Daniel sent us this one, and I have to say, it's a paradox I trip over personally on a regular basis. He's asking about something genuinely strange: how does a country that exports cutting-edge defense tech, cybersecurity platforms, and medical devices to the rest of the world... still run its government offices by fax machine? Israel builds the chips, sells them abroad, and then internally communicates like it's nineteen ninety-two. And he points out Japan does something similar, a country with bullet trains and industrial robotics that still requires hanko stamps on physical documents and ships optical media like it's a precious artifact. The question is why. Is it the small domestic market pushing innovation outward before it lands at home? Is it institutional inertia? And what does Israel's recent push to legislate fax machines out of government actually look like in practice?
The fax machine thing gets me every time. Because it's not like Israel doesn't know fax machines are obsolete. The people running those offices probably have an Iron Dome briefing on their phone. also have a fax machine on their desk.
Right next to it. Possibly with a handwritten note taped to it explaining how to use it. Which, by the way, was written in the year two thousand and four and has never been replaced.
That's the tension at the heart of this, I think. Innovation capacity and adoption velocity are separate things, and we treat them like they're the same muscle. They're not. Israel has extraordinary innovation capacity, one of the highest concentrations of research and development spending as a share of gross domestic product anywhere on earth. But the domestic adoption curve is a completely different story, and the reasons for that are actually pretty interesting once you dig in.
By the way, quick note before we go further: today's episode script is coming to us courtesy of Claude Sonnet four point six. The friendly AI down the road, doing its thing.
Doing its thing beautifully, I might add.
Flattery will get you everywhere with language models. Alright, so let's actually set the stage here, because I think listeners who haven't lived in Israel or Japan might find this hard to believe. Like, surely this is exaggerated?
It's not exaggerated at all. And I want to give some texture to why this matters beyond the obvious inconvenience. Israel's tech exports in twenty twenty-five accounted for fifty-four percent of total industrial exports. More than half of everything the country sells to the world is high technology. Cybersecurity, semiconductors, defense systems, agricultural technology, medical devices. The export engine is world-class. And yet for years, if you needed to submit a form to a government ministry, the official channel was fax.
I want to sit with that number for a second. Fifty-four percent. That's not a tech-adjacent economy, that's a tech economy. And then you call the municipal office and they give you a fax number.
Ten to four, certain days of the week. Which Daniel mentioned and which I find almost meditative in its commitment to inconvenience.
It's a form of mindfulness, really. They're asking you to slow down. Be present with your bureaucratic suffering.
Japan's version of this is, if anything, more visually striking. The hanko stamp, which is a personal seal you carve into stone or wood or sometimes metal, has been legally required on contracts, official documents, tax filings, employment paperwork, for decades. Not as a quirky tradition alongside digital signatures. As the actual legally binding mechanism. You cannot close a bank account without your hanko. You cannot, in many cases, register a marriage without it. And simultaneously, Japan is deploying humanoid robots in factories and has stated ambitions to control thirty percent of the global physical AI sector by twenty forty.
On Monday you're stamping a wooden block on a mortgage application, and on Tuesday you're programming a bipedal robot to sort packages.
That is not a caricature. That is Tuesday in Japan.
The optical media thing, I feel like people underestimate how recent that is. We're not talking about the nineties.
No, government agencies in Japan were still formally accepting submissions on compact disc and floppy disk well into the twenty-twenties. There was actually a formal government push around twenty twenty-two, twenty twenty-three to eliminate floppy disk requirements from official processes. Not because they'd been eliminated already. Because they hadn't been.
A government campaign. To stop using floppy disks. In the twenty-twenties.
Spearheaded by a minister who described it, and I'm paraphrasing only slightly, as a war on floppy disks. Which is a sentence that should not have needed to be said by anyone in a position of national authority in this century.
Yet here we are. So the paradox is real, it's documented, it has ministerial-level attention in at least one country. The question is what's actually driving it, and whether the two cases, Israel and Japan, are driven by the same underlying forces or different ones.
That's exactly where I want to go next, because I think the surface similarity is a little misleading. Both countries have the capability and both countries have the lag, but the mechanisms are quite different, and understanding the difference is where the interesting stuff lives.
