Company attributes
Other attributes
Synthetaic is a synthetic data and artificial intelligence company. It develops training data for artificial intelligence when there is limited or no starter data. Synthetaic uses a combination of novel generative artificial intelligence and 3D modelling similar to that used in video game technology to generate large data sets.
It claims that it is able to produce large data sets with little to no starting data in order to move beyond data access and quality issues that can limit the application of artificial intelligence in certain industries and use cases. Synthetaic also says it is able to model and detect common occurrences for modelling of edge cases as well as ordinary and prevalent ones. Their modeling workflows include 3D modeling and generative artificial intelligence to deliver an end-to-end training pipeline.
Synthetaic suggests its synthetic data training is useful in security scenarios where artificial intelligence is applicable. It claims to be especially useful for limited data situations and in remote areas where automated and artificial intelligence can aid security operations, such as identifying weapons used by illegal poachers.
Synthetaic notes its synthetic data training for artificial intelligence is useful for real-time feedback in medical situations, especially with rare diseases or disorders which already have limited traditional data such as COVID-19.
For COVID-19 research, Synthetaic used a custom artificial intelligence workflow and hybridized data synthesis approach to project COVID-19 infection cases, investigate potential virus mutations, and find correlations that could lead to treatment options.
While working on preservation efforts for the Sumatran Rhino, Corey Jaskolski came up with the idea for founding Synthetaic. This came after he developed a 3D representation of the Sumatran Rhino and was struck by how real the representation looked. He thought this would help in training artificial intelligence in applications where good data is hard to come by. He saw this as a way of growing data faster, cheaper, and more aligned with training artificial intelligence.
