DataCamp founder Dieter De Mesmaeker’s Conveo has raised $50M from DST Global and Balderton to scale AI-powered consumer video research.

The startup is developing causal "world models" trained on synthetic data that can adapt to new situations from only a handful of examples, targeting applications from robotics to healthcare.

kausable raises €12M to build AI models that learn and adapt without retraining. German startup kausable, which develops reasoning-first AI models designed to adapt to changing environments without continuous retraining, has secured €12 million in a seed funding round led by UVC Partners and Entourage, with participation from HTGF, Mätch VC, and several AI industry and academic angel investors. * Established in 2025 by Johannes Haux, Dr Benjamin Herdeanu, and Gregor Ramien, kausable is a Heidelberg-based AI startup developing frontier AI models that can learn from limited examples and adapt to new situations without requiring constant retraining. * The company builds causal "world models" that learn cause-and-effect relationships rather than relying solely on patterns identified from large datasets. Its models are trained using synthetic causal data, enabling them to understand how systems behave and apply that knowledge to unfamiliar scenarios with only a small number of examples. * The technology is designed to improve AI efficiency in areas such as robotics, healthcare, forecasting, and industrial systems, where adapting to changing environments remains a major challenge. Details of the deal. * With the fresh capital, the company plans to expand its nine-person team, advance its rapid-learning frontier AI model, and move from research toward commercial applications by testing its technology with early customers across sectors including robotics, healthcare, forecasting, and industrial systems.