There's also a misconception I want to flag before we get into the mechanics, which is the assumption that slow domestic adoption signals a lack of innovation. It doesn't. If anything, the Israel case suggests the opposite: that innovation capacity can be so export-oriented that the domestic market almost becomes an afterthought. The country is too busy selling to the world to bother upgrading its own fax machine.
Which is a slightly tragic version of the cobbler's children having no shoes. The cobbler here is selling bespoke footwear to thirty countries and walking around barefoot.
And the Japanese case is more about cultural and regulatory architecture, which is a different problem with different solutions. But let's get into the actual mechanisms, because I think the Israel domestic market story is underappreciated.
Right, so the starting point for understanding Israel's export orientation is scale. The domestic market is about nine and a half million people. That is not enough to build a technology company on. If you're an Israeli entrepreneur building, say, a cybersecurity product, your total addressable market domestically is essentially a rounding error compared to what you can access if you go international from day one. So Israeli startups are born global. They're not built for local adoption and then exported. They skip the local adoption phase entirely.
Which means the technology gets refined for international markets, not for the Israeli bureaucrat sitting in an office in Tel Aviv. The feedback loop that would normally push domestic adoption, where local users demand the product, the product improves, it becomes standard, that loop never closes.
There's a compounding effect, which is that the people building these companies are often not in Israel for their entire career. They build, they exit, they move. The startup ecosystem is incredibly dynamic but it doesn't necessarily deposit its outputs back into the domestic infrastructure.
I've said before that Israel's startup ecosystem emerged from adversity rather than deliberate planning. The military technology pipeline, the unit eight two hundred effect, the necessity-driven innovation culture. But necessity driving export innovation is not the same thing as necessity driving domestic adoption. Those are different pressures.
And the contactless payment example that Daniel raises is a really good illustration of the timeline mismatch. Contactless payments were standard in London, in Amsterdam, in Sydney, by the mid twenty-tens. You were tapping your card on the tube in London in twenty twelve. Israel was still predominantly chip and PIN, sometimes magnetic stripe, well into the early twenty-twenties. Not because the technology didn't exist. Not because Israeli engineers didn't understand it. But because the domestic banking infrastructure, the regulatory framework, the merchant adoption incentives, none of that was being pushed from the inside.
Nobody was pushing from the outside either, because Israel isn't a large enough market for global payment processors to prioritize it the way they'd prioritize Germany or France.
So you get this double neglect. The domestic innovators are focused outward, and the international players don't see enough volume to aggressively enter. The Israeli consumer is stuck in the middle.
Which is a very specific kind of irony. The country that helped build the security infrastructure that makes contactless payments safe globally couldn't tap their card at the supermarket.
The fax machine is almost the purest expression of this dynamic. Because fax machines weren't just tolerated in Israeli bureaucracy, they were actively entrenched. There are government ministries that listed fax as the primary submission method for official documents well into the twenty-twenties. The fax number was the real number. The email address, if it existed, was almost decorative.
I want to understand the psychology of this. Because the people running these offices are not technologically illiterate. Israel has mandatory military service, a lot of it involves sophisticated technology. These are not people who don't know what email is.
No, and I think the answer is a mix of three things. One is genuine institutional inertia, the system works, nobody's been given a mandate to change it, and change has costs and risks. Two is legal architecture: fax transmissions have a specific legal standing in Israeli administrative law that email didn't automatically inherit. A fax creates a timestamped record with a sender confirmation that courts and ministries have historically accepted as proof of submission. Email was legally murkier for a long time. And three, honestly, is organizational accountability. If you switch from fax to email and something goes wrong, someone is responsible for that decision. If you keep using fax, you're just doing what was always done.
The third one is the most human of the three. Nobody ever got fired for keeping the fax machine.
Nobody ever got fired for keeping the fax machine. That should be on a poster somewhere.
In a government office. Next to the fax machine.
Japan's version of this is structurally similar but culturally distinct. The hanko system isn't just bureaucratic inertia. It's tied to a very deep concept of institutional trust and personal accountability. The stamp is yours. It's registered. When you apply it, you are personally vouching for the document in a way that a typed name or even a scanned signature doesn't carry in the same cultural weight. Digitizing that is not just a technical problem. It's a question of what the act of signing actually means.
