InstaDeep raises $100M to inject enterprise decision-making with AI

janvier 25, 2022 Par admin 0
InstaDeep raises $100M to inject enterprise decision-making with AI

Developers.

Image Credit: Getty Images

Join gaming leaders, alongside GamesBeat and Facebook Gaming, for their 2nd Annual GamesBeat & Facebook Gaming Summit | GamesBeat: Into the Metaverse 2 this upcoming January 25-27, 2022. Learn more about the event. 


AI has the potential to generate meaningful returns for the enterprise. Responding to a 2018 PricewaterhouseCoopers survey, 54% of business executives say that their adoption of AI within the workplace has led to a boost in productivity. A separate 2019 McKinsey report found that 44% of firms using AI achieved a reduction in business costs in departments where AI is implemented.

But barriers stand in the way of deployment, including a lack of production-grade data and expensive tools and development processes. Among the top challenges enterprises face in adopting AI is an absence of in-house talent. Indeed, laments over the AI talent shortage in the U.S. have become a familiar refrain. In 2018, Element AI estimated that of the 22,000 Ph.D.-educated researchers globally working on AI development and research, only 25% are “well-versed enough in the technology to work with teams to take it from research to application.”

Increasingly, rather than hire their own talent, companies are turning to specialized vendors to fulfill their predictive analytics goals. According to Deloitte, 59% of organizations are meeting their AI needs by adopting third-party enterprise software with AI. One of the more prolific providers is InstaDeep, a London-based firm that bills its products — which leverage AI to solve problems across logistics, energy, biology, and electronic design — as “decision-making.” InstaDeep today announced that it raised $100 million in a funding round led by Google, BioNTech (the company behind Pfizer’s COVID-19 vaccine), Chimera Abu Dhabi, Deutsche Bahn’s DB Digital Ventures, G42, and Synergie.

Reinforcement learning

InstaDeep was founded in 2014 by Karim Beguir and Zohra Slim in Tunis in North Africa, with little more than two laptops and $2,000. Slim, who’s completely self-taught, was leading a team of developers in India prior to cofounding InstaDeep. Beguir, a graduate of France’s Ecole Polytechnique and former program fellow at NYU’s Courant Institute, is a steering committee member of Deep Learning Indaba, an organization whose mission is to strengthen machine learning and AI in Africa.

Employing GPU-accelerated computing, machine learning, and reinforcement learning, InstaDeep builds AI systems to tackle challenges across a range of industries and organizations. In reinforcement learning, a system is given a set of actions that it can apply to its environment (e.g., a map of possible transit routes). The system — which usually starts knowing nothing about the environment — receives rewards based on how its actions bring it closer to a goal. As the system gradually receives feedback from the environment, it learns sequences of actions that can maximize its rewards.

InstaDeep’s AI-powered circuit board designer.

InstaDeep staffs a team of consultants and data scientists that design AI solutions around a company’s data structure and software architecture. After defining a business objective and piloting an AI solution, InstaDeep helps to implement the solution and provides ongoing support.

InstaDeep developed an AI-accelerated protein design platform, DeepChain, by training AI language models on billions of amino acids — enabling users to discover designs and validate them with molecular simulations. (Protein discovery is a key — albeit typically a time-consuming and expensive key– step in the drug discovery process.) Another of InstaDeep’s platforms, DeepPCB, can automatically create printed circuit board blueprints in less than 24 hours, ostensibly shaving weeks off a process that’s normally completed manually.

InstaDeep recently partnered with Google AI, one of Google’s AI research divisions, to create an early detection system for desert locus outbreaks across the African continent. Desert locust outbreaks — which have become more common following the migration of billions of locusts in 2019 and 2020, after cyclones on the Arabian Peninsula — threaten the food security and livelihoods of millions of people. By combining data from the U.S. Food and Agriculture Organization with climatic and environmental data from NASA and the International Soil Reference and Information Centre, InstaDeep claims it was able to create outbreak-forecasting models that can generalize across many different countries.

InstaDeep has also completed a study with BioNTech that investigated ways AI can be applied to aid in predicting new COVID-19 variants. The “Early Warning System” (EWS) that the two firms prototyped combines structural modeling of the spike protein — the protein that allows the virus to attack cells — with AI algorithms to flag potential high-risk variants. The EWS has identified over 90% of the World Health Organization-designated variants of concern in sequences of the virus uploaded online, InstaDeep claims, including Alpha, Beta, Gamma, Theta, Eta, and Omicron — on average two months in advance.

Growing solutions

InstaDeep claims it has worked with other customers to increase efficiency across the industrial supply chain, make decisions on ride-hailing vehicle fleet size, and predict machine failure in factories and warehouses. Currently, the company is pursuing a “moonshot” product to automate railway scheduling with Deutsche Bahn, the largest rail operator and infrastructure owner in Europe.

“[W]e see wide-ranging opportunities to deploy our AI products to tackle complex real-world problems,” Beguir said in a statement. “The funding round is a tremendous vote of confidence from our partners.”

The challenge for InstaDeep, which competes with AI-focused consultancies including Fractal Analytics, is overcoming the institutional barriers to successful AI deployments. In a 2020 survey, O’Reilly found that a lack of knowledge about modeling and data science, understanding business use cases, and data engineering were major blockers in the enterprise. Executives also cited a lack of quality data as a reason that their organization was struggling to successfully implement AI.

Still, the interest — and investment — in AI isn’t waning. According to Gartner, 33% of tech providers plan to invest $1 million or more in AI within the next two years.

“Rapidly evolving, diverse AI technologies will impact every industry,” Gartner managing VP Errol Rasit said in a recent blog post. “Technology organizations are increasing investments in AI as they recognize its potential to not only assess critical data and improve business efficiency, but also to create new products and services, expand their customer base and generate new revenue. These are serious investments that will help to dispel AI hype.”

With the new funds, InstaDeep plans to expand its high-performance computing infrastructure and launch new products in emerging verticals. The company says it’s also aiming to grow its workforce across its offices in Paris, Lagos, Dubai, and Cape Town.

VentureBeat

VentureBeat’s mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact.

Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:

  • up-to-date information on the subjects of interest to you
  • our newsletters
  • gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
  • networking features, and more

Become a member