Synthetaic is a new company focused on the development of data – specifically, visual content – designed for the purpose of training artificial intelligence systems.

Corey Jaskolski, the company’s founder and CEO, brings experience from both National Geographic, where he was recently honored as Explorer of the Year, and a firm specializing in 3D media. He explained that his work with National Geographic highlighted a critical need for expanded data sets within the field of conservation.

This combination of backgrounds might seem unusual, but Jaskolski explained that he was engaged in a project aimed at automatically recognizing poachers and endangered species from video recordings. A significant challenge he encountered was the limited availability of existing images – both of individuals engaged in poaching (who understandably avoid being photographed) and of certain vulnerable animal populations in their natural habitats – necessary to effectively train AI for detection purposes.

He noted that other organizations are attempting to generate synthetic data for AI training through the creation of complete 3D environments, essentially reconstructing the world for AI learning. However, he pointed out that this method can often be excessively costly.

Conversely, Synthetaic’s (pronounced “synthetic”) methodology integrates the skills of 3D artists and modelers with technology built upon generative adversarial networks, resulting in a more cost-effective and readily scalable solution, according to Jaskolski.

The company also confirms the effectiveness of its results by evaluating AI systems trained on Synthetaic’s generated images using real-world data.

For its initial endeavors, Jaskolski stated his intention to collaborate with organizations dedicated to positive global impact, including Save the Elephants (leveraging the technology for animal population monitoring) and the University of Michigan (developing AI capable of identifying various brain tumor types).

Jaskolski further emphasized that Synthetaic’s clients do not require pre-existing AI expertise, as the company delivers a comprehensive, “end-to-end” solution.

The company announced today that it has secured $3.5 million in seed funding, led by Lupa Systems, with additional investment from Betaworks Ventures and TitletownTech (a collaboration between Microsoft and the Green Bay Packers). Having now raised a total of $4.5 million, Synthetaic is also a participant in the Betalab program, a joint initiative by Lupa and Betaworks focused on startups with the potential to “fix the internet.”