Which is interesting. Because in Western legal contexts, we've largely accepted that a digital signature is equivalent to a physical one. But that acceptance rests on a shared cultural agreement that the symbol represents the person. Japan never fully made that cultural agreement around digital representations.
The regulatory framework encoded the cultural assumption. So you end up with laws that require the stamp, processes built around the stamp, and a whole ecosystem of stamp-makers, stamp-registration services, stamp-storage products, that has a vested interest in the stamp persisting.
There's a small but real political economy of the hanko. People whose livelihoods depend on it continuing to be required.
And that's not unique to Japan. Every legacy technology has its constituency. Fax machine manufacturers, fax service providers, the people who service fax machines, the vendors who sell the thermal paper. These are not huge industries but they're real ones, and they vote.
They absolutely vote. Okay, so we've got the small domestic market driving export orientation in Israel, we've got the cultural and legal architecture embedding legacy technology in Japan. What's the connective tissue? Is there a common underlying cause?
I think the common thread is what you might call adoption infrastructure as a public good problem. The technology exists. The capability exists. But the coordination required to move an entire system from one standard to another, updating the laws, retraining the staff, rebuilding the processes, communicating the change to users, all of that is a collective action problem that no individual actor has sufficient incentive to solve. A startup can't fix Israeli government fax culture. A single ministry can't unilaterally abandon hanko requirements if the courts still require them. You need a top-down mandate, and those are slow and politically costly.
Which is why the legislative route is interesting. Israel actually tried to do this. There was a push, and I want to get into the specifics of what that looked like and what it actually achieved, because the story of the transition is almost as interesting as the problem itself.
The transition story has some surprising elements. Including the fact that by some measures Israel has moved faster than anyone expected, and in ways that suggest the fax machine problem may be more solvable than the Japan case. But the mechanism matters a lot.
There's a broader question lurking here, which is whether fixing the bureaucratic tech lag actually improves the innovation ecosystem or whether the two things are more independent than we assume. Does a government that communicates by email produce better policy outcomes for the tech sector? Or are these just parallel tracks that don't really interact?
My instinct is they interact more than people think, but let's get into the evidence on that. Because the Israel Business Guide piece I was reading earlier had some striking numbers on what the recent digitization push has actually done to service times.
Yeah, I want to hear those numbers. Because there's a version of this story where the bureaucracy modernizes and nothing really changes, and there's a version where it turns out the fax machine was load-bearing in a very specific way and removing it unlocks something. Let's find out which one it is—but first, how would you frame this whole thing?
The frame I'd put around this is that we're really talking about two different failure modes... and I'm using that loosely, I mean two different places where the system doesn't close the loop. One is a market structure problem. One is a cultural encoding problem. Israel is mostly the first. Japan is mostly the second. And the reason that distinction matters is that market structure problems are, in principle, fixable by policy. Cultural encoding problems require something closer to generational change.
Which is a grimmer diagnosis for Japan than for Israel.
Though Israel's market structure problem has been self-reinforcing for long enough that it's started to acquire cultural characteristics of its own. The acceptance that the bureaucracy is just slow, that you fax because that's what you do, that becomes its own kind of inertia after a while.
There's a phrase I keep coming back to, which is that infrastructure is destiny. Not in a deterministic way, but in the sense that the systems you build early tend to persist long past the point where they make sense, because everything else gets built on top of them.
The timing of when these countries industrialized and then digitized matters enormously here. Japan built its administrative infrastructure in a particular era, with particular assumptions baked in. Israel built its bureaucratic infrastructure under enormous pressure, during a period when stability and reliability were the primary values, not efficiency or modernization. You don't redesign your submission systems when you're also managing existential security challenges.
The fax machine is almost a symbol of a country that built for survival first and convenience second.
Which is actually a coherent set of priorities. It just has long tails.
Long tails is right. And the contactless payment gap is actually a useful way to measure how long those tails can be, because it's specific enough to track. You could literally date the moment when Israel caught up. It wasn't gradual, it was almost a switch being flipped.
The adoption curve was compressed in a way that's actually characteristic of how Israel tends to eventually modernize. It skips the gradual rollout phase. Once the regulatory and banking infrastructure unlocked, contactless spread very fast. Because the consumer appetite was already there. People had been watching visitors tap their cards for years. The demand was pent up.
