Here is the truth that took me a long time to say plainly: we do not have a healthcare system. What we have is an extraordinarily expensive, extraordinarily sophisticated sickness management infrastructure — built around the moment of crisis rather than the prevention of it. And almost everyone inside the machine knows it.

I have filmed in hospitals across five continents. I have sat with oncologists in Dubai, neurologists in Boston, health economists in Geneva. The conversation always arrives at the same place. The system is reactive by design. It responds to disease. It does not prevent it. And the reason is structural, not accidental: reactive medicine is, commercially speaking, far more profitable.

Think about it from first principles. If you are healthy, you generate no revenue for the healthcare system. The moment you are sick, you become enormously valuable. Diagnostics, drugs, specialists, hospitalizations, follow-up treatments — the meter runs from the moment of diagnosis. A system optimized for prevention would be a system optimizing itself out of its primary revenue stream.

When I spoke to the experts featured in this episode of The Futurist, the consensus was jarring in its clarity. The current model is not sustainable. Demographic aging, rising chronic disease burdens, and accelerating treatment costs have put every major healthcare system in the world on a trajectory toward fiscal collapse. The question is not whether the system will break. The question is whether AI arrives before it does.

The variables that make prevention possible — genetic predisposition, blood biomarker trends, lifestyle data, family history, environmental factors — are all measurable today. What has been missing is the intelligence to connect them into a coherent, individual-level prediction. That intelligence now exists. It is called artificial intelligence, and it is arriving at exactly the moment the old model is failing.

The inversion of care.

What AI enables is a fundamental inversion of the care model. Instead of waiting for you to develop Type 2 diabetes, an AI-powered health system can analyze your biomarker trends over years and identify — long before any symptom appears — that you are on a trajectory toward metabolic disease. The intervention is a conversation, a lifestyle adjustment, a dietary change. Not an ICU admission.

The shift that is coming is not incremental. It is structural. And the people I met while making this episode — scientists, CEOs, economists, artists — are each, in their own way, building different pieces of the same future. A future where the machine that is supposed to keep you well actually does.

When I walked into the Quris-AI lab, Isaac Bentwich did what great scientists always do: he showed me the thing instead of telling me about it. In the center of the room was a robotic system that most people would walk past without comprehension. What it was doing was conducting thousands of drug tests — simultaneously, automatically — on what Bentwich calls miniaturized human organs. Tiny liver constructs. Tiny brain tissue. Grown from actual human blood samples.

The architecture is deceptively simple in concept, breathtaking in execution. Take hundreds of known drugs — compounds that have already been through clinical systems, some of which passed, some of which failed. Run them through these miniaturized human organs. Collect the data using nanosensors, microscopy and robotics. Feed that data into an AI. And train the AI to recognize the signatures of safety and toxicity at the molecular level.

The result is an AI that has, in Bentwich’s framing, seen hundreds of safe drugs and hundreds of unsafe ones. When a new drug candidate arrives, the AI can look at its molecular signature and say, with a precision that no animal model can match: this is likely to be safe in humans — or not.

A barrier built from your own blood.

The blood-brain barrier — the selective membrane within the brain’s blood vessels that permits only certain molecules to enter — has historically made neurological drug development one of the hardest problems in medicine. Alzheimer’s drugs fail at catastrophic rates. Bentwich’s system allows scientists to test precisely how a given molecule interacts with a patient-derived blood-brain barrier. Not the average barrier. Your barrier. Built from your blood.

The commercial case is irrefutable. Failed clinical trials cost the pharmaceutical industry an estimated $53 billion annually. Quris-AI’s platform, if it performs at scale, does not just save time — it saves the industry’s most expensive resource: a decade of human testing on a molecule that was never going to work in humans. The $28 million seed round the company raised in 2022 was, at the time, a record for an Israeli biotech seed. That is not a coincidence.

What Bentwich is building is not just a better drug testing platform. He is building the infrastructure for personalized medicine. For a world in which the question is not whether a drug is safe — but whether it is safe for you. That is the shift. That is what makes this important.

There is an uncomfortable truth embedded in the pharmaceutical industry’s business model, and Michel Vounatsos is one of the few people in the world positioned to speak to it from the inside. Drug companies do not just develop medicines. They develop intellectual property. They control patents. They set prices. And they have built, over decades, a commercial architecture that is extraordinarily effective at generating returns for shareholders — and deeply imperfect at delivering accessible treatment to patients.

When one in four Americans cannot afford to fill a prescription, the system is not malfunctioning. The system is functioning exactly as designed. The pricing power of pharmaceutical companies is not incidental to the model. It is the model.

Can AI break the monopoly?

The question that matters — the one I wanted to explore in this episode — is whether artificial intelligence changes the economics of drug development enough to break open the monopoly structure. If AI can compress the timeline from discovery to approval by five to ten years, and cut the cost of development by half, the calculus changes. The justification for charging $80,000 a year for a neurological treatment becomes harder to sustain.

