In the AI era, Apps are becoming an overabundant commodity.

According to RevenueCat, in early 2022, approximately 2,000 new subscription Apps were launched each month. By early 2026, this number had surged to 14,700, meaning nearly 500 new products flooded the market daily.

At this rate, the businesses most ripe for disruption are the “App factories” that profit from bulk App development.

Yet, interestingly, while the entire industry is still searching for answers on AI application commercialization, these low-key “App factory owners” are already reaping substantial profits.

Take the Turkish company Codeway, for example. In just a few years, it has developed over 60 Apps, more than half of which are AI-related. One of its Apps, Learna, which teaches spoken English, generates $5-6 million in monthly revenue on the iOS platform alone.

Chinese company Hangzhou Ruiqi is no slouch either. Its product portfolio has expanded to around 36 Apps, with at least 24 centered around AI. Its most successful plant identification App, PictureThis, generates approximately $12 million in monthly revenue across both platforms.

From AI-powered foreign language tutors to plant identification, these seemingly insignificant niche demands can support monthly revenues in the millions or even tens of millions of dollars. Behind this lies an important shift in the AI application layer:

AI has made developing Apps easier, but it hasn’t made monetizing them equally simple.

As technical barriers continue to lower, the ability to identify demands, acquire users, and sustain monetization becomes increasingly critical. And this is precisely where App factories have leveraged their strengths built over the past decade.

Today, Silicon Guy discusses why AI has, paradoxically, ushered in a second golden age for App factories.

/ 01 / A New Wave of Low-Key “Factories” in the AI Application Layer

The AI application layer has produced not only star products like Harvey (legal AI), Rogo (financial AI), and Cursor (programming AI) in recent years.

It has also given rise to a group of relatively low-key (low-key) players who are quietly making money: the App factories.

These companies share a common trait: they rarely bet all their resources on a single product. Instead, they operate dozens of Apps simultaneously, covering diverse needs such as image processing, language learning, health management, and lifestyle tools.

In the past, they primarily monetized through bulk development of utility Apps. Now, in the AI era, this strategy is driving a new wave of growth.

The most typical example is the Turkish company Codeway.

Founded in 2020, Codeway had only 9 App products in early 2024. Today, its portfolio has expanded to over 60, with more than 30 being AI-related, becoming its primary product direction.

More importantly, these AI products have started generating significant revenue.

Currently, Codeway’s most powerful product is Learna, an AI-powered foreign language tutor in your pocket.

Its function is straightforward: it allows users to practice spoken English with AI anytime, anywhere.

For example, if preparing to go abroad, users can directly select scenarios like airport check-in, hotel check-in, or restaurant ordering. The AI acts as staff, and users respond verbally. The AI continues the conversation, correcting pronunciation, word choice, and grammar in real time.

To make conversations more lifelike, Codeway later replaced Learna’s voice system with ElevenLabs.

This change was directly reflected in monetization.

An A/B test disclosed by Codeway and ElevenLabs showed that after switching voice systems, Learna’s trial-to-paid conversion rate increased by 9%, and user lifetime value rose by 11%.

The product launched in September 2024. In just two years, it has accumulated over 15 million downloads.

According to Sensor Tower estimates, by September 2026, Learna’s downloads on the iOS platform alone reached approximately 2 million in the past 30 days, with store revenue between $5-6 million.

A seemingly simple AI spoken English practice tool now generates millions of dollars in monthly revenue on a single platform. But Learna is just one product line under Codeway.

Its AI photo editing App, Retake, generates around $2 million in monthly revenue, while the general-purpose assistant Chat AI brings in about $600,000. AI meeting recorder Smart Noter generates around $400,000, and AI writing tool TypeAI about $100,000.

From English learning and photo editing to meeting recording, Codeway has successfully launched a series of profitable AI products across entirely different demands.

Another Turkish company, HubX, follows a similar strategy.

Founded in 2022, HubX has developed over 40 products, with at least 15 explicitly identified as AI applications. In 2024, the company established an AI Lab to train and fine-tune models for different App scenarios.

Unlike Codeway, HubX lacks a standout superstar product and instead achieves success across multiple fronts.

For example, its Home AI is dedicated to solving interior design problems.

In the past, ordinary people wanting to visualize their renovated home often needed a designer to create renderings. With Home AI, users simply take a photo of a room, select a style like modern or minimalist, and the AI redesigns the sofa, walls, color scheme, and even the entire spatial layout.

Living rooms, kitchens, bedrooms, bathrooms, and even exterior facades and gardens can all generate separate renovation plans.

This seemingly niche demand has also proven to be a viable business.

Home AI launched in April 2024. By September 2026, it had approximately 500,000 monthly downloads on the iOS platform alone, with Sensor Tower estimating monthly revenue around $900,000.

Another product, AI Video, targets short video creation.

Upload a photo of a person, and the AI can make them dance; a static old photo can be transformed into a video. Users can also generate scenes from text, add voiceovers and lip-syncing, and export vertical content suitable for TikTok, Instagram Reels, and YouTube Shorts.

What once required filming, editing, and animation can now start with a single image. This product generates around $1 million in monthly revenue on iOS alone.

Further down the lineup, general-purpose assistant Nova, tattoo design generator Tattoo AI, avatar and image creator DaVinci, and AI photo and video tool Momo each generate tens of thousands of dollars in monthly revenue.

Today, this approach is being adopted by more App factories.

Take AIBY, founded in 2018, which now operates over 50 products. Since 2024 alone, it has added 29 new Apps, 19 of which are AI-related, accounting for 65.5%.

