2025 Who Predicted this

by Alexander Rose, December 31st 2024

‘Tis the season to look over past predictions to see how we got here, and this year being a nice quarter century, inspired me to look back at more than just the last year. At 52 years old, this means taking a look at the events that shaped the world in the second half of my life. What got me curious was how many of these events were not only predictable, but were in fact actually predicted. And most importantly, what can we learn from it all?

What is a bit surprising is that nearly all of the major world events I looked at turned out to be predicted in some way, often by many people or organizations. But while correct predictions do happen all the time, picking that needle out of the haystack of punditry is the true art. The other element that emerges is that even when all the right people are saying all the right things, it is often very difficult to change our behavior or use that information in a useful way.

Remember at the beginning of the pandemic? It was easy to see from history that coming out of lockdown would be long, arduous, and require the development of a vaccine. But few of us thought when it all began that we would not be returning to a similar life for several years.

As any futurist or prognosticator will tell you — when you predict bad outcomes and they don’t come to pass, people forget it easily, but if you predict a good outcome that does not come to pass, it can follow you forever. In addition for a prediction to be taken seriously, it has to seem reasonable to be believed, but the actual future is under no such constraint.

And this leaves us with one of the trickier issues with making any prediction — timing. One of the projects I helped usher into the world with The Long Now Foundation is Long Bets. After helping adjudicate many of the predictions and bets in that arena, I was struck by how many were correct predictions, but were ‘wrong’ because they misjudged the time scale. Events like earthquakes, tsunamis, asteroid impacts and even pandemics are 100% predictable events. But knowing when they will hit, and at what magnitude, continues to elude us. As my friend and futurist Paul Saffo is famous for saying: “Never mistake a clear view, for a short distance.”

So with all that in mind here are some of the events from our first 25 years into this new century and millennium that struck me:

• 2000 The Y2K non-event and the dot-com crash
• 2001 9/11 and the subsequent 2 decades long war on terror
• 2008 Financial Crisis
• 2009 Bitcoin and Blockchain
• 2010 Air travel disruption after the Eyjafjallajökull eruption
• 2020 Global Pandemic
• 2022 Expansionist Russia and the Ukrainian Conflict
• 2024 Computers pass the Turing Test
• 2000–2025 “Fake News” and the loss of agreed upon truths
• 2025 Technology predictions vs reality and other references

The Y2K non issue: One of the few people I remember from the time that correctly predicted that Y2K would not be a big deal was computer scientist Danny Hillis. He correctly identified that we actually don’t trust technology that much, as it often fails for mundane reasons like power failures, or bad software updates. (I have also since learned that there were many others who predicted along similar lines like futurist Paul Saffo in this 1999 article.) It has been pointed out that one of the reason’s Y2K was a non-event was precisely because some countries did take it seriously, and fixed many critical systems. This is no doubt true, but if we look at the rest of the world who did vastly less to prepare for this (Russia being the largest example), there were very few documented issues, and none of them very serious. Interestingly something that haunted computers even worse, and was never addressed at scale, was the Y2K leap year exception issue. But I think it is worth noting that time-frame plays a key role here. If we had widespread technology failures now, the effects would be felt much more acutely than in the year 2000. As we saw with the 2024 Crowdstrike software update failure, the tendrils of technology have crept much further into our daily lives. I happened to have been in an airport on that day, and saw over half of the flights get canceled, not because there was a problem with the planes or weather, but because half of the airlines used the same software system that failed due to a poorly tested update.

Economic crashes: Living in the Bay Area in the 90s it seemed unfathomable that so many of the new meteoric tech companies could fail so fast. Though, as we look back on it, it is obvious that few had a real business model. But there were people who did predict that crash, like Robert Shiller who published Irrational Exuberance just before it happened. (He then also went on to correctly predict the 2008 housing crash.) The people who I knew in large-scale real estate were also amazed that the housing market lasted as long as it did, and were not surprised by the 2008 crash. So I would categorize both of these as predicted events, it’s just that few people comprehended the scale, or adjusted their lives accordingly. I remember asking Jeff Bezos how Amazon survived that time. He told me how when the stock was quadrupling every quarter, he would tell his employees, “you are not all 4x smarter, and when the stock goes down, you will not be 4x dumber. Focus on the customers and the business, not the hype.”

