Sooth Labs, a newly launched artificial intelligence research company founded by former Meta researchers and Carnegie Mellon University professors, has raised approximately $50 million in a seed funding round to develop AI models designed to forecast the probability of future geopolitical, economic, and real-world events. The financing values the Pittsburgh-based startup at roughly $335 million, including the new investment, as it seeks to bring predictive AI capabilities to enterprises operating in increasingly uncertain environments.

The seed round is being led by Felicis and includes participation from AI pioneers Yann LeCun and Jeff Dean. The company has also attracted support from Ardent Venture Partners, according to funding records associated with the transaction. The backing from some of the most recognized names in artificial intelligence reflects growing investor confidence in Sooth Labs’ vision of building AI systems capable of predicting future events rather than simply generating content.

Founded by Yaser Sheikh, Ruslan Salakhutdinov, Chuck Hoover, David LaRose, and Shih-En Wei, Sooth Labs combines expertise in artificial intelligence research, computer vision, machine learning, and large-scale AI systems. Several members of the founding team previously held senior research positions at Meta while also maintaining academic leadership roles at Carnegie Mellon University, where they contributed to advances in AI and computer vision.

Unlike conventional large language models that primarily generate text or answer questions based on existing information, Sooth Labs is developing foundation models specifically trained to estimate the probability that future events will occur. The company’s technology is designed to process massive amounts of multimodal information, including text, audio, video, images, and structured datasets, enabling organizations to make better-informed decisions in rapidly changing environments.

According to the company, its predictive AI models are intended to help businesses anticipate geopolitical developments, market shifts, economic trends, supply chain disruptions, and other complex events that can significantly affect strategic planning. Rather than offering deterministic predictions, the models generate calibrated probability estimates, allowing decision-makers to assess uncertainty and risk more effectively.

The newly raised capital will support continued research, expansion of engineering and scientific teams, and the development of large-scale AI infrastructure required to train increasingly sophisticated predictive models. Building foundation models capable of forecasting real-world events requires enormous computational resources as well as access to diverse datasets spanning multiple domains and formats.

The company’s leadership believes advances in artificial intelligence have reached a point where models can move beyond understanding existing information toward reasoning about possible future outcomes. By combining multimodal learning with probabilistic forecasting techniques, Sooth Labs aims to create AI systems that assist organizations facing high-stakes decisions in sectors such as finance, insurance, defense, energy, logistics, and real estate.

The participation of Felicis further strengthens the company’s position within the competitive AI startup ecosystem. The venture capital firm has a long history of investing in early-stage artificial intelligence and enterprise software companies, while investors Yann LeCun and Jeff Dean are widely recognized as two of the world’s most influential AI researchers. Their involvement provides not only financial backing but also significant technical credibility as Sooth Labs begins commercial development.

In addition to investor support, the company benefits from guidance provided by prominent technology leaders. Meta Chief Technology Officer Andrew Bosworth is serving as an advisor to Sooth Labs, bringing operational experience from one of the world’s largest AI research organizations.

The emergence of Sooth Labs comes amid rapidly growing investment in artificial intelligence startups developing foundation models for specialized enterprise applications. While many AI companies have focused on conversational assistants, coding tools, image generation, and enterprise productivity, Sooth Labs is pursuing a distinct category centered on forecasting future events through probabilistic reasoning. The approach seeks to provide organizations with decision-support systems capable of quantifying uncertainty across highly complex scenarios.

With approximately $50 million in seed financing and backing from leading venture investors and globally recognized AI researchers, Sooth Labs is entering the market with significant resources to pursue its ambitious research agenda. As organizations increasingly seek AI tools capable of improving strategic decision-making under uncertainty, the company aims to establish itself as a leader in predictive artificial intelligence by developing models that forecast the likelihood of real-world events before they unfold.

Related Articles