To bring back the dire wolf, Colossal’s team extracted usable DNA from a 13,000-year-old tooth and a 72,000-year-old ear bone. They reconstructed the genome, identified the 20 key traits that made a dire wolf a dire wolf, and rewrote 14 genes in 45 engineered eggs. Those eggs became embryos. Those embryos were carried by surrogate hound mixes. Those surrogates gave birth to three healthy pups.

It worked. On the first attempt at this scale. That is not a small thing.

The mammoth program has already produced what the team calls “woolly mice” — ordinary lab mice gene-edited with mammoth DNA traits, including the distinctive shaggy tawny fur — as a proof that the genetic approach works before they try it on something the size of a bus. The Tasmanian tiger program has reconstructed a 99.9% accurate genome from a 110-year-old skull. A partnership with filmmaker Peter Jackson and the indigenous Māori people of New Zealand is underway to attempt to bring back the moa, a giant flightless bird that disappeared after human settlement of the islands.

Each project is different. Each one pushes the tools further. And each time the tools get pushed further, they get more useful for everything else.

The Part Nobody’s Talking About

There’s a piece of the Colossal story that gets almost no attention in the mainstream coverage, but that may turn out to matter more than the animals themselves. It’s the artificial womb program.

Here’s the problem it solves. To bring back an extinct species, you usually need a living relative that’s close enough in biology to carry the embryo. For the woolly mammoth, that relative is the Asian elephant. But the Asian elephant is itself endangered — and Lamm has been consistent and clear that Colossal will not put endangered animals through invasive reproductive procedures to serve the mammoth project. So they’re building a womb instead. An artificial one. A device that can take a fertilized single-cell embryo and bring it all the way to a live birth without a living surrogate.

Lamm has said publicly that he expects this to work — with a small mammal first, then scaling up — by the end of 2026. If it does, the implications go well beyond anything Colossal is currently working on. A technology that can grow a living mammal from a single cell to birth in an artificial environment is not a tool that stays in one laboratory. What it means for the survival of critically endangered species, for the future of reproductive medicine, for conservation biology across the board — that conversation is only just beginning.

Lamm has drawn a clear line: Colossal won’t apply this technology to humans or non-human primates. But drawing a line around your own use of a tool doesn’t make the tool disappear. It just means someone else will eventually have to decide where to draw theirs.

Startup speed meets science—extinction is shifting from fate to choice. What once took decades now happens in years, turning resurrection into a commercial, repeatable reality.

A Tech Company Wearing a Lab Coat

What makes Lamm’s approach genuinely different from how science normally works is that he’s running Colossal the way he ran his software companies. Fast. Capital-intensive. Outcome-focused. Willing to hire the best people from academia and pay them what the private sector pays, which is not what universities pay. Willing to fail publicly and talk about it openly, which scientists are not traditionally trained to do. And obsessively focused on the question of how the technology gets commercialized — because without a business model, there is no mission.

The result is a company that has moved faster in four years than the academic de-extinction community moved in twenty. That’s not a criticism of the scientists who came before. It’s a description of what happens when you apply startup discipline to a problem that used to live entirely inside universities.

The dire wolf is alive. The woolly mouse exists. The Tasmanian tiger’s genome is reconstructed and waiting. These are not predictions about the future. They are things that have already happened.

The mammoth is next. And after that, if Ben Lamm has his way, extinction itself becomes optional.

Next week: Form Bio — the AI scientific software platform that Colossal built for itself, then realized the entire life sciences industry needed.

80 Years to an Overnight Success: The Real History of Artificial Intelligence

Thomas Frey — Sat, 04 Apr 2026 21:58:36 +0000

By Futurist Thomas Frey

From a shirtless philosopher in 1943 to ChatGPT — the people, the breakthroughs, the winters, and the single idea that refused to die

The Man Without a Shirt

In January 2026, Marc Andreessen sat down for an 81-minute podcast conversation on the a16z show and did something most technology commentary doesn’t bother to do: he started at the beginning. Not the beginning of this AI cycle, or large language models, or even deep learning. He started in 1943 — with a paper, a seaside villa, and a neurophysiologist who, in archived footage from 1946, can be seen discussing the future of computing without a shirt on, apparently unbothered by the formality the topic deserved.