Vounatsos, to his credit, does not pretend that the industry he led is without fault. He understands the system with the intimate knowledge of someone who spent three decades inside it. And he sees — clearly — the role that AI will play in reshaping it. Faster development cycles. More personalized treatment matching. Better prediction of clinical outcomes before billion-dollar trials are launched.

The next decade of pharmaceutical innovation will be shaped by how quickly the industry can metabolize AI into its core development process. Companies that move early — that rebuild their discovery pipelines around AI-driven clinical prediction, that adopt organ-on-chip validation as standard practice — will have a decisive advantage. Those that protect legacy processes will find that advantage increasingly eroded.

When Daniela Rus talks about the future of robots, she does not talk about science fiction. She talks about your kitchen. Your car. Your medical appointment. She imagines a world in which robots are as common as smartphones — not as autonomous agents replacing human agency, but as partners extending it.

The distinction matters enormously, and it is one that most public discourse about AI and robotics gets completely wrong. The popular narrative — that robots are coming to take your job — assumes a zero-sum relationship between human capability and machine capability. Rus’s research tells a different story: the most powerful deployment of robotics is not replacement. It is augmentation.

The Second Machine Age.

The concept that Rus and economist Erik Brynjolfsson — two very different thinkers working from very different disciplines — have each converged on is what they call the Second Machine Age. The first machine age mechanized physical labor: the steam engine, the assembly line, the industrial robot. The second mechanizes cognitive labor: analysis, decision-making, pattern recognition, language.

For the organizations and leaders I work with, this is the insight that reframes everything. The machines that are arriving in your workplace are not going to lift boxes or weld steel. They are going to read contracts, analyze market data, draft proposals, review medical images. They are going to do the work that, until now, only a university-educated human could do.

What Rus is building at CSAIL goes well beyond industrial robotics. Her team is developing soft robots — machines made from materials that can deform, adapt, and interact with the human body in ways that rigid machines cannot. The Mini Origami Surgeon she describes in our conversation: a surgical tool you swallow, that deploys inside the body and can be remotely controlled. Surgery without incision. Medicine without trauma.

The future Rus describes is not dystopian. It is not a world of mass unemployment and idle humans. It is a world in which routine tasks — the cognitive equivalent of lifting boxes — are taken off the human plate, and people are freed to do the things that machines are genuinely bad at: empathy, creativity, moral reasoning, genuine human connection.

Here is the question that has been nagging at every boardroom conversation about AI for the past three years: we have deployed the tools. We have done the training. We have announced the AI strategy. So why hasn’t the productivity curve moved?

Erik Brynjolfsson has a name for this. He calls it the productivity paradox. And his research — conducted across decades, first at MIT and now at Stanford — offers both the uncomfortable diagnosis and the genuinely hopeful prognosis.

The 1920s told us this would happen.

The diagnosis: organizations adopt new technologies while preserving old structures. They layer AI onto broken processes. They give individual employees better tools without redesigning the system those employees operate within. And so the technology’s potential is captured at the individual level — this person is faster, that team is more efficient — but never propagates to the organizational level where GDP-moving productivity gains live.

The historical pattern is consistent. Electricity was invented in the 1880s. It did not show up meaningfully in productivity statistics until the 1920s. Why? Because factories were initially designed as if they still ran on steam power — one central drive shaft, workers clustered around it. It took a generation of engineers to redesign factories around the specific affordances of electricity: distributed power, flexible layout, individual machine motors. The technology was not enough. The redesign was the thing.

The prognosis — and this is the part that matters for the decisions you are making right now — is that the AI dividend is not gone. It is delayed. The organizations that will capture it are the ones doing what Brynjolfsson calls co-invention: not just deploying AI tools, but simultaneously redesigning the organizational structures, workflows, and management practices that surround those tools.

This is a leadership challenge more than a technology challenge. The technology already exists. What is scarce is the organizational courage to redesign around it — to change hiring, to change incentive structures, to change the metrics by which success is measured.

Franz Schubert died at 31. He left behind one of the most extraordinary bodies of work in the history of Western music — and one of its most famous mysteries. His Symphony No. 8 in B minor, composed in 1822, contains only two movements. What he intended for the third and fourth, no one knows. The symphony sat unfinished for decades before it was even performed. It became iconic precisely because of what it was missing.

In 2019, Huawei commissioned Lucas Cantor to do what 200 years of musicologists had declined to attempt: finish it. The tool he was given was AI — specifically, Huawei’s Mate 20 Pro smartphone, fed with MIDI files of Schubert’s known works, trained to generate melodic material in Schubert’s voice.

The form without the ghost.

What Cantor discovered in the process of working with an AI collaborator was not what he expected. He had spent most of his career as what he calls a self-described Luddite — skeptical of technology’s intrusion into the creative process. The experience of working with AI shifted him fundamentally. Not because the AI was brilliant. But because of what it revealed about what human creativity actually is.

