The Truth About Digital Banking: The Fragile Ecosystem Behind 1.46 Billion Users

There are 368 active digital banks, two-thirds of which do not have formal banking licenses.

Written by: Francesco Andreoli, Head of Developer Relations at Consensys

Compiled by: Chopper, Foresight News

Everyone is focused on rounds of financing, yet no one pays attention to the waves of disappearances. Thus, I have documented both.

Six months ago, I began tracking digital banks (Neobanks). I realized that no one could clearly state how many are still operational in the market. Not even analysts selling reports for $4,000, venture capitalists funding such projects, or competing founders could provide accurate answers.

As of July 2026, there are 368 verified digital banks that are still operational. I continuously track each institution based on publicly available data from neobankbeat.

However, the number that truly reshaped my understanding of this industry is not 368. It is the names I had to exclude during the process of compiling the list.

368 Active Digital Banks

Let’s first discuss the undeniable rise of the industry. By aggregating data disclosed by various companies in the database, we find that the digital banks we track collectively serve approximately 1.46 billion users. This is not a forecast or a target market projection, but the actual customer scale reported by each institution.

The geographical distribution of users will overturn the preconceived notions of readers from Western fintech media: 817 million users are located in Asia. WeBank alone serves over 400 million users, surpassing the total number of users of all digital banks in the U.S. and Europe combined. Nubank has 131 million more customers than all U.S. digital banks combined. The European benchmark company Revolut has over 50 million users, which is impressive, but just a small fraction compared to the Asian market.

The focus of industry innovation has also shifted. Among the digital banks established after 2020 that are still alive today, 30% belong to native Web3 self-custody applications, which do not hold user funds. In contrast, only 4% of similar institutions established in the 2010s fall into this category. Regardless of how you view cryptocurrencies, builders have made their choices through action.

Thus, the rise of digital banks is real. The 368 institutions can be divided into three distinctly different waves of development: 254 traditional challenger banks, 58 fiat and crypto hybrid platforms, and 56 Web3 native platforms; supported by 106 infrastructure service providers and backed by 219 investment institutions. The complete ecological context is clearly visible.

Next, we address the truths that will not appear in any financing pitch deck.

Among the 368 institutions, only 127 hold complete banking licenses.

Yes, platforms named “banks” in app stores, two-thirds do not possess banking qualifications. They rely on partnerships with licensed banks, electronic money licenses, or lesser-known card issuing institutions to gain operational permissions. The vast majority of customers cannot distinguish the boundaries.

This is not an inconsequential technical detail, but a core structural risk of the entire industry, which has already led to multiple incidents:

• 2018 WaveCrest: Visa terminated cooperation, leading to the overnight shutdown of dozens of crypto card projects;
• 2020 Wirecard: A €1.9 billion funding gap caused the freezing of numerous digital banking services in Europe that relied on it;
• 2024 Synapse: The bank-as-a-service (BaaS) provider went bankrupt, and ordinary users in the U.S. realized that the so-called “FDIC protection” was not what they imagined, as the underlying system recording fund ownership encountered issues;
• 2026 Ready: History repeats itself, only the protagonists have changed to a new batch of players.

When a legitimate bank goes bankrupt, deposit insurance will compensate depositors. Once the infrastructure that digital banks rely on collapses, customers can only wait in line for bankruptcy liquidation.

The disappearance of this industry is always silent, which is precisely the most alarming aspect.

At the beginning of compiling this database, I did not anticipate that the exclusion process from the list would never stop.

Just this month, five institutions have been removed from the list: liquidated, acquired, or quietly transformed. There were no press releases, no retrospective summaries. Digital banks do not collapse dramatically like FTX. Instead, the app stops updating, customer service no longer responds; one day, the official website domain redirects to a partner's webpage, and hundreds of thousands of customers either have to move their funds or lose them altogether.

No media writes obituaries for the digital banks that have disappeared. Fintech media is keen to report on new projects launching and financing news, with advertising revenue and industry resources concentrated here. The companies that go bankrupt remain forever hidden from view. Thus, a new generation of founders continues to step into the same traps, thinking they are the first to discover the problems.

This is precisely why we equally emphasize the birth of new institutions and the exit of institutions. The data from failure cases is far more valuable than financing news. Press releases cannot teach lessons.

“Can artificial intelligence solve the profitability problem”? Is it really so?

Almost all digital banks’ financing PPTs now mention artificial intelligence. We verified each of the 368 platforms against regulatory disclosures, filing documents, and actual products on the ground, rather than merely relying on marketing materials.

Only 67 platforms have achieved large-scale implementation, accounting for 18%. The remaining 300-plus are either still in pilot phases, merely “exploring research and development,” or directly borrowing models from partners to claim proprietary technology.

Who are the true players implementing AI digital banking? The answer is surprising. Most of the industry’s AI application pioneers are not well-known large companies, but credit institutions from emerging markets like Nigeria, the Philippines, Mexico, and Bangladesh. These regions have imperfect credit systems and rely on models to grant credit to individuals lacking credit records, which is not just a product highlight but the foundation for the survival of these enterprises. While Western markets discuss AI banks, the global south has already achieved large-scale implementation, with survival needs driving innovation.

The Truth Revealed by the Ecological Landscape

Looking at the ecological infrastructure, 106 service providers support 368 consumer-facing brands. In this layer of the system, a few partner banks, BaaS platforms, and card processing service providers simultaneously support dozens of upper-layer brands. The high concentration that consumers find difficult to perceive is the source of the next Synapse-like crisis.

This is the true face of the industry in 2026. The industry is experiencing a transformative growth, with 1.5 billion people using banking services via their phones, many of whom are accessing financial services for the first time; yet the entire system is built on underlying service providers that the public knows almost nothing about. Two-thirds of the platforms will struggle to survive if upstream partners encounter crises.

Finally, I offer three predictions, while accepting that they may be proven wrong:

• The licensing gap will close from both ends. Strong unlicensed platforms will acquire or apply for banking licenses; weaker platforms will gradually exit the market in 2027, and players in the middle ground will disappear.
• The first AI credit model failure will occur within the next credit cycle. Of the 67 large-scale implemented models, the vast majority have yet to experience complete economic downturn pressure. Some teams are about to encounter risk scenarios not covered in AI training data.
• The next generation of financial service clients will no longer be natural persons. Financial underlying channels for AI intelligent agents, intelligent agents managing wallets autonomously, self-issuing cards, and machine-to-machine payments currently have only 7 companies laying out plans. The current landscape resembles the Web3 track of 2021: few participants, niche models, but with long-term structural opportunities.