That man was Warren McCulloch. His observation — that computers could one day be built on the model of the human brain, using neural networks rather than pure mathematical logic — was the road not taken for most of the next eight decades. Andreessen’s point was simple and important: what feels like an overnight revolution is actually the payoff on an 80-year bet made by a small group of people who spent most of that time being ignored, defunded, and told they were wrong. Understanding that history explains why what’s happening now is different from everything that came before — and why it is probably not going to stop.

Warren McCulloch, the shirtless philosopher who’s thinking on the first artificial neuron proved a radical idea: intelligence could be built, not programmed—setting a path that would take decades to fully unfold.

1943: The Paper That Started Everything

Warren McCulloch — a neurophysiologist — and Walter Pitts — a mathematical prodigy who had run away from home as a teenager to attend university lectures and was, at the time, technically homeless — published “A Logical Calculus of the Ideas Immanent in Nervous Activity” at the University of Chicago. The paper proposed the first mathematical model of a neural network: an artificial neuron that received inputs, applied weighted thresholds, and fired an output based on logical rules.

The idea embedded in it was radical: that the logic of the human brain could be formally described and computationally replicated — not mimicked through clever programming, but actually replicated through interconnected units that learned by adjusting their own weights. The computer industry took a different road: building literal mathematical machines to execute explicit instructions at enormous speed. The neural path would take 80 more years to fully develop. But it was always there, tended by a minority who believed it was the more important direction. John von Neumann cited the paper. Norbert Wiener found it foundational. Marvin Minsky, later one of AI’s central figures, was influenced by McCulloch and built an early neural network in 1951 using 3,000 vacuum tubes to simulate 40 neurons. The seed was planted.

“Can machines think?”—Alan Turing’s one question in 1950 ignited a field that is now reshaping what it means to be human.

1950–1956: Turing’s Question and the Birth of a Field

In 1950, Alan Turing published “Computing Machinery and Intelligence” and opened with the question that became the field’s defining provocation: “Can machines think?” Rather than get lost in philosophy, he proposed a practical test — if a machine could convince a human judge through text conversation alone that it was human, that was sufficient evidence of intelligence worth taking seriously. The Turing Test was born.

Six years later, John McCarthy organized a two-month workshop at Dartmouth with an audacious premise. McCarthy, Marvin Minsky, Nathaniel Rochester of IBM, and Claude Shannon — the father of information theory — claimed that “every aspect of learning or any other feature of intelligence can in principle be so precisely described that a machine can be made to simulate it.” It was at Dartmouth in 1956 that McCarthy coined the term “artificial intelligence.” The field had a name, researchers, and ambition. What it lacked for several decades was the ability to deliver.

1957–1969: The Perceptron, the Promise, and the First Winter

In 1957, psychologist Frank Rosenblatt built the Perceptron at Cornell — the first artificial neural network capable of learning from data, updating its own internal connections based on errors. The Navy funded it. The New York Times declared it would one day “walk, talk, see, write, reproduce itself and be conscious of its existence.” By the mid-1960s, Perceptrons were everywhere.

Then Minsky — who had been Rosenblatt’s classmate at the Bronx High School of Science — published “Perceptrons” in 1969 with Seymour Papert. The book proved mathematically that a single-layer network could not solve basic logical functions like XOR. Funding collapsed. Researchers fled to symbolic AI. The first AI Winter arrived. The tragedy, which Minsky later acknowledged, was that the book also noted multi-layer networks could solve XOR — but nobody yet knew how to train them. That problem would take seventeen more years to crack.

Jeffrey Hinton’s backpropagation didn’t win immediately—but it made deep learning possible, waiting decades for compute to catch up and unlock modern AI.

1986: The Algorithm That Changed Everything

Geoffrey Hinton spent the AI Winter years convinced the brain’s massive parallelism held the key to machine intelligence, continuing to work on neural networks when doing so was roughly equivalent to professional suicide. In 1986, Hinton, David Rumelhart, and Ronald Williams published “Learning Representations by Back-Propagating Errors” — the paper that solved the credit assignment problem and made training multi-layer networks mathematically tractable.

Backpropagation worked by running the network forward, measuring output errors, then propagating that error signal backward through every layer — adjusting each connection proportionally to its contribution to the mistake. Neural networks could now learn all the way down. A second brief spring followed, then a second winter. The 1990s saw expert systems — elaborate rule-based programs — briefly dominate before proving too brittle and expensive to maintain. Funding dried up again. But backpropagation was real. The tool existed. It was waiting for computers fast enough to use it at scale.