Its AI image and video generator ARTA generates around $300,000 in monthly revenue on iOS in 2026. Coin ID (coin recognition), HitMeal (diet tracking), and OnSkin (skin care tool) each generate tens of thousands of dollars in monthly revenue.

Hangzhou Ruiqi has taken this model of productizing niche demands even further.

Based on a rough count of publicly available products in the App Store, its portfolio has expanded to around 36, with at least 24 centered around AI.

Plants, insects, birds, fish, mushrooms, rocks, coins, vinyl records—virtually every identification demand can be turned into a standalone App. Its most successful product, PictureThis, generates approximately $12 million in monthly revenue across both platforms.

From Codeway to HubX, and then to AIBY and Hangzhou Ruiqi, AI is becoming a new growth engine for App factories, driving a fresh wave of expansion in the bulk production of Apps.

/ 02 / Why Are App Factories Reaping AI Dividends?

At this point, you’re probably wondering:

Logically, AI should make App development easier, weakening the App factories’ original advantages. So why are they the first to reap AI’s benefits?

In Silicon Guy’s view, there are three main reasons:

First, AI reduces development costs, making many previously unviable small-scale businesses profitable.

American App factory Bending Spoons disclosed a set of data: in Q1 2025, less than 10% of the company’s code PRs were written or co-written by AI; a year later, this proportion exceeded 90%, with about 70% directly completed by AI.

For App factories, this efficiency gain is further amplified.

Because they typically operate dozens of products simultaneously, code, model interfaces, subscription systems, and even advertising strategies can be reused. Combined with AI’s reduction of design, development, and maintenance costs, fewer resources are needed for a new App.

As a result, many demands previously too small to justify a project now have the potential to become viable businesses.

Take AIBY’s Rock ID, which does one thing: take a photo of a rock and tell you its type.

Peekly estimates that this App generates only about $5,000 in monthly revenue on Google Play. Similar Apps for mushroom, insect, or trading card identification may generate only a few thousand dollars per month.

In the past, this revenue might not even cover a development team’s costs. But now, as long as development, maintenance, and acquisition costs are low enough, even a few thousand dollars in monthly revenue can be profitable.

A single factory operating dozens of similar products can accumulate small gains into substantial revenue.

AI has given commercial value to demands previously ignored due to insufficient market size.

But cost reduction is just the first layer. AI’s greater value lies in commercializing demands that were previously technically unfeasible.

Take photo-based food calorie calculation.

Back in 2011, DailyBurn launched an App called Meal Snap, which estimated calorie counts from food photos.

But the technology was unreliable. TechCrunch tests found that for a small handful of cashews, the App’s calorie range spanned from 150 to 614 calories.

A few years later, the product was removed from the U.S. App Store.

Over a decade later, Cal AI revisited this concept. It uses AI to identify food, estimate portion sizes, match nutritional databases, and calculate calories along with protein, fat, and carbohydrates. According to the founder, the average accuracy of a single food scan now exceeds 90%.

The result? By March 2026, when acquired by MyFitnessPal, Cal AI had accumulated over 15 million downloads, with sales exceeding $40 million in the previous 12 months.

The same concept—photo-based calorie counting—went from technically flawed and unpopular to a business generating tens of millions in annual sales.

From AI-powered foreign language tutors to interior design and photo-to-video generation, the underlying logic is the same.

Advances in vision, voice, and large language models have expanded the problems Apps can solve. Tasks previously poorly handled by software can now deliver experiences users are willing to pay for.

Lower costs have expanded the survival space for niche demands, while model improvements have further expanded the markets App factories can enter.

But here’s the question: when anyone can develop Apps at low cost, why are these factories still the ones making money?

This brings us to the third reason: AI is amplifying the commercialization advantages App factories have already built.

RevenueCat data shows that nearly 15,000 new subscription Apps launch each month, but only 4.6% generate $10,000 in monthly revenue within two years.

Developing an App is getting easier, but actually selling it remains difficult.

For subscription-based Apps, downloads are just the first step. Whether users are willing to pay, how much acquisition costs, and how long subscriptions last determine whether a product can turn a profit.

And this is precisely where App factories excel.

Take HubX, which has developed two dedicated systems.

The first, XRay, scans millions of products on the App Store and Google Play hourly, estimating downloads, revenue, and growth trends to identify rapidly growing niche markets.

In other words, it determines what is currently profitable and what the next App should be.

The second, MarketingX, handles selling the product after development.

From the first day of advertising, it predicts how much revenue new users will generate in the future, then evaluates whether ads are worth continuing based on acquisition costs.

If acquiring a user for $1 yields only $0.50 in future revenue, advertising is cut quickly. If it yields $3 and remains profitable after other costs, launch (advertising) is scaled up.

This system has supported HubX in developing over 40 products without external financing, covering over 600 million users cumulatively, and maintaining consistent profitability.

For these companies, each App is a commercial experiment: identify demand, develop the product, test monetization, scale advertising, double down on what works, and abandon what doesn’t.

Success rates for individual products may not be high, but as long as testing costs are low enough and successful products yield high enough returns, the entire business can keep running.

And AI has made this system even more efficient.

Lower development costs mean more products can be tested for the same budget; stronger model capabilities mean more demands can be attempted; and as competition intensifies, the market insight, acquisition, and subscription operation skills App factories have accumulated become even more critical.

Thus, the rise of App factories in the AI era reveals an interesting shift:

When developing Apps becomes less valuable, the abilities that determine whether an App can make money become more valuable.

The product development and commercialization systems Codeway, HubX, and others have built over the past decade are now unlocking even greater value in the AI era.

By Yuan Yuan