How Super Apps & Platforms Are Turning AI On Its Head

I usually present the US and EU view of AI, so in this article, I will cover a third perspective, Asia. It’s one of the few markets that Big Tech companies face significant competition. We assume that the biggest, best model wins, which is why we focus on the US so much.

However, the US AI labs have faced strong competition from Asian companies that focus on serving local markets with homogenous needs. In the US, startups and Big Tech focus on delivering products to a US consumer and assume that will generalize globally with small changes. The reality is much different.

The Rise Of Super Apps & Platforms

Markets like South Korea and Japan are small, so startups scaling in those markets grow quickly but hit the customer ceiling just as fast. They have been forced to come up with a novel business model to keep scaling the super app and platform. Rather than growing through new customers, they grow by offering new services and products on their platform.

Big tech grows by offering a single service to as many customers as possible. Asian tech companies scale by offering a smaller customer base as many services as possible. This necessity drove the super app and platform business model. Asian startups scale to support so many customer needs that they take over entire domains of those customers’ lives.

Most Asian countries have homogenous populations with very similar needs. What works in the US or EU rarely fits those needs, so Big Tech struggles to get a foothold. Super apps and platforms niche down and customize products to support a single country. Learning to customize products for one market means it will be easier to expand into the next market by repeating the localization process.

Super apps work for two reasons. The super app’s platform makes deploying new features and localizations less expensive and speeds up delivery. The super app’s customer base is more likely to use new services on the platform than to go elsewhere. Customer acquisition costs and churn are much lower for super apps than single service platforms.

The Advantages Of Super Apps & Platforms

Advantages can be grouped into three main categories: acceleration, cost reduction, and strategic positioning.

Acceleration

Faster Time-to-Market: AI platforms like Amazon Prime and TikTok create new opportunities for monetization and accelerate the development of novel features and products. Platforms also streamline the transition from prototypes to market-ready products.

Rapid Maturity Progression: Robust platforms can quickly advance a company's data and AI maturity, enabling them to move from basic data gathering to advanced applications like knowledge graphs, multi-agent systems, and simulations in a shorter timeframe.

Improved Development Cycles: AI platforms enhance internal processes, leading to faster model training, product development, and testing. This creates a flywheel effect where each iteration accelerates subsequent ones.

Cost Reduction

Lower Data Gathering and Training Costs: Data and AI platforms reduce the cost per unit of data gathered and the cost per unit of training data used. This is achieved through efficient data curation and streamlined model training processes.

Optimized Resource Utilization: Platforms facilitate the auditing and elimination of redundant data, unused datasets, and inefficient models, leading to significant cost savings.

Strategic Positioning

Enhanced Competitiveness: Companies leveraging AI platforms can gain a competitive edge by developing and deploying innovative AI solutions faster than their rivals.

Unique Data as a Moat: Platforms facilitate the creation and curation of proprietary datasets, establishing a unique advantage that is difficult for competitors to replicate.

Agility and Adaptability: AI-native platforms enable businesses to quickly adapt to changing market conditions and customer needs by supporting dynamic workflows and intent-driven interactions.

A Potential AI Super App

A South Korean startup, Wrtn, launched an AI super platform two years ago and is following the ‘aggregator’ strategy. It aggregates multiple models and services into a single platform. It has localized versions of ChatGPT, Perplexity, X’s Radar, and Character AI. The startup has already scaled in the South Korean consumer markets, reaching 5 million monthly active users in October, and plans to expand to Japan next.

The platform is an enabler for each expansion and localization. Wrtn creates a new localization layer, and everything else stays the same. Wrtn doesn’t build all the LLMs itself. The company partners with Anthropic, OpenAI, and many other AI labs…so it’s just a wrapper, right?

In most cases, being a wrapper around those services isn’t enough to develop a viable business model. Wrtn has developed a viable business model through localizations and its pricing model. Super apps can bundle services into a single subscription, but Wrtn has an even stickier approach.

The super app is almost entirely ad-supported. Users only pay for a few character customizations. In the Big Tech model, getting access to similar functionality requires multiple paid subscriptions: ChatGPT, Character, and X Premium. The business model is what turns AI on its head. We can buy everything separately or get everything in a single app for free.

Wrtn has room to grow. Every time an AI lab releases a new product, Wrtn can integrate it into the super platform, and customers get even more value. They spend more time on the super app, and Wrtn gets to serve more ads. It’s a virtuous cycle built to thrive on new entrants into the AI market instead of being disrupted by them.

In Big Tech, each service is siloed. Wrtn’s unified platform can streamline updates and scaling of LLMs within the ecosystem, leading to faster product innovation cycles. The simplicity and free tier make the platform an obvious choice for price-sensitive customers. Customization only makes sense if it can scale across services, so single-service AI labs don’t have the right business model to enter those markets.

How Will This All Play Out?

So far, Wrtn and the AI labs are taking a partnership approach. Anthropic used the company as a case study for how its LLMs can support advanced AI platforms. Big Tech companies have tried and failed to gain traction alone in the past. Google, Amazon, Uber, and many others eventually followed the partnership paradigm.

So far, super platforms have failed to gain traction in the US and EU. This time might be different. We’re seeing cracks in consumer tolerance for multiple subscriptions in the streaming market. People have subscription fatigue and multiple AI services will likely hit the same wall.

There’s also a massive divide between power AI consumers and those who don’t use any of the services. Wrtn has gained traction with South Korean customers who haven’t used AI services in the past. The focus on entertainment and accessibility is a winning combination for driving adoption with consumers who are on the fence about AI apps. Customer-focused businesses win, and super platforms have an advantage there.

It will be interesting to watch how this round plays out.