iGenius.ai, a European AI scale-up valued at more than $1 billion, develops opensource large language models that respect strict data security rules to ensure that corporate clients can safeguard their intellectual property. The company’s AI platform is used by some of the world’s largest organizations including financial services company Allianz, electric utility company Enel, Intesa Sanpaolo, one of the top banking groups in Europe, global shipbuilding group Fincantieri and several governments, including Italy’s.
“We are attracting customers in the U.S.,” says Uljan Sharka, co-founder and CEO of the Italian scale-up. “We tell them if you want the best models go to Silicon Valley. If you want good performance and large language models you can trust you can talk to us.”
Sharka is one of dozens of AI entrepreneurs invited to the global AI Action Summit which is being co-hosted by the French and Indian governments and will take place in Paris February 10 and 11. The conference will address, among other things, concerns about control of GenAI by a few dominant players, the role of opensource and how to equitably distribute AI’s benefits globally and build responsible and trustworthy AI.
“Europe is well-poised to win a GenAI-led digital Renaissance by offering an alternative to American and Chinese models that embeds European values and builds trust,” says Sharka.” It is a different way of building tech, and it is an advantage,” he says.
Approximately 80% of the most valuable enterprise data -including personal information, financial transactions, trade secrets, and intellectual property -cannot be exported and fine-tuned to centralized models and/or open models with a limited license, explains Sharka, This is especially true of generative AI, which merges data and intellectual property irreversible. For example, financial institutions sharing sensitive data with centralized LLMs can lead to potential data breaches or misuse, which can expose proprietary trading strategies, signal market intentions, and enable market manipulation, he says.
In January iGenius.ai announced the launch of Colosseum 355B, a new state-of-the-art Large Language Model (LLM) with 355 billion parameters, which targets highly regulated industries with an alternative approach. Built with the latest NVIDIA technology, Colosseum 355B enables regulated organizations to create regionally-specific LLMs with Continual Pre-Training (CPT). Additionally, it makes model building at scale accessible to more enterprises, a capability that was previously limited to hyperscalers, says Sharka.
Colosseum 355B is designed for both CPT and fine-tuning. CPT allows enterprises to own proprietary LLMs by adding domain-specific knowledge and enabling long-term scalability, without sacrificing general knowledge, says Sharka. Customers can then use fine-tuning to create powerful task-specific adaptations.
Colosseum 355B was developed on a single cluster with more than 3,000 NVIDIA H100 GPUs. This environment enables iGenius to rapidly scale its AI R&D and pre-training iterations, culminating in a highly capable model that supports over 50 languages, excels at coding, and is optimized for efficient deployment and resource utilization, cutting inference costs by 50%, says Sharka.
Sharka says its model’s ability to continuously pretrain allows companies to create their own general intelligence. GenAI’s real power is to give companies the ability to transform their own offers into intelligent products and services, says Sharka, and “this will only be possible if companies build their own general intelligence.”
Once corporates have that foundation they can launch effective virtual agents, he says. iGenius’ platform can be accessed through natural language-based applications across mobile, desktop and embedded end-user experiences. By introducing a conversational paradigm as the end-user interface of business data, iGenius aims to enable access to AI agents by all relevant employees.
Sharka cites the example of a leading financial institution which is using iGenius’ AI agent Crystal to help its financial advisors. Each morning the financial advisors can ask the agent, in natural language which customers he or she should meet with today or which customers need help, based on proprietary information in the company’s database. “This is about a company using its own LLM to amplify their employees’ potential,” he says.
Sharka emphasizes that GenAI is crucial to Europe’s competitiveness, helping transform Europe’s industrial champions like Airbus or Leonardo, global industrial group that focuses on aerospace, defense and security, into powerful tech companies.
The European capital market is undervalued, says Sharka. “Europe has €35 trillion in savings. If we create AI-drive industrial champions that are at the heart of every supply chain in the world we can increase their value 500X even 5000-fold and Europe will become the fastest growing capital market in the world.”
Sharka believes European AI companies have what he calls a “third mover advantage.” “The first mover OpenAI monetizes in a very aggressive way. Second movers – other hypersecalers- are copying and pasting at scale, competing on small increments of performance. Our models are better at other things – namely embedding European values and safeguarding data,” he says.
Sharka compares the current state of play to the evolution of the personal computer. The Olivetti Programma 101, also known as P101, was one of the first “all in one” commercial desktop programmable calculators, a forerunner of personal computers. HP, Microsoft and other US companies copied the Italian company’s model and beat it at its own game due to better sales channels. Then Apple came along and took a different approach.
“We have the chance to become the Apple of GenAI by safeguarding data privacy and incorporating human rights and other values at the heart of European culture by giving people around the globe an alternative to the U.S. and Chinese models,” he says.
iGenius aims to complete a major data center project in the south of the Italy this summer, using Nvidia technology and requiring a $1 billion investment spread over five years, its CEO said. The cost of the project had prompted the start-up to extend a funding round from an initial target of €650 million.
Sharka said the project’s new supercomputer will be able to execute 115 billion of billions”of calculations per second, making it one of the leading supercomputers in Europe
iGenius is part of The World Economic Forum’s Global Innovators community. It is a Gartner recognized Cool Vendor in AI Core Technologies and received an honorable mention in the 2024 Gartner Magic Quadrant.
To access more of The Innovator’s Startup Of The Week articles click here.