The Quiet Years: LeCun, Bengio, and the Believers

Through the 1990s and 2000s, a small community kept the neural network program alive at the margins. Yann LeCun at Bell Labs demonstrated that convolutional neural networks could read handwritten digits reliably enough for real bank check-processing systems — actual commercial deployment, quiet and largely unnoticed. Yoshua Bengio at the University of Montreal published foundational work on language models and distributed word representations — intellectual precursors to the large language models that would arrive two decades later. Hinton, LeCun, and Bengio — who would share the 2024 Nobel Prize in Physics for their contributions to machine learning — continued building theoretical and empirical foundations through years when the dominant sentiment was that deep learning was a dead end. They were wrong about that. The rest of the field was wrong about them.

2012: The Moment That Started the Current Era

In October 2012, Geoffrey Hinton, his student Alex Krizhevsky, and Ilya Sutskever entered the ImageNet visual recognition competition with a deep convolutional neural network called AlexNet. They won — and the margin shocked everyone. AlexNet achieved a top-5 error rate of 15.3 percent. The next best entry was 26.2 percent. That is not incremental improvement. It is a discontinuity — the kind of gap signaling that one team was playing a fundamentally different game. The key was the combination of deep neural network architecture, a massive training dataset, and GPUs repurposed for the parallel matrix mathematics training required. Hinton later summarized it with characteristic dryness: “Ilya thought we should do it, Alex made it work, and I got the Nobel Prize.”

Within months, every major technology company was hiring neural network researchers. Google acquired a startup Hinton had founded. Facebook opened an AI lab. The money that had twice abandoned the field came back — and this time the technology actually worked at real scale on real problems with real economic value. The third spring arrived. Unlike the first two, it did not end.

When Google’s DeepMind mastered Go and defeated Korea’s Lee Sedol, it wasn’t just a game—it was the moment human intuition met a machine that could outthink it.

2017 and Beyond: Transformers, Scale, and the Arrival

In June 2017, eight Google researchers published “Attention Is All You Need” — perhaps the most important research paper in AI history. The transformer architecture replaced sequential recurrent networks with self-attention: a mechanism letting every part of a sequence simultaneously consider every other part, weighted by relevance. Transformers could be trained in parallel, scaled to far larger datasets, and — crucially — their capabilities improved in ways not fully predictable from smaller versions. Scale the model, scale the data, scale the compute: get a qualitatively better system. This scaling law drove everything that followed.

OpenAI’s GPT series demonstrated the trajectory — GPT-1 in 2018, GPT-2 in 2019, GPT-3 in 2020 with 175 billion parameters — each generation capable of things the previous one could not do at all. DeepMind’s AlphaGo in 2016 mastered Go well enough to defeat the world’s best human player. AlphaFold in 2020 solved the protein folding problem that had challenged structural biologists for 50 years. Then on November 30, 2022, OpenAI released ChatGPT. One hundred million users in two months — the fastest adoption of a consumer technology in history. Not because it introduced new capabilities, but because a conversational interface made the full power of a large language model legible to anyone with a browser. Millions of people sat down, typed a sentence, and watched something that felt like thinking happen in response.

The People Who Made It Happen

McCulloch and Pitts provided the foundational concept. Turing provided the philosophical framework and organizing question. McCarthy named the field. Minsky built its institutional architecture, even as his 1969 book nearly killed it. Rosenblatt gave the field its first learning machine. Hinton kept neural networks alive through two winters, solved the training problem, and watched with a mixture of pride and growing concern as the systems he helped build became more capable than he had anticipated. LeCun gave the field convolutional networks and the first proof that learned representations could outperform hand-engineered features at real-world scale. Bengio provided much of the foundational theory and became the field’s most prominent voice on safety. Sutskever co-authored AlexNet, co-founded OpenAI, and drove the GPT series before leaving to found Safe Superintelligence. Ilya’s former mentor Sam Altman made the decision to release ChatGPT publicly — turning an abstract technical debate into a mass cultural experience.

Andreessen’s 80-year framing is not historical interest for its own sake. It is a structural argument about where we are. The technologies that reshape civilizations almost never arrive on the schedule their inventors expect. They require the convergence of the right idea, the right hardware, the right data, and the right moment of public readiness. Usually the idea comes first and waits decades for the rest. What began as a shirtless philosopher’s conversation about building machines on the model of the brain has become the most consequential technological transition of our lifetimes. It took 80 years. It was worth the wait.

