The game-changing impact of AI on the photo industry

Zeev Farbman

Photo: Lightricks, Ricky Rachaman

Any silver linings? Well, if there are any, he doesn’t offer much in this interview. Yes, over time general purpose software could conceivably be replaced by software that’s created on the fly that would allow a specific person to do a particular task at a given moment. But that’s still way out there, albeit perhaps more doable than we might have thought, according to Farbman. And for the time being there’s still work to be done to keep the current user base happy.

That’s it.

As someone who has been knee-deep emerged in previous major industry transformations (from film to electronic publishing, prepress and printing; from digital to smartphone cameras; and from PC photo software to smartphone photo apps), I can only appreciate Farbman’s candor when rubbing in the level of disruption AI is bringing to the photo industry.

But let me add some context, perspectives and recommendations.

I’ll zoom in on:

• The lessons we’ve learned from that other not so long-ago tech revolution that upended our photo industry: mobile photography

• Understanding how today’s AI tech revolution is different from when going mobile was all the rage

• One often useful model to develop a business strategy that leverages new opportunities triggered by AI 

1. Remember the mobile photography revolution? 

One message Farbman accentuates is that AI dramatically lowers the barriers to entry. Anyone with a creative idea and some programming resources can license or open-source generative AI or large language models and crank out innovative AI-powered apps. 

In that regard, today’s AI revolution is not that different from the low-barrier to entry days when the mobile app ecosystem was established 15 years ago. In those days innovative mobile apps were able to upend desktop software, allowing nimble startups to gain market share by addressing new use cases or catering to existing use cases in more effective ways. 

These mobile-first startups put the onus on incumbents to rearchitect their tech so that they too could become part of this exciting new mobile world. The alternative being: not develop mobile solutions and simply assume this new computing platform was never going to be serious enough to matter to their customers. 

So what happened in this new low-threshold mobile world? Innovative startups came to market with solutions that effectively catered to new (mostly younger) users, or to new use cases (say, social photo sharing) or existing use cases addressed in new ways (such as photobook creation on the go). What’s more, they were able to offer their apps for free or close to free (remember the days that a $2.99 perpetual license photo app was considered to be expensive)? And quite a few succeeded, were even acquired or became unicorns: they not only include Lightricks, but also Instagram, Snapchat, TikTok, Picsart, VSCO, and Perfect Corp. Many others thrived in profitable segments and became permanent fixtures in the photo ecosystem.

And finally, whether barely on time or simply too late, the incumbents did eventually come to the conclusion that they couldn’t stay on the sidelines and had to be part of this mobile world as well: take desktop photo editing software vendors like Adobe, various photo print product sellers, stock photography juggernauts, and even digital camera vendors (who finally mastered frictionless photo transfers with smartphones and the cloud).

But in most cases, coming late to the party came with a penalty when the newcomers had already carved out their stakes, resulting in permanent market loss for the incumbents, or – less obvious but often even more substantial – cause the incumbents to miss out on opportunities to expand their business when the mobile transformation was still in its infancy and the landgrab was still going on.

2. How is today’s AI revolution different?

Fast forward to today’s AI revolution. We’re of course still super early and the dust has certainly not settled yet, but what strikes me as dramatically different now is how fast photo industry incumbents have come to the realization that they need to incorporate AI tech to develop new features, new products, or simply to improve their existing solutions. 

While for many years after smartphone photography had already taken root, the chorus of naysayers in our photo industry was still loud and clear, most of today’s photo industry incumbents have already gotten the message – you’d better be part of this and do so fast, otherwise you’ll be a relic in no time. Take Adobe (Firefly), Google Photos (Magic Editor), Canva (Magic Edit), and visual-focused social media companies like Meta (Imagine with Meta AI, see also below), Snapchat, and TikTok. Add to that equation the stock photography players (see our recent Visual 1st The Future of Stock Photography panel), photo print-focused developers (Masterpiece AI, myprint.ai, PlanetArt), AI-powered photo enhancement solutions (Viesus, Let’s Enhance, EyeQ) and photo curation (Zenfolio, Mobius Labs, and the speakers in our AI Image Curation 2.0 panel at Visual 1st) – the list goes on…

What made today’s existing industry players react so fast, while so many incumbents were so slow to act when smartphone photography first took off? 

For one, many of today’s players might actually still remember a thing or two from the smartphone platform shift that caused them to lose a big chunk of their business or to miss out on sizable opportunities. “This can’t happen to us again!” might be their thought.

Another factor (but again, it is still early to see it this is a long-term phenomenon), is that today’s AI-powered solutions can be offered within existing image creation, enhancing, curation, or printing products. Today’s incumbents don’t have to build new solutions for a completely new technology and sales & marketing platform, as was the case with the smartphone app platform. 

In addition, as mentioned earlier, todays’ barriers to adopt AI tech are quite low. This doesn’t only provide opportunities for newcomers, but also for incumbents to relatively quickly adapt by incorporating AI into their products. 

Offering AI tech integrated inside existing products also gives incumbents a leg up to compete with emerging startups, who face the challenge of needing to attract users from scratch. In addition, the combo of generative AI with traditional photo editing features appears to get more user traction at this stage than “pure generative AI” apps that solely offer text-to-image or photo-to-image features. Examples galore: Adobe Firefly, Google Photos Magic Editor, Canva Magic Edit, Picsart Ignite, etc.