The attacks of 9/11: Enough has been written about this one already. I think the most notable part about this aside from the level of tragedy, is that this was an entirely predictable and predicted event. It was just a failure of imagination on the part of the systems by which we prepare and defend the United States. It was so far out of the realm of possibility, that even when the terrorists were in flight school, and showed no interest in learning how to land, that it went completely unreported. It is a horrible, yet perfect example of how difficult it is to believe a prediction that does not seem plausible, even when all the information is telling you the implausible thing is the one most likely to happen.

Bitcoin and the Blockchain which were launched around 2009 have had an extremely volatile path to reaching $100k/coin and making the impact that they have. There are certainly many who bet on this, and have done extremely well because of it. Tim Draper is one of the more prominent people to publicly predict BitCoin value, and below you can see Peter Diamandis correctly predicting the blockchain was going to be important as far back as 2015. So again this was predicted, but hard to say if you could have called it reliably predictable.

2010 Air travel disruption after the Eyjafjallajökull eruption: I included this one because I see it as one of the truly unpredicted events of the last 25 years. While the eruptions of volcanoes like this one have some predictability, the disruption to air travel for over a month was a complete blind side. The particularly abrasive cloud of dust that became airborne over a large swath of Northern Europe meant that planes could not fly without damaging their turbine engines resulting in the largest interruption of air travel since World War II. To my knowledge, this was a truly unpredicted Black Swan event.

The Global Pandemic: This is another subject that plenty has been written about. It was entirely predicted, and predictable. I include several references below ranging from the Department of National Security, Rockefeller Foundation, or even the hauntingly accurate blockbuster movie Contagion. That movie was made a decade before the pandemic by the Skoll Global Threats Fund, and advised by epidemiologists like Larry Brilliant to help the world imagine and understand what might happen in a real pandemic. Even with all that, people were largely blind sided by the events as they unfolded. And even the people who predicted it to a high degree of accuracy, did not predict elements like the toilet paper shortage, what it would do to a generation of children, or how fast we would be able to create and deploy multiple vaccines.

Expansionist Russia and a drawn out Ukranian war: While the possibility of Russia invading the Ukraine was on many analysts bingo cards, the scale and length of this development certainly surprised me, and eluded most long-term predictions. Intelligence agencies did see the military build up at the border in the months leading up to it, but I have not seen any citable predictions about the conflict playing out the way it has. It seems few could imagine how well Zelenskyy, a former reality TV star, would lead the Ukrainian defense, or how long Putin would sustain the effort. On those fronts I would classify this as one of the other true Black Swan events of the last quarter century.

Computers pass the Turing Test: I would argue that the current versions of AI such as Chat GPT basically pass what is called the Turing Test. In his 1950 landmark paper pioneering computer scientist Alan Turing predicted that:

“[by about the year 2000] it will be possible to programme computers… [that] play the imitation game so well that an average interrogator will not have more than 70% chance of making the right identification [of human vs machine] after five minutes of questioning.”

While Turing was off by about 25 years, and some may argue we are not there yet, I do think artificial intelligence largely passes the test at this point. In the right circumstances, most people can be fooled into thinking computer generated images, writing, and even art are from an actual person. Just ask chat GPT to write you a poem about an obscure subject, it is astonishing (or take a look at how it summarized this article). And while Alan Turing’s timing was off, many have predicted we would achieve this around now, one of the more notable being technologist Ray Kurzweil. And while the Turing test may have now been passed, I think we can all agree that we have not achieved human-like general machine intelligence. This is the more surprising part about the development of AI to me. If you showed anyone from the year 2000 our current version of it, they would feel as though we have achieved full machine sentience. But really what we keep doing is moving the goalpost down the road, and maybe human-like sentience is not what we should be looking for. Maybe we should be looking for a new kind of consciousness that we have yet to define.

“Fake News” and the loss of agreed upon truth: I have to say that this particular one is the event that is both the most surprising to me, and the most disturbing. I remember in the 1990s as the digital and communications revolution began, almost everyone who was creating our future had the sense that all we had to do was connect everyone, and dis-intermediate the big companies from news, and everything would be better. Very few people predicted that we would end up in such a fractioned information landscape, where people’s news is fed to them by algorithms that show them stories they either already agree with, or that they so violently disagree with, that they won’t believe it. There were voices that predicted some of this for sure, but I see it as another a huge failure of imagination that it would be at this scale. Most of all, I am at a loss as to how we recover from this. There is no longer a way to have everyone in a single conversation with mutally agreed upon and verified facts.