The AI generated melodies. It produced harmonic patterns that were consistent with Schubert’s known stylistic tendencies. It was, in the most literal sense, capable of producing music. And yet. The thing it could not do — the thing no AI can do — is have had Schubert’s life. His illness. His relationships. His particular 19th-century Viennese consciousness. The music AI generates is technically correct and experientially hollow. It has the form without the ghost.

The full symphony premiered at Cadogan Hall in London to standing ovations. It has since been performed in Barcelona, Beijing and Mexico City. Audiences receive it as a curiosity, a technical achievement, a conversation starter. What they are responding to, Cantor believes, is not the AI’s contribution. It is his — his choices, his arrangements, his channeling of the AI’s suggestions through a human consciousness that knew Schubert, loved Schubert, and cared about getting it right.

The lesson for every creative industry — music, writing, film, design — is not that AI will replace human creativity. It is that AI will raise the stakes of human creativity. When a machine can produce the technically competent, the differentiator becomes the irreducibly human: the biography, the wound, the specific perspective that no training dataset can replicate.

I have spent years arguing that the most important question about AI is not how powerful can it get? but who does it serve? The default answer to that question — the answer embedded in most of the AI investment, most of the talent, most of the compute — is: it serves whoever can pay for it.

Milind Tambe has spent his entire career building a different answer. And the work is as concrete and as extraordinary as anything I have encountered while making The Futurist.

From LAX to Uganda.

The 2007 phone call that changed Tambe’s career came from the chief of security at Los Angeles International Airport. The problem was specific: limited checkpoints and canine patrols, multiple road entries, and a very real terrorism threat. How do you deploy limited resources to maximize security against an adversary who is watching your patterns?

Tambe’s answer was ARMOR: a game-theoretic AI system that generated randomized patrol schedules that were optimal against a strategic adversary. The system deployed at LAX in 2007. It then expanded to the US Coast Guard, the Federal Air Marshals Service, and the TSA. It was the first operational deployment of computational game theory for security. Tambe and his students built it.

But the work that moves me — the work that I think represents what AI can become when we have enough courage to point it at the right problems — is what Tambe has done in public health and wildlife conservation. His PAWS system (Protection Assistant for Wildlife Security) uses AI-driven patrol planning to protect endangered animals from poaching in Uganda, Cambodia, and other at-risk territories. His HIV prevention work used social-network algorithms to deliver life-saving information to homeless youth in Los Angeles — reducing HIV risk behaviors with measurable, peer-reviewed statistical significance.

The productivity paradox that Erik Brynjolfsson describes — the gap between AI’s theoretical potential and its measurable economic impact — does not exist in Tambe’s work. That is because Tambe’s work is not optimizing a process that already works. It is solving problems that no other tool has ever been able to address at scale.

The documentary work came first. Blockchain City — featuring Imogen Heap — was among the first serious documentary treatments of blockchain technology as a civic and urban phenomenon. The Digital Kingdom examined the UAE’s technology transformation with access that most Western journalists could not obtain. Bitcoin Dilemma explored the economic and social implications of cryptocurrency before the mainstream conversation had even begun.

The Futurist on Amazon Prime Video is the culmination of that body of work. Available in 25+ countries across 45+ OTT platforms, it brings the interview-driven intelligence format that Khan has been developing for two decades to a global streaming audience. Season 2’s 13-episode slate — including The Futurist: Water, featuring Richard Gere — represents the most ambitious production to date.

The book. The frameworks. The record.

The book that defines this moment in Khan’s career is UNDISRUPTED (Wiley, 2025) — a USA Today and Publishers Weekly bestseller that presents a seven-pillar framework for organizational transformation: Engagement, Learning, People, Collaboration, Innovation, Execution and Culture. It has been adopted by executive teams at Fortune 500 companies as a planning resource. It won the Bronze Medal at the North American Book Awards (2025, Business Leadership) and was shortlisted for the Thinkers50 Future Readiness Award.

The proprietary frameworks Khan has developed over 25 years — AI-IQ™, the Future Readiness Score (FRS™), the AI Readiness Index (AIRS™) and the Decision Velocity Framework (DVF™) — represent a body of diagnostic IP benchmarked across 2,000+ organizations in 40+ industries. When he walks into a boardroom or onto a conference stage, he is not offering speculation. He is offering pattern recognition across a dataset that very few individuals on the planet have access to.

He has been recognized as a Global Top 25 Futurist. A Thinkers50 Distinguished Honoree. Named to Playboy Germany’s Menschen 2026 list alongside Sylvester Stallone and Gavin Newsom. Featured by Fast Company Turkey as a leading global voice on AI strategy. He has appeared on CNN, BBC, Bloomberg, Fox, and ABC Australia, and has delivered three TEDx talks.

The personal dimension of Khan’s work is inseparable from the professional. He is based in Toronto with his family. His faith is the foundation from which he approaches everything — not as a constraint on inquiry, but as its grounding. The questions he asks of the world’s most powerful technologists come from a place of genuine curiosity, genuine care, and a genuine belief that the future is not something that happens to us. It is something we build.