Ten Years Ago I Made 72 Predictions About 2026. Here’s the Honest Report Card — and What 2036 Actually Looks Like

Thomas Frey — Tue, 24 Mar 2026 16:08:45 +0000

By Futurist Thomas Frey

A decade-old list, graded in real time — plus the next ten years

Back in August 2016, I sat down and published a piece called “72 Stunning Things in the Future That Will Be Common Ten Years from Now That Don’t Exist Today.” I covered 3D printing, VR, drones, driverless cars, the Internet of Things, health tech, AI, and transportation. I gave myself a decade. The decade is up. Time to pay the bill.

The short version: some of it landed almost exactly right. Some of it was right in concept but wrong on timing. A few items missed completely. And one category — AI — I almost certainly undersold rather than oversold, which is the kind of mistake I find most interesting to examine.

The Solid Hits

The VR and AR predictions held up remarkably well. Theme park rides mixing physical experiences with VR — fully real and widespread. Live sports in virtual reality — done, including NFL, NBA, and soccer broadcasts. VR therapy for physical and psychological conditions — now a recognized clinical modality used in hospitals for pain management, PTSD treatment, and phobia exposure therapy. VR and AR tours in real estate — completely standard. That entire category was largely on target.

The health tech predictions also aged well in aggregate. Telehealth checkups without a doctor’s appointment — COVID accelerated that from a nice-to-have to a healthcare pillar almost overnight. AI-controlled prosthetic limbs — real, advancing rapidly, and genuinely changing lives. Ingestible data collectors with sensors — early commercial versions exist, and continuous glucose monitors have become mainstream for diabetics and increasingly popular among health-conscious people who aren’t diabetic at all. Real-time blood scanners are still evolving, but the direction was right.

The drone predictions were solid, particularly fireworks launched from drones — that specific prediction now has an entire FAA-approved industry behind it, with pyro drones appearing at major stadiums and city celebrations across the country. Bird-frightening drones for agriculture, livestock monitoring drones, and drone use in entertainment all landed as predicted. Drone racing viewed through VR headsets became a legitimate organized sport with professional leagues and broadcast deals, another clean hit.

On AI, I predicted that best-selling books and legal documents would be written by artificial intelligence, that AI would select movies, music, and menus based on personal preferences and moods, and that AI hackers would emerge as a serious threat. All of that is not just real — it’s so thoroughly embedded in daily life that most people have stopped noticing. Netflix recommendations, Spotify playlists, AI-drafted contracts and briefs — these are baseline expectations now, not futuristic concepts.

Biometric payment systems were on my IoT list, and fingerprint and face recognition payments are now so standard they barely register as technology. 360-degree video cameras at major urban intersections are common in cities worldwide. Everywhere wireless connectivity — through Starlink, expanded cellular infrastructure, and other systems — is now real and still expanding. Robotic bricklayers are operational. And a privacy bill of rights materialized, though unevenly — GDPR in Europe, CCPA in California, and an ongoing global patchwork of digital privacy regulation that continues to evolve.

The future didn’t miss—it’s just running late. The technology works, but regulation, trust, and adoption are still catching up.

The Partial Hits — Right Direction, Wrong Timing

Driverless cars are the most prominent partial credit. I predicted driverless car hailing apps, large fleet ownership of autonomous vehicles, and in-car work and entertainment systems — all real, but not yet common in the way I imagined. Waymo is operating in a handful of U.S. cities. Tesla’s robotaxi network is expanding. But the mass adoption I envisioned by 2026 hasn’t arrived. The technology largely works. The regulatory framework, insurance ecosystem, and public trust are still catching up.

I predicted crash-proof cars, specifically citing Volvo’s pledge to achieve that by 2020. That was not met. Advanced collision avoidance systems are now standard on most new vehicles and are saving lives — but truly crash-proof is still a work in progress. EV charging in under five minutes is not yet standard, though the technology is advancing rapidly and that milestone is genuinely within reach in the next few years.

3D printed replacement teeth and custom-fitted shoes and clothing from in-store scanners exist in prototype or limited commercial forms, but haven’t reached the mass retail ubiquity I described. Same-day dental crowns printed in-office are now common in dental practices, so the teeth prediction is closer than it looks. The clothing and shoes have the technology behind them but the consumer journey hasn’t fully standardized. These feel like 2028-2030 arrivals rather than 2026 ones.