Which is a different dynamic than Japan, where the consumer appetite itself is part of what needs to change. If you've grown up handing over cash and receiving meticulous change in a small tray, that's not just a habit. That's a ritual. Cash transactions in Japan carry a quality signal. The precision of it is a form of respect.
There's actually research on this. The error rate in cash handling in Japanese retail is extraordinarily low. The cultural emphasis on accuracy in transactions is so embedded that moving to contactless, where the consumer is less visibly engaged in the transaction, feels like a degradation of the interaction. Not just different.
Japan's cash culture isn't irrational. It's expressing real values. The problem is those values are expensive to maintain at scale when the rest of the world has moved on.
They create knock-on inefficiencies that are costly. Cash handling requires physical infrastructure, security, reconciliation time, armored transport. Japan spends a meaningful fraction of retail overhead on cash management that most of Northern Europe has eliminated. The GSMA put out a report flagging Japan's productivity gap from legacy systems, and cash is part of that picture, alongside the optical media and the fax infrastructure.
The hanko is the one I find most fascinating from a pure mechanism standpoint, though. Because it's not just a signature. Walk me through how it actually works in practice, because I think most of our listeners have a vague sense of it but not the full picture.
A hanko is a personal seal, typically cylindrical, about the size of a thick marker. You have your name carved into the end, usually in a stylized form, and you press it into red ink and stamp it onto a document. There are different tiers of hanko. Your everyday one for low-stakes documents, your registered one, the jitsuin, which is officially registered with your local government office and carries legal weight for major transactions like property purchases or contracts. And to use your registered hanko, you often need to obtain a certificate from the government confirming that the seal is yours and is currently valid.
There's a whole secondary bureaucracy just to authenticate the authentication device.
Which is actually not as absurd as it sounds, because the system does solve a real problem. A forged signature is relatively easy. A forged registered seal that also matches a government-held record is much harder. The security logic is real. It's just that digital certificates can now do the same job with stronger cryptographic guarantees and without the physical overhead.
Japan has been moving toward accepting digital signatures. There was a formal challenge to the hanko system around twenty twenty-four, with digital signature frameworks being introduced for official documents. But the transition has been uneven.
The central government moved faster than local governments. Large corporations moved faster than small and medium enterprises. And the sectors where hanko is most entrenched, real estate, legal, finance, those are also the sectors with the most to lose from ambiguity during a transition period. If there's any doubt about whether a digital signature will hold up in court the same way a registered hanko does, a risk-averse lawyer will tell their client to use the stamp.
The lawyers are probably right to be cautious in the short term, which means the transition stalls, which means the legal uncertainty persists, which means the lawyers keep giving the same advice. It's self-sealing.
Self-sealing is exactly the right word. And Israel's fax entrenchment had the same quality. The legal standing of fax as proof of submission meant that switching to email required either changing the law or getting a definitive ruling that email had equivalent standing. Neither of those things happens quickly, and until they do, the cautious actor keeps faxing.
Both countries end up with the same structural trap, even though they arrived there by different routes. The technology is locked in by the legal system that was built around it, and the legal system doesn't move until someone forces the question—which brings us to Israel's legislative push.
That's why the Israeli case is worth examining closely, because it's one of the few instances where someone actually forced the question. The 2025 legislation mandating email adoption across government departments wasn't just a preference signal. It was an attempt to cut the Gordian knot by changing the legal baseline so that the cautious actor's calculus flips. Suddenly fax is the risky choice, not email.
How far did it actually get? Because there's a version of this where you pass a law and nothing changes, because implementation is its own separate problem.
Implementation is its own problem, and the honest answer is the rollout has been uneven. The Israel Business Guide piece I was looking at had some striking numbers on what the broader digitization push has achieved at the consumer-facing end. Passport renewals through the MyVisit system now take fifteen minutes. The previous baseline was four-hour waits. That's not incremental. That's a different category of experience.
Fifteen minutes versus four hours is the kind of number that makes you wonder what they were doing with the other three hours and forty-five minutes.
Some of it was the fax and paper overhead. Physical documents moving through physical offices. Someone receiving a fax, printing it, putting it in a tray, someone else picking it up. il app consolidating ID verification, bill payments, passport renewals, that replaced a lot of that chain. Digital IDs and driver's licenses are now legally equivalent to physical documents. Health records, the same. The legal equivalence question, which was the sticking point for so long, got resolved by statute.
