Meta is using AI to build and launch new standalone apps faster, and Mark Zuckerberg says several more consumer products are already on the way. The interesting part isn’t what Meta has released yet, but what happens when a company with billions of users can suddenly experiment at startup speed.
During Meta’s second-quarter earnings call, Zuckerberg said AI is accelerating product development across the company, making it easier for teams to turn ideas into working apps.
Meta has already launched Instagram Instants for sharing disappearing photos, Forum for Facebook Groups and Seller for frequent Facebook Marketplace merchants.
The company has also tested an AI-generated bedtime-story experience and released a gaming app built with help from AI coding tools.
This is a fairly dramatic shift for Meta, which hasn’t exactly been great at creating successful standalone apps from scratch.
The company has tried this routine before through Facebook Creative Labs and its experimental NPE Team, producing a small parade of apps with names like Slingshot, Rooms, Paper, Tuned and BARS.
Most arrived quietly, lived briefly and eventually disappeared into the same digital closet where old Facebook features go to think about their choices.
AI could change the math by making those experiments faster and cheaper.
Instead of spending years developing one heavily polished product, Meta can build smaller apps, release them to limited audiences and quickly see whether anyone actually cares.
Most will probably fail, but failure becomes considerably less painful when the company can move from idea to launch without assembling a miniature civilization of engineers around every concept.
Meta also has something most startups would happily trade several limbs for: distribution. Its family of apps now reaches about 3.6 billion people daily, which means a new Meta app doesn’t have to wander into the App Store and politely introduce itself.
Facebook, Instagram and WhatsApp can place it directly in front of an enormous existing audience.
Threads is the clearest example of how powerful that combination can be. The platform has reached 500 million monthly active users, helped by its connection to Instagram and Meta’s recommendation systems.
Meta says large language models are improving those recommendations by understanding what posts are about, identifying trends and helping engineers test ranking changes.
Every post appearing in Instagram Feed and Reels is now processed by an LLM that analyzes its topic and tone.
That doesn’t guarantee Meta’s next wave of apps will be good. Faster development can produce sharper ideas, but it can also produce a truly impressive quantity of stuff nobody requested.
The difference is that Meta can now test those ideas without betting the entire company cafeteria on each one.
The bigger takeaway is that AI may be turning major tech companies into permanent app incubators. Meta isn’t merely adding chatbots to Facebook and Instagram.
It’s using AI to shorten the distance between someone saying, “What if we built this?” and millions of people receiving a notification about it.
Somewhere inside Meta, another standalone app is probably being assembled right now.
Whether it becomes the next Threads or the next forgotten icon hiding in your purchase history is the fun part.
Meta has launched or tested several standalone apps and experiences, including Instagram Instants, Forum for Facebook Groups, Seller for Facebook Marketplace merchants and an AI-generated bedtime-story tool.
Meta says AI coding tools are helping its teams move from an idea to a working product more quickly. That allows the company to test smaller apps with real users without spending years developing each one.
Meta can promote new products across Facebook, Instagram, WhatsApp and Threads, giving them immediate access to billions of existing users. That built-in distribution makes it easier for Meta to test and grow new apps than a typical startup.
|