The smart IoT household items — smart beds, smart plates tracking nutrition, smart mailboxes — have partial implementations but haven’t reached the seamless mass-market penetration I expected. Eight Sleep’s smart mattress is a real product used by hundreds of thousands of people. Continuous glucose monitors track what you eat and how your body responds. The infrastructure is forming; the widespread daily use hasn’t quite arrived.

The Misses

Hyperloop — ultra-high-speed tube transportation — was prediction number 64, and I said it was something “the only thing lacking is a few people capable of mustering the political will to make it happen.” A decade later, most hyperloop ventures have quietly folded or dramatically scaled back ambitions. Virgin Hyperloop shut down its passenger program. The technology proved far more expensive and complex than its promoters suggested, and the regulatory and infrastructure challenges were even more formidable than I acknowledged. That one missed.

Electric cars winning the Daytona 500 and Indy 500 hasn’t happened. Electric racing series exist and are growing — Formula E is real and exciting — but the major traditional races haven’t converted. That was probably too specific a prediction, conflating the trajectory of EV adoption with the far more conservative pace of change in established motorsport institutions.

Personal drone transportation — unmanned aviation for individual people — I listed as prediction 57, and while eVTOL air taxis are being tested and certified, they are not yet common by any definition. This one needed more time, and the honest timeline is probably closer to 2028-2032 for meaningful urban deployment.

Self-retrieving shoes and robotic follow-behind luggage were creative ideas that haven’t materialized in any practical sense. Some prototype robotic luggage exists. Nobody is calling their shoes by name yet.

What I Undersold

The AI section is where I was least bold, not most bold. I predicted AI-written documents and AI content recommendations — which happened exactly as described. But I completely missed the civilizational magnitude of what large language models would become by 2026. I didn’t predict that AI would write code well enough to replace junior programmers, that it would generate photorealistic images on demand, that it would hold multi-hour conversations indistinguishable from human interaction, or that the entire global economy would be reorganizing itself around AI adoption in real time. My AI predictions were right but timid. The future was much bigger than the list.

The future doesn’t arrive evenly—it seeps in early, spreads fast, and suddenly becomes the new normal before most people notice.

What 2036 Actually Looks Like

The lesson from grading the 2016 list is that transformation usually arrives on schedule — just unevenly distributed. Things that seemed far off are already here for some people. Things that seemed imminent took longer than expected. With that humility established, here is where the next decade is heading.

By 2036, humanoid robots will be genuinely common in warehouses, hospitals, and manufacturing settings, and will be beginning to appear in homes. The Optimus, Figure, and other platforms being tested today will have completed their first commercial deployments and will be in their second and third hardware generations. The workforce disruption this creates will be the dominant political and economic story of the late 2020s and early 2030s.

Autonomous vehicles will have finally crossed into genuine mass adoption in most major cities. The regulatory and insurance frameworks that have delayed deployment in 2026 will have been resolved by necessity — too many people will have used autonomous ride services in too many cities for the holdouts to maintain their position. Owning a personal car will begin to feel unnecessary for urban residents in the way owning a horse began to feel unnecessary after World War I.

AI will be so embedded in daily professional life by 2036 that describing it will feel like describing oxygen. Every knowledge worker will have AI systems that know their work style, priorities, communication patterns, and professional history. The question won’t be whether to use AI but how to maintain the distinctly human judgment and creativity that AI cannot replicate. That will be the skill that commands premium compensation.

Personal health monitoring will have crossed a threshold where most chronic disease is managed in real time rather than treated after the fact. Continuous monitoring of blood glucose, cardiac rhythms, inflammation markers, and hormonal levels — all via non-invasive wearables — will give individuals and their physicians a real-time biological picture that makes today’s annual physical look like guesswork. Personalized drug dosing and AI-driven treatment recommendations will be standard practice.

Space will have moved from aspiration to infrastructure. The first permanent human presence on the Moon — research teams, not tourists — will be underway. Orbital data centers, powered by solar energy and cooled by space vacuum, will be handling a meaningful portion of global AI compute. The idea that all of civilization’s intelligence runs on Earth will already seem like a transitional phase rather than a permanent condition.