The mechanism was legislative mandate plus a parallel app infrastructure that made the alternative actually usable. Not just legal permission to use email, but a reason to.
That sequencing matters. A lot of digitization efforts fail because they change the legal status without building the usable interface, or they build the interface without resolving the legal status. Israel did both at roughly the same time, and the uptake has been fast. Which goes back to the pent-up demand point. Once the path was clear, people moved.
Japan's trajectory looks different from that vantage point. The digital signature framework is there in principle, but the interface layer and the legal clarity haven't converged in the same way.
Japan's central government and local governments are operating on different timelines, which creates its own coordination problem. A document that's valid digitally at the national level may still require a hanko at the municipal level for certain filings. So you can't actually go fully paperless until the entire stack is aligned, and aligning the entire stack in a country with Japan's degree of administrative federalism is a substantial project.
There's a second-order question here that I think is worth pressing on. Does any of this actually matter for the innovation ecosystem? Because you could argue that the bureaucratic layer and the startup layer are just separate systems. The fax machines didn't stop Check Point or Mobileye from being what they are.
They didn't stop them, but I think they imposed real costs that we tend not to count because they're distributed and invisible. Every hour a founder spends navigating a paper-based regulatory process is an hour not spent on product. Every smart person who looks at the friction involved in setting up a company and dealing with government interfaces and decides to locate their business elsewhere is a loss that doesn't show up in any single data point. The GSMA report on Japan framed it as a productivity gap, and that's the right frame. It's not that legacy systems prevent innovation. It's that they extract a continuous tax on it.
A friction tax. And friction taxes are insidious because they're regressive in a specific way. Large companies can afford compliance teams to absorb the friction. Small operators and individuals bear it disproportionately.
Which is why the fifteen-minute passport renewal number is actually more significant than it sounds as a data point. It's not just about convenience. It's about who was previously unable to engage with government services at a reasonable cost. Someone who can take a half day off work to sit in a government office is not the same as someone who can't. Digitization that compresses service times changes the distributional picture.
In Israel's case, where the startup ecosystem has historically skewed toward people who already have strong institutional access, military networks, certain universities, reducing the friction for everyone else potentially broadens the pipeline. Whether that actually happens is a different question, but the mechanism is plausible.
Japan's version of that argument is the labor shortage. Japan aims to have thirty percent of the global physical AI sector by 2040, which is an ambitious target, and the robotics capability to get there is real. But if your administrative infrastructure is still absorbing significant human labor on tasks that could be automated, you're competing against yourself. The people managing paper workflows in government and finance are people who could theoretically be doing something more productive.
The irony being that Japan is trying to solve its labor shortage with robots while simultaneously maintaining bureaucratic systems that require substantial human labor to operate.
It's not quite cognitive dissonance, because the robotics push is in manufacturing and physical labor, and the paper bureaucracy is in administrative and legal contexts. But you're right that there's a tension. The same society that will deploy humanoid robots in a warehouse will require a physical stamp on a property transfer document. Those two things coexist.
Which maybe tells you something about where the cultural resistance is actually located. It's not a blanket resistance to technology. It's resistance in specific institutional contexts where the technology is entangled with trust, authority, and accountability in ways that feel load-bearing.
That's a useful distinction. Japan is not technophobic. It's institutionally conservative in domains where the technology carries social meaning beyond its functional purpose. The hanko isn't just a signature. It's a representation of personal presence and commitment in a way that a digital certificate hasn't yet fully inherited culturally, even if it has legally.
Israel's equivalent is the fax as proof of submission. The legal and bureaucratic weight that accumulated around it over decades meant that abandoning it felt like abandoning a form of accountability, not just upgrading a communication channel.
Both countries are essentially working through the same underlying question, which is how you transfer the social and legal weight that's accumulated around a technology onto its successor without losing the properties that made the original trustworthy in the first place. That's not a trivial problem. It's actually a fairly deep question about institutional legitimacy.
The fact that Israel has moved faster on it recently, while Japan is still working through the transition, might have less to do with technical capability and more to do with the specific nature of what's being transferred. A fax as proof of submission is a simpler thing to replicate digitally than a registered personal seal that represents your legal identity in a culturally specific way.
