Skin In The Game: Why Banks Are Taking Stakes In AI Labs

by · Inc42

SUMMARY

  • Banks are moving beyond buying AI tools, becoming investors, strategic partners and co-builders with AI startups, particularly with an eye on creating AI-powered experiences for their customers
  • As AI moves deeper into critical banking workflows, institutions want greater control, customisation and influence over the technology they depend on
  • After a wave of global banks backing AI labs in the West, India is also beginning to see this shift with HDFC Bank and Canara Bank’s CoRover bet and IDFC FIRST Bank’s collaboration with Sarvam AI
  • Added to Saved Stories in Login

Rogo, a US-based financial-services AI startup, raised $30 Mn this month. What made this unique was that some of the world’s largest banks, including Barclays, BNP Paribas, Citi, MUFG and Société Générale, invested in this company, even though it has VC backing too. 

Nine global banks are now on Rogo’s cap table, representing nearly $20 Tn in combined assets.

Rogo isn’t an isolated case. Large banks and their venture investment arms are increasingly taking big bets on AI companies. 

Such investments have grown at a compounded annual rate of 21% since 2023, per an Evident AI analysis of the 50 biggest global banks, with Wells Fargo, Citi and Goldman Sachs the most active US bank investors. 

Recent deals around the world and in India back this trend. Citigroup and Spain’s Santander have both taken stakes in Tokyo-based Sakana AI, which builds foundational models and custom AI for financial services. 

In India, examples include HDFC Bank and Canara Bank’s investments in CoRover as well as IDFC Bank’s partnership with Sarvam to build an R&D lab with the aim of developing the world’s first self-improving bank.  

Three forces seem to be pushing banks past the old vendor relationship. Deep integration into regulated, high-stakes workflows requires more trust and customisation than an arm’s-length license can provide. 

Owning a stake or co-building the R&D gives banks influence over a roadmap they’re increasingly dependent on. And for BFSI institutions sitting on enormous proprietary data and distribution, contributing that data to jointly built AI systems can be more valuable than a services contract.

“License provides a means to access technology, while investment or co-development may result in a more profound collaboration between a bank and a startup. Cocreation gives the bank an ability to affect product architecture and work flows instead of simply fitting the bank’s processes into a ready-made solution,” said Mukesh Pandey, founder of consultancy firm Rupyaa Paisa.

Whether this becomes the default model for bank-AI relationships, or stays concentrated among the largest, most AI-forward institutions, is still an open question. 

So is what this means for smaller AI startups negotiating with a bank that is simultaneously their client, their investor and increasingly their co-developer. The ones that end up without any bank partnerships or strategic investments may struggle to get the same access to data, distribution and trust.