The honest summary of grading the 2016 list is that the direction was right more often than not — the technologies I pointed to were real and consequential. The errors were mostly in the magnitude and timing. I was too conservative on AI and too optimistic on autonomous vehicles and hyperloop. If that pattern holds for the 2036 projection — and it probably will — then the next decade will be bigger than this column describes in some areas, and slower than it describes in others. That’s the nature of forecasting. The future always surprises on the upside in unexpected places, and disappoints in the ones you were most confident about.

Predictions Archives - Futurist Speaker

The Asimov Manifesto

We Are Already Living in the World He Warned Us About

Efficiency Is Not a Virtue When the Goal Is Destruction

The Child Who Grows Up Afraid of the Sky

Morality Must Be Built In, Not Bolted On

The Five Principles We Must Enshrine

What We Build Next Defines Who We Are

Related Articles

The Neutral Sky: Why Space May Be the Only Fair Ground for AI in the Developing World

What Orbital Edge Compute Actually Is

The Eye in the Sky, Reconceived

The Neutrality Problem, Honestly Examined

The Honest Ceiling and the Real Floor

Related Articles

The Token Revolution: How the Global South Becomes the Global Brain

The System That Was Always Running

The Reinforcing Loop That Is About to Flip

What a Token Economy Actually Means in System Terms

The American Miscalculation: Fixing the Wrong Variable

The Intellectual Service Economy Follows the Data

Related Articles

The Data Centers That Will Float

Why the Most Radical Solution to the AI Energy Crisis Is Already at Sea

The Ocean Has Been Waiting for This Conversation

What a Floating Data Center Actually Is

Microsoft Proved the Concept. Nobody Scaled It.

Why Wave Energy Changes the Calculation

The Questions That Deserve Direct Answers

The Infrastructure Inversion

Related Articles

The Cyberattack That Could Shake the World Is No Longer a Thought Experiment

What Changed, and When

The Anthropic Model Nobody Gets to Use

The Scale Problem

The Arms Race Is Already On

What This Actually Means

The Bottom Line

Related Reading

Breaking: The Company Using Biology to Eat the Plastic Crisis

A Microbe That Eats Plastic for Breakfast

Why This Connects to Everything Else

The Hard Question

The Bigger Picture

Related Reading

Form Bio: The Operating System for Science

The Problem Nobody Had Solved

What Form Bio Actually Does

The Pattern Behind the Spinout

Why This Matters Beyond Biology

Related Reading

Colossal Biosciences: The Moonshot That Became a Business

What They’re Actually Building

The Numbers Tell the Story

The Part Nobody’s Talking About

A Tech Company Wearing a Lab Coat

Related Reading

80 Years to an Overnight Success: The Real History of Artificial Intelligence

The Man Without a Shirt

1943: The Paper That Started Everything

1950–1956: Turing’s Question and the Birth of a Field

1957–1969: The Perceptron, the Promise, and the First Winter

1986: The Algorithm That Changed Everything

The Quiet Years: LeCun, Bengio, and the Believers

2012: The Moment That Started the Current Era

2017 and Beyond: Transformers, Scale, and the Arrival

The People Who Made It Happen

Related Reading

Birthplace of Neural Networks: McCulloch & Pitts at UChicago University of Chicago — The 1943 paper that started everything, and the institutional context that made the McCulloch-Pitts collaboration possible

The Real Story of AI: From Turing to ChatGPT DatBot.AI — A detailed narrative of the full AI timeline: two winters, three booms, and the technology that finally worked

Marc Andreessen’s 2026 Outlook: AI Timelines, US vs. China, and the Price of AI a16z / Spotify — The January 2026 podcast where Andreessen traces AI’s arc from the 1943 neural network paper to today’s reasoning models

Ten Years Ago I Made 72 Predictions About 2026. Here’s the Honest Report Card — and What 2036 Actually Looks Like

The Solid Hits

The Partial Hits — Right Direction, Wrong Timing

The Misses

What I Undersold

What 2036 Actually Looks Like

Related Reading

Birthplace of Neural Networks: McCulloch & Pitts at UChicago
University of Chicago — The 1943 paper that started everything, and the institutional context that made the McCulloch-Pitts collaboration possible

The Real Story of AI: From Turing to ChatGPT
DatBot.AI — A detailed narrative of the full AI timeline: two winters, three booms, and the technology that finally worked

Marc Andreessen’s 2026 Outlook: AI Timelines, US vs. China, and the Price of AI
a16z / Spotify — The January 2026 podcast where Andreessen traces AI’s arc from the 1943 neural network paper to today’s reasoning models