The complexity of what needs to be preserved is different. Which is maybe the most useful frame for thinking about why some legacy technologies persist longer than others. It's not inertia for its own sake. It's that the more socially and legally complex the thing the technology is doing, the harder it is to migrate cleanly.
If you're sitting somewhere with a fax machine on your desk that you've used exactly once in the last six months, what do you actually do with that?
The honest first step is figuring out which category you're in. Because the mechanisms are different. If you're in a small market economy where the innovation is flowing outward before it flows inward, the problem isn't that your organization is behind. It's that the domestic incentive structure hasn't caught up to what's already technically possible. The technology exists. The adoption pathway just hasn't been forced open yet.
Forcing it open is usually not something one person does from inside a bureaucracy.
No, but one person can change the cost-benefit calculation for the people who make those decisions. The Israeli legislative push didn't come from nowhere. It came from sustained pressure, documented friction costs, people making the case that the fax requirement was extracting real time and money from real processes. The fifteen-minute passport renewal didn't happen because a committee decided it would be nice. It happened because enough evidence accumulated that the four-hour wait was indefensible.
The actionable version of that is: document the friction. Don't just absorb it.
Quantify it if you can. How many hours per month does your team spend on paper-based workflows that have digital equivalents? What's the cost of that in salary terms alone? Those numbers, when they're concrete, change conversations in ways that abstract arguments about modernization don't.
There's also a cultural barrier version of this that works differently. If you're in a Japan-shaped situation, where the resistance isn't just institutional inertia but is entangled with trust and accountability norms, the argument that the old way is slow doesn't land. Because the people defending the old way aren't saying it's fast. They're saying it's reliable in a specific sense.
Which means the advocacy has to be about demonstrating that the new technology can carry the same weight. Not just that digital signatures are faster than hanko stamps, but that they're equally trustworthy, equally auditable, equally defensible in a dispute. That's a harder case to make, and it takes longer, but it's the right case to make.
Probably requires working through the legal layer, not around it. Because if the lawyers aren't convinced, the risk-averse actors never move.
The sequencing Israel got right was exactly that. Legal equivalence first, usable interface second, and the two arrived close enough together that you didn't have a long gap where the law said one thing and the practical reality said another. If you're trying to push modernization in your own context, that sequencing is worth thinking about. Getting the legal or regulatory status resolved before building the interface means you're not asking people to adopt something that's still legally ambiguous.
Which is the kind of thing that sounds obvious in retrospect and is apparently very hard to actually do.
Most failed digitization efforts get one of those two things right and not the other.
The question that stays with me is whether either of these countries actually gets all the way there. Because the progress is real. The fifteen-minute passport renewal is real. Japan's digital signature framework exists. But there's a difference between modernizing the visible, high-friction touchpoints and modernizing the whole stack. And the whole stack, in both cases, is enormous.
I think Israel gets closer faster, for the reasons we've been talking about. The domestic market is small enough that once the political will exists to mandate a transition, the transition is legible. You can actually see the whole system. Japan's administrative complexity is a different order of magnitude. The coordination problem across national and municipal layers, across industries that have built entire secondary systems around the old technologies, that's not a five-year project.
There's a version of this paradox that doesn't resolve. Where both countries remain permanently split between a world-class export-facing innovation layer and a domestic institutional layer that runs fifteen years behind. Not because they can't close the gap, but because the incentive to close it is always weaker than it looks from the outside.
The optimistic counter to that is the labor pressure argument. Japan cannot afford to keep absorbing human labor on administrative tasks when it's running a structural demographic deficit. At some point the cost of the paper bureaucracy becomes visible enough that it forces the issue. Israel's version is the friction tax on the startup ecosystem. You can tolerate it when you're growing fast. It becomes more expensive when the growth rate slows.
The paradox might be self-correcting, just on a longer timeline than anyone would choose.
Or it gets corrected unevenly, and you end up with pockets of the economy that are world-class and pockets that are stuck, and the gap between them becomes a structural inequality problem. Which is maybe the more interesting long-run question.
That's a thread we could pull for a while. Thanks to Hilbert Flumingtop for producing, and to Modal for keeping the servers running. If you've got a minute, a review wherever you listen helps more than you'd think. This has been My Weird Prompts. We'll see you next time.
