Vaultree has cracked the code that will define the future of encrypted ML, AI and Fully Homomorphic Encryption (FHE). In an industry where the biggest technology giants and research institutions have struggled, Vaultree has succeeded in a mission once thought impossible. What began in the humble setting of an Irish dairy farm is now poised to lead an encrypted global data, ML & AI revolution, fundamentally redefining how data is protected and utilized.
The inadequacies of traditional encryption methods have shackled the industry. For decades, organizations have faced an unyielding conflict between data security and data utility. Existing encryption paradigms—securing data at rest and in transit—offer only a superficial layer of protection. The glaring vulnerability remains – when data is used, it must be decrypted – exposing it to significant risks and astronomical energy costs.
Vaultree’s breakthrough obliterates these limitations. Through relentless innovation and an uncompromising vision, they have accomplished what was once deemed unattainable: the ability to process data while it remains fully encrypted “at scale.” This is not just a technical milestone—it is a transformation that redefines the very foundation of data security, encrypted ML & AI. This isn’t just theoretical—Vaultree’s technology is already operating in production environments and is engaged with some of the largest healthcare and financial services organizations, solving these real-world challenges. Enabling organizations to tackle their most pressing business use cases whilst working on encrypted data was previously unimaginable. Vaultree is aggressively expanding in the US and has partnered with some of the world’s largest organizations, such as Google, continuing to forge new partnerships as they scale globally.
Vaultree’s founders, driven by an unyielding belief that data privacy is a fundamental human right, set out to solve these complex mathematical challenges. The mission was clear: create a future where organizations can extract the full value of their data without ever compromising security. And they have succeeded. Vaultree’s cryptography research team has achieved the unthinkable—solving the at-scale and runtime processing capabilities of IN-USE and Fully Homomorphic Encryption without the need for special purpose hardware. After years of stealth-mode research and development, Vaultree has unveiled Next-Generation FHE (NG-FHE) schemes, and alongside their latest public release, Vaultree has published a scheme as per the latest release. This is just the beginning of their aggressive expansion and Vaultree invites the public to collaborate and review.
With Vaultree’s technology, sensitive data is now processed without ever being decrypted, ensuring that data remains secure during every operation. But the vision extends beyond just protecting data—it is about enabling the future of encrypted ML & AI. Organizations can now fully harness the power of ML & AI without exposing their sensitive data to risks. For the first time, ML models can be trained on fully encrypted data, ensuring that organizations maintain the highest levels of data security while unlocking the true potential of their data, with unprecedented possibilities for innovation, growth, and competitive advantage.
Also Read: Cyberbit Secures Investment to Accelerate Global Growth
Vaultree has built the world’s first scalable encrypted transformer, capable of performing 4 to 10 billion vectorized FHE operations per minute—on a standard MacBook Pro M1. You can see a demo video here: Vaultree Encrypted Transformer Demo. This is a game-changer that opens a world of possibilities for ML & AI-driven innovation. Vaultree is leading the charge towards a future where encrypted ML & AI are not just possible, but ubiquitous. Their NG-FHE schemes are designed with a single goal in mind: to solve real-world business challenges and enable organizations to operate on always-encrypted data.
As the CEO and Co-Founder of Vaultree, Ryan Lasmaili, asserts, “Our mission is clear: to truly enable an encrypted future where data privacy, ML & AI are not just possibilities, but expectations. We are here to empower every organization, large or small, to harness the full potential of their data and AI without ever compromising security.”
Vaultree’s commitment to open-source innovation is a testament to their profound belief in collaboration and shared progress. They have launched a GitHub demo library for Encrypted Machine Learning, providing the global community with an unprecedented opportunity to experience the power of their NG-FHE implementation firsthand. The library shows Encrypted Time-Series Forecasting—dubbed ‘Phineus,’ Vaultree’s answer to Meta’s Prophet—for Encrypted Facial Recognition and Encrypted PageRank, while also keeping its eye on the future of Encrypted Quantum Machine Learning with a partnership in this space.
Vaultree is also expanding its footprint into academic and industry collaboration. They officially partnered with several leading research institutions. “We need to break the cycle of misalignment between different sectors to ensure we are razor-sharp focused on solving real-world problems and enabling a foundation for the future of data enablement,” says Lasmaili.
Looking ahead, Vaultree is at the forefront of developing quantum-resilient Data-In-Use Encryption schemes. They are devising methods that will automate the development of new schemes, rapidly test them, and enable their use in real-world applications. As Ryan Lasmaili emphasizes, “Just like at the beginning of the Covid pandemic: we didn’t have 10 years to find a new vaccine, and the same urgency applies now with Data-In-Use Encryption. Vaultree will continue to make more announcements next year and is eager to partner with proactive research labs, finding new ways to prove the security aspects and develop new secure FHE or other Data-In-Use Encryption schemes that we can deploy for data-in-use scenarios is critical.”
Source: Businesswire