The Spanish deep tech scale-up Sherpa.ai has developed a SaaS platform for privacy-preserving AI model training through Federated Learning. Simply explained, the platform ensures the AI models are trained with data that never leaves its original location, thereby protecting sensitive personal and enterprise data.

Led by the recognized AI innovator Xabi Uribe-Etxebarria, Sherpa.ai has all the chances of becoming a global success story. And we are proud that the company is a member of the EIC Scaling Club’s Digital Security & Trust market group.

AI with privacy in mind

The company’s “secret sauce” lies in its advanced federated learning platform, a privacy-first AI solution that enables organizations to train machine learning models collaboratively without transferring any actual data. 

This is a game-changer in heavily regulated industries like healthcare and finance, where sharing data across institutions is often impossible due to legal, ethical, or competitive concerns. Instead of requiring data to be centralized, Sherpa.ai’s platform sends models to local nodes (for example, hospitals or banks), where training occurs locally. Then, only the updated parameters are shared and securely aggregated.

Sherpa.ai’s technology is designed with privacy by design principles, meaning that privacy and data protection are built into the system from the start, not added later. The result is a platform that offers fast and safe deployment of AI solutions with real-world impact on operational efficiency and prediction accuracy. 

“By harnessing the power of AI to act on big data, corporations can become more productive, reduce costs, and enter new markets,” explains Sperpa.ai founder and CEO Xabi Uribe-Etxebarria.

The revenue is generated through subscription fees, which can vary based on the level of service and the number of users.

A team with a proven track record

Uribe-Etxebarria is a recognized entrepreneur and AI visionary, named one of MIT Technology Review’s top innovators under 35 and one of the 100 most influential people in Ibero-America by El País. 

Initially, Sherpa.ai was built to be an alternative to consumer-facing voice assistants, such as Siri and Google Assistant. Under Uribe-Etxebarria's leadership, the company strategically evolved into the federal learning platform it is now.

Behind Uribe-Etxebarria stands a stellar advisory board and leadership team with experience in both technology and policy. 

To name a few, Tom Gruber, former CTO of Siri, is the company’s Chief AI Strategy Officer. Joanna Hoffman, former marketing executive at Apple and NeXT, is Uribe-Etxebarria’s right-hand. Whereas Thomas Kalil, former deputy director for Science and Technology Policy at the White House, joined Sherpa.ai as Senior Advisor for Innovation. And that’s not it – the impressive lineup goes on.

A high-impact pivot

Like any scale-up company, Sherpa.ai has faced challenges. Its success with their initial product – the Spanish-language voice assistant that surpassed 5 million users in 2019 – was eventually threatened by global giants like Apple and Amazon that began adding Spanish to their services. That made it impossible for Sherpa.ai to keep up in the long term, and the company pivoted toward federated learning.

This pivot has paid off. Today, Sherpa.ai can take pride in forming several strategic partnerships and achieving several milestones. 

In 2021, the company closed a $8.5 million (approx. €7.1 million) funding round to double down on their new focus: B2B privacy-first federated learning services. The scale-up has also received a €1.4 million European Innovation Council grant.

A year later, Sherpa.ai reached a deal with Telefónica Tech to deliver federated learning solutions across Spain. This partnership laid the foundation for use cases across healthcare and finance, supporting early disease diagnosis and fraud detection.

Over the years, the scale-up has received numerous awards. Most recently, Sherpa.ai was selected as one of the World’s 5 Enterprises in the “AI Implementer of the Year” category at The AI Summit London.

In March 2024, Sherpa.ai was among the first 48 companies to join the EIC Scaling Club.

About the EIC Scaling Club

The EIC Scaling Club is a curated community where 120+ European deep tech scale-ups with the potential to build world-class businesses and solve major global challenges come together with investors, corporate innovators and other industry stakeholders to spur growth.

The top 120+ European deep tech companies will be carefully selected from a pool of high-growth scale-ups that have benefitted from EIC financial schemes, other European and national innovation programmes, and beyond.

The EIC Scaling Club is an EIC-funded initiative run in partnership by Tech Tour, Bpifrance (EuroQuity), Hello Tomorrow, Tech.eu (Webrazzi), EurA and IESE Business School.

Subscribe to our newsletter here to stay up-to-date!