3. How to develop a successful biz strategy that leverages AI tech innovation?

As a photo product vendor, what’s the best way to develop a strategy that leverages the opportunities of AI and defends yourself from AI-powered competitors?

While there are many exercises possible to develop a forward-looking business strategy, in my executive workshops I’ve found the good-old Ansoff matrix to be a very useful brainstorming model to come to data-driven decisions on the optimal strategies for growing your business:

The four quadrants to brainstorm on are:

• Market Penetration: focus on existing markets with more or less existing products (that might have more AI tech under the hood)
• Could you leverage AI platforms to make your products better, cheaper, more efficient, or could you product variations that add value for specific customers and warrant a higher price?

• Market Development: focus on new markets with more or less existing products (that might have more AI tech under the hood) 
• Could you leverage AI platforms to offer versions of your existing products to new markets that weren’t feasible to enter before AI?

• Product Development: focus on existing markets with new AI-powered products
• Could you leverage AI platforms to develop significantly different products to sell to your current users?

• Diversification: focus on new markets with new AI-powered products)
• Could you leverage AI platforms to expand into more lucrative markets with new products?

Obviously, each of these strategies have their pros and cons, the balance of which will differ depending on the markets you’re in, your customers’ use cases, your current or anticipated competitors, your team’s tech, your marketing, sales and operations resources, your funding, your long-term goals – and a lot more.

But the bottom line is: The AI transformations in the tech industry at large, and our photo industry in particular, force us all to strategically think outside the box as to how to move forward with our businesses. Business as usual is not an option.

And a few more things...

Meta. Imagine with Meta AI. How could I have imagined it would come to this? I’m really excited about a Meta product: Meta’s new and shiny Imagine with Meta AI AI text-to-image creator. Images are high quality high-res. And fast to create. And free. And not part of Facebook or Instagram (you just need a Meta account). And will in the coming weeks have invisible watermarking according to Meta “for increased transparency and traceability.” (There’s already a visible watermark.) 

Imagine with Meta AI is based on Meta’s EMU AI image-generation model, which uses a process called "quality-tuning." Unlike traditional text-to-image models trained with large numbers of image-text pairs, Emu focuses on "aesthetic alignment" after pre-training, using a set of relatively small but visually appealing images. But no surprise, at Emu's heart is a massive pre-training dataset of 1.1 billion text-image pairs pulled from Facebook and Instagram. 

di support. Hello pixolo. True to its roots of lower case company naming tradition, di support has re-emerged from bankruptcy protection and rebranded as pixolo. pixolo has new shareholders, including di support founder Ralph Naruhn, who sold the majority of his shares in 2019 and is now once again the largest shareholder. 

pixolo will no longer be an operator of instant print systems in the retail sector. Having the coolest looking kiosk in the market apparently does not guarantee success. 

Instead, the strategy forward is to grow its biometric ID (passport photo) B-2-B systems and develop e-commerce solutions and individualized on-demand photo and lifestyle products, including B2C photo apps, under the pixolo brand.

Click. Authenticating your smartphone photos. With camera vendors, including Leica, Nikon, Sony, and Canon now on board to support the Coalition for Content Provenance and Authenticity (C2PA) industry guidelines so that photos taken by their cameras can have encrypted metadata that proves their origin – for instance, showing they weren’t generated by AI – the question has been: how about smartphone photos?  

If you can’t wait for the major smartphone vendors to come on board and implement C2PA: there is Click, a smartphone camera app that not only supports C2PA, but lets you digitally sign your photos by simply swiping them. 

Nice touch: it stores the provenance data, plus location, time and specific smartphone camera info, as unalterable data on the blockchain – which is where I always thought C2PA data should reside – that can be verified on Click’s public website or its app.

Fujifilm. Instax – the fad that wasn’t a fad. Heralded by observers as a short-lived youngsters obsession with retro, Instax has proven the observers to be wrong. Initially released in 1998, Instax cameras have not only proved their staying power, but their sales keep on growing.

But really, growing by this much??? Instax currently amounts to 65% of the sales in Fuji's entire Consumer Imaging division, which includes all of Fujifilm’s cameras including Instax. Fuji's goal is to increase this by $1.04 billion (15% YoY growth) by end of the current fiscal year (March 2024). Who said that smartphones killed the camera market? Or wiped out any reason to print your photos?

Google. Gemini coming to you, wherever you are. It took Google a while to respond to GPT-4 but the company is on a roll now. Gemini, Google’s new and shiny multimodal LLM is making its way into Bard, the Pixel 8 Pro , and for now developers and enterprise customers: Google Generative AI Studio and Vertex AI. 

But that’s just the beginning. According to CEO Sundar Pichai, Gemini will eventually be integrated in Google’s search engine, its ad products, the Chrome browser, and more.

Midjourney. AI image creation for the rest of us. Hurray, for rest of us: soon, Discord will not be the only place where you can have access to Midjourney. One of the very best AI image generators will soon do what all its rivals have been doing for a long time: offering its AI image generator on a dedicated web page. Midjourney has started beta testing this.

Happy Holidays and the best wishes for 2024!

Hans Hartman