If you just want the answer to the question, scroll down to the second bold-faced line later in the article. However, you really need to get there via some set up, so I hope you read my whole treatise.
______________________________
"It seems that we are in a period of time we reached the end of miniaturization! We can't go to smaller microchips, the strength of the currents inside the chips will cause the electricity to skip around circuits and cause problems." — Internet forum post
When you read something like the above quote on the Internet, you're seeing an uninformed opinion that's wrong. I can't state that more bluntly. Unfortunately, the Internet has this amplification effect, so once someone posts their paranoia online, that can quickly swell into first legitimate concern and eventually full acceptance.
In my talks with semiconductor experts, they all seem to believe that we've got at least a decade's worth of a reasonable (slower) form of Moore's Law (miniaturization) ahead of us (and I don't believe cameras are near current state-of-the-art, so they already have quite a bit of catching up to do). Moreover, we've been through these "Moore's Law is ending" suggestions before. The first time I heard that expressed is now almost twenty years ago.
As most of you know, I'm a technophile, and I'm optimistic that science continues to take us on a journey of discovery that generates new and better technologies in the future. Assuming, of course, that we continue to pursue and pay attention to science, which unfortunately isn't a given. I fear we're currently headed into the modern equivalent of the Middle Ages.
At the heart of statements like the above are people pondering what happens next with "cameras." I put that word in quotes because I think what we call a camera is likely to change in the coming decades. You can already see that in some forms. For instance, "cameras" in Tesla vehicles have been used in everything from court cases to art. One could claim that instead of a device that you hold in your hands and look through, Tesla "photographers" just drive their cameras to where they want to take a photo.
I get the "what's next for photography" question near constantly. Partly because I was right about predicting the transition from film to digital, as well as the transition from DSLR to mirrorless. Most of those questions are not actually about photography, though. They're gear questions, as in "I'm using a Nikon ZX; what am I going to be using in ten years, and will Nikon still be making cameras?" I don't know the answer to the first part of the question at the level of detail the person probably wants, but the answer to the second half is "yes, but maybe not in the form you're thinking of."
The interesting thing about being at the front edge of the tech business for so long is that you see in the failures the future. I'll give you a very simple personal example: back when we first imagined Go and the PenPoint operating system, we quickly stumbled on the use of tabs in the UI. Indeed, the whole PenPoint user experience revolved around moving between tabs to get access to different data. Recognize something about that statement? Does the Web browser you're currently using have tabs to move between data? Go failed, but quite a few of our ideas have had full and productive lives ;~)
Ideas that are basically correct in tech do live on. Here's another one that people have now dismissed but shouldn't: Lytro's light field capture. Like Go, Lytro burned though nine figures of capital, only to disappear. Yet I still remember quite well the excitement among the Hollywood crowd when Lytro demonstrated their Immerge system at NAB in 2016. I had originally thought that it was the three-dimensional virtual reality that was the selling point—witness today's Apple Vision Pro attempting the same thing—but it wasn't. In a cab with a Disney executive I found out the real reason they were interested, and that was the volumetric video aspect of the system: because the entire space in front of the "camera" was fully mapped in 3D, doing special effects—"replace that cardboard box with a robot"—could be easily done. Actually "easily imagined to be done."
Volumetric video is still coming. The ideas, the scholarly work, the physics, the geometry, everything about the idea still percolates, it's just not fully brewed into useful consumer products yet.
Which brings me to my answer to the question "what will cameras be like in ten years?" First, they will have multiple inputs (sensors). Second, they will integrate what we know about the world with what the sensors measure (AI). Third, user control interaction will (mostly) disappear and be replaced by user processing interaction. My marketing slogan? "The world is your camera."
This is a tricky bit for the Japanese camera companies. I'd say that virtually all of the "innovation" in cameras we've seen out of Tokyo has played off of things others discovered and proved before them. Digital image sensors didn't get invented in Japan. More like Bell Labs. But once out of the labs, the notion of converting photons to electrons and recording that became a pretty clear goal, particularly since the way we had been doing that revolved mostly around chemicals emanating from Rochester, New York. That allowed the Japanese to disrupt an entire industry.
Disruption has always been part of business cycles. The trick is to find the next technology that allows you to create that disruption.
You might think that the AI built into your camera today—virtually all focus systems use one now—means that the Japanese are on top of things. They aren't. The type of AI you see in your camera today is mostly the machine learning derived AI we were talking about in Silicon Valley ten to twenty years ago. The current focus system AI is just past the problem that Netflix encountered back in 2006 when they were trying to figure out "movie recommendations." Personally, I can actually see what the focus systems were trained on, and what they weren't. They're just babies with limited abilities at the moment.
My sense is that the Japanese are trailing the US and China in what we think about in current AI engines and uses. That doesn't mean they always will, however. It's a potential point of disruption to the Japanese camera industry, and I'd be looking closely at DJI and what they're doing in that aspect, because that's where the next generation of "cameras" may really come from.
Of course, smartphones are also in the mix, as well. Too many people get caught up in the limitations of small image sensors when they discuss smartphone "cameras." Smartphones are actually towards the forefront of the #2 and #3 of what I said a camera would be like in ten years (five paragraphs above).
Thus, my final thought is this: the camera industry is going to be disrupted again. Perhaps in five years, maybe in ten years, but absolutely within twenty years. That disruption will come from China or the US (as long as things continue to progress the way they are currently). Japan is on a clock.
-----------------
Bonus: Oh, I just have to write it. My fingers are way ahead of me here. 21st century cameras will be communicative. There, I said it (again; I first wrote that in 2007). That means that they don't exist in a vacuum by themselves. They integrate into all your other technology. Naturally. Conveniently. In real time. The Japanese like to think that they're making systems. Nope. Their product is (or should be) part of a bigger system they don't control.
-----------------
Some of you want a higher level of detail than the three points I made. I can actually do that. However, I still consult with tech companies and startups from time to time, so I'll reserve my detailed thoughts for those discussions.
