Some people love AI and think it’s the greatest thing ever, and some people hate AI and think it’s the single worst thing to ever happen to humanity. I’m more in the middle, but leaning slightly closer to “single worst thing” than “greatest thing ever”, as I have a somewhat pessimistic outlook on it because I see a lot of negative aspects. For that reason, I avoid AI as much as practical, but I have found it to be helpful occasionally, and use it from time-to-time. For example, last year I had a last-minute opportunity to rent an RV from Miami to Boston for a steal-of-a-deal, but I only had a few days to plan the entire trip, so I utilized AI to create the route and decide the stops. It did a good job overall—better than I could have done on such short notice, but probably not as good if I had had months to plan it—but more importantly, it took a big task off of my plate that I just didn’t have enough time for. As someone once told me, AI is better than a complete novice, but not as competent as an expert. Every so often I find a good use for it.

One negative aspect of AI has been its effect on website traffic. I noticed a reduction in visitors to Fuji X Weekly beginning in the early spring, and it has only increased since. Initially I chalked it up to that it’s been a historically slow year for Fujifilm X-series cameras; had Fujifilm released a couple of models, perhaps traffic would be up and not down. Also, I’ve written a few less articles in 2026 than I had at this same time in previous years, so maybe that plays a role. Still, those using the Fuji X Weekly App are up notably. My Instagram account has been steadily growing, despite my general incompetence at all things social media. I couldn’t quite put a finger on why web traffic was down until I talked to several other people with websites (not all Fujifilm or even photography related) who noticed something similar. That sent me down a rabbit hole, and I found the answer: AI is reducing traffic to most websites across the internet.

I guess the thing to know is that (prior to 2026) about 70% of visitors to Fuji X Weekly (and this goes back many years) are either regulars who frequently visit this site or those who follow a link from somewhere across the web (Facebook, Reddit, etc.), while the other 30% come via web searches (mostly Google). They type into the search bar Kodachrome and Google shows them my article History & Poetry of Kodachrome. Or they search history of Instax and Google brings up my article Fuji’s Instant Film: The Immensely Interesting Story of Instax. It’s from this group—the 30% that come from web searches—that traffic is down. So far this year, only about 25% of visitors came from a search engine, and last month it was closer to 20%. I still rank very high (at or near the top) of many Fujifilm-related searches, though, so what gives?

The most obvious reason is Google Gemini, which summarizes answers to web searches. Many people now just read the AI summary. There’s no need to click the article, and in fact only about 1% of those who read the Gemini answer will click the resource that was used to create the answer. In other words, it’s copying other people’s work and basically taking the credit. This actually isn’t good for Google, as (according to Yahoo Finance) it has had quite the negative effect on their valuation due to lower ad revenue. People use ChatGPT and other AI programs as alternatives to Google. About 75% of ChatGPT customers use it as an alternative to Google at least sometimes, and 25% do so regularly; however, less than 3% will click on links provided by ChatGPT, and instead find the AI answer sufficient and trustworthy (despite it getting things wrong sometimes). Almost everyone with a website is affected by this, but especially those who rely on Google searches.

Another problem is AI vibe coding. Basically, anyone can have AI create any website or app (or camera firmware) that they want, despite perhaps no personal coding skills or experience with whatever it is they are making it for. But, AI cannot create anything new, it can only imitate what already exists. If someone put in hours and hours of work to create something, as long as AI can analyze it, it can copy it, or make some new version of it. Just like Gemini does for web searches, AI vibe coding does something similar for apps and such. That’s why there has been an explosion of new apps (of all types) over the last few months. Someone even made a facsimile of Adobe’s software suite. Nothing that is AI vibe coded is original, it’s all copycats and riding coattails.

When it comes to AI vibe coded camera firmware, that’s super risky. PTP (which is what the Fuji X Weekly App uses for the App-to-Camera Recipe Transfer feature) is simple and stable, but there are minuscule risks with it, and it still needs to be handled carefully by developers to mitigate those tiny risks. When we’re talking about firmware, the risks are so much greater, probably 1,000,000 times greater or more! Add AI vibe coding into it, and it’s probably much more risky than that. It’s not so much if someone’s camera will become a permanent paperweight, it’s a matter of when and how many, and whether it will be yours. Personally, I wouldn’t load any unofficial third-party firmware into any of my cameras, the risk is much too large. Even if it seems to be working fine, you have no idea if some rarely used setting combination will be what wrecks it, or if by using it the camera will wear out much quicker.

Whenever AI “borrows” something from you, it devalues your work, at least a little. When Google’s Gemini summarizes the content of your webpage, people won’t bother to click on it. When a person uses AI to vibe code an app that’s similar to someone else’s, that devalues the original app. When someone uses AI to copy Adobe’s software, that encourages people to cancel their Adobe subscription. AI companies are getting billions of dollars from this. Someone who can’t create anything original—can only coattail off of someone else who has actually created something with blood, sweat and tear—gets a little peace of the pie, perhaps undeservedly. The creative people who actually create things, whether art or tools or something else, are the ones screwed over by this. In a sense, your work is “taxed” by AI, simply for existing somewhere on the internet. The incentive to be creative and original has certainly been reduced. Why put in tons and tons of work just to have AI more or less copy it in a matter of seconds?

As I’ve thought about this, I believe the natural response by creative people who actually create things will be to remove their creations from AI. Not just avoid using AI in the making of it, but keep it away from AI completely. The only way to do that is to keep it off the internet altogether. Perhaps some platform will emerge that has an AI blocker to prevent AI from having access to whatever is on their platform; otherwise, more and more I think artists and inventors will simply not have their creations appear anywhere online. The more disconnected from AI, the more valuable it will be. This seems inevitable. Those who don’t will suffer a similar fate as “microstock” photography, which has basically died, or at least is greatly struggling.

I don’t have any sort of conclusion, and I wish I could point to some great silver lining. I hope that governments will do better at placing guardrails and holding AI companies accountable—I don’t have a ton of faith in that, but it is very necessary and already overdue, so maybe it will happen. Some are saying that this is the beginning of the end of humanity, and others are saying a utopia will soon be ushered in. I don’t believe either of those two scenarios are likely; they’re both wrong in my opinion (but what do I know?). What I can say is that the more “artificial” things become, the more people desire what is real. The pendulum swings. For photographers, that means making real photos, which also means printing our photos, because physical is real. That’s my closing encouragement to you: print more of your photographs. Go to real places and have real experiences, and do it with your camera. Then put those pictures on a wall or in a book.