Artificial intelligence testing platform provider Distributional Inc. announced today that it has raised $19 million in new funding to support its mission of making AI reliable for enterprise use.

Founded in 2023 by Scott Clark, the former general manager of AI software at Intel Corp., Distributional is building an enterprise platform for consistent, adaptive and reliable AI testing. The company’s platform tests the consistency of any AI or machine learning application, giving AI engineering and product teams confidence in the reliability of their AI applications.

Distributional argues that, unlike traditional software testing, AI testing needs to be done more consistently and adaptively over time on a meaningful amount of data due to AI being inherently “probabilistic and dynamic.” Added into the mix is the ongoing need to avoid operational risks by deploying faulty products due to a business’s financial, regulatory and reputational bottom line.

The platform tests applications such as generative AI, which is claimed by Distributional to be particularly unreliable since it’s prone to non-determinism or varying outputs from a given input. The company says generative AI is also more likely to be non-stationary, with many shifting components that are outside of the control of developers.

Distributional helps automate AI testing with intelligent suggestions on augmenting application data, suggesting tests and enabling a feedback loop that adaptively calibrates these tests for each AI application being tested.

The platform allows AI product teams to continuously identify, understand and mitigate AI risks before they affect customers. By proactively addressing potential issues, the service ensures the reliability and consistency of AI applications across their lifecycle.

Distributional’s platform also offers an Extensible Test Framework that allows teams to gather and enhance data, run tests and respond to alerts through adaptive calibration or debugging. It does so while seamlessly integrating with existing datastores, workflow systems and alerting platforms to provide a self-managed solution within customer environments.

Additional features include a Configurable Test Dashboard and Intelligent Test Automation that allows teams to collaborate on test workflows, analyze results and scale AI testing effortlessly. The features can also help fine-tune testing processes across all AI applications to ensure that teams can maintain reliability and adapt to dynamic AI environments.

Two Sigma Ventures LP led the Series A round, with Andreessen Horowitz, Operator Collective, Oregon Venture Fund, Essence Venture Capital and Alumni VenturesGroup also participating.

Image: SiliconANGLE/Ideogram