If I can leave you with a prediction from me about the next 25 years, it is that this will be our biggest challenge of this next quarter century

Alexander Rose, December 31, 2024

---

Past Predictions and some analysis:

One of the better set of technology predictions I found for 2025 is this one by Peter Diamandis “The World in 2025” published in 2015 in the Singularity Hub. I should say that I know Peter a little bit, and I appreciate that he is one of the rare people brave enough to give positive predictions about the future. In many ways he lives the adage that the best way to predict the future is to build it yourself. That said, several of these predictions were not accurate, but I discuss them here, not to show how wrong he was, but to illustrate how difficult it is to predict technological developments. I think he did a great job overall, especially calling out things like a “Jarvis like AI” and the importance of the Blockchain way back in 2015.

1. A computer under $1000 that can run at 10,000 trillion (10¹⁶) cycles per second, roughly equivalent to a human brain: When he predicted this in 2010 we had just about reached 1 billion cycles/sec (1×10⁹), or 1Ghz processor speeds in home PCs, and we are now barely hitting 6 billion cycles or 6Ghz (6 × 10⁹). However in terms of operations per second (OPS) in 2010 we were at about 70 billion OPS (7 × 10¹⁰) with the Intel i7 processor, and the recently released $250 Nvidia Jetson can do 70–250 trillion operations per second (7–25 × 10¹³). So there have been large gains, but Moore’s law has largely flattened in terms of cycle rates for home PCs, and is just starting to take off again with the desire to add AI to everything. So this prediction was pretty far off as even if we go with operations per second. We have however achieved 2 × 10¹⁸ operations per second with the fastest super computers (and over 1000 qubits in quantum computing), but how that is equal to the human brain is more ambiguous.
   [UPDATE Jan 7th 2025: NVIDIA announced a $3000 consumer computer at CES with a PetaFLOP processor (10¹⁵ floating point operations/sec) making a huge leap in the operations per second front.]
2. A trillion sensor economy of 100 billion connection devices: Currently it is estimated we have about 20 Billion connected devices that would equal maybe a ¼ trillion sensors. So off by about a factor of 4. He also cited CISCO claiming the Internet of Things (IoT) market would reach $25 Trillion, but it is currently at about $1 Trillion with current estimates of it reaching $4T by 2032.
3. Perfect Knowledge “you’ll be able to know anything you want, anytime, anywhere, and query that data for answers and insights”: While I think the spirit of this has been achieved with the types of answers we get from Chat GPT, I am not sure that I would call what we have ‘perfect knowledge’. We are at a low point in terms of conflicting and dis information, conspiracy theories and unchecked facts being used widely in everything from policy decisions to vaccines and food choices.
4. 8 Billion Hyper Connected people: It looks like as 2024 comes to a close we are currently at about 6 billion people connected to the internet. This has roughly doubled the 3 billion people connected in 2010, but still a couple billion short. I do think we are not far off from connecting many more people with the advent of increasingly less expensive satellite based broadband.
5. Disruption of Healthcare: While many of the precursors Diamandis points out have come to pass such as widespread wearable sensors, robotic surgeons, AI diagnostics, and big new players like Amazon entering the market. The costs of healthcare have continued to rise at similar rates, and one of the only large changes we saw in the US happened with the ACA in 2010 where people with pre-existing conditions could not be turned down. So while I would say that the technology in this space has moved considerably, the inertia of the industry has yet to be truly disrupted.
6. Augmented Virtual Reality: “The screen as we know it — on your phone, your computer and your TV — will disappear and be replaced by eyewear” Again big leaps were made in this space, and it is true that there are many augmented reality options out there, it has in no way replaced the TV or computer screen. I think it was greatly under-appreciated how silly people feel wearing these glasses, and waving their hands around in space like a crazy person.
7. Jarvis like AI: This prediction could not be more true. He predicted that the early versions of Siri would be eclipsed with something more akin to Jarvis from the Iron Man movies. However he called out very large super-computing efforts like IBM Watson and DeepMind as the driving force for this, butit was breakthroughs by smaller companies like Open AI that really changed this landscape.
8. The Blockchain and Bitcoin: He very correctly predicted in 2015 that it was the beginning of something that would have a large effect.

Some of the places I looked for predictions include Long Bets and the US Dept of Natl Intelligence 2008 report on what they thought the world would look like in 2025. They did a pretty good job overall. Most notably, their prediction of a possible global pandemic. Luckily their prediction of how bad it could have been were not realized. They predicted up to tens of millions of deaths in the US, we had roughly 1.2m directly attributed to COVID.