Lessons learned building an AI company from the ground up

Description

Companies in all industries have begun implementing AI, with varying success. Although AI technology seems to be everywhere, implementing AI in practice is a real challenge. The technology needs to be scalable, trusted by the humans that use it, and easily accessible for those with limited AI expertise. Nicole Eagan shares the unique insights on building practical and successful AI applications Darktrace has gained from its 4,000+ deployments.

A successful AI application needs to be scalable, install quickly, learn on the job, and not require ongoing tuning of AI models. Companies can’t assume customers have any prior expertise deploying AI on their businesses networks. Additionally, you needn’t worry about being a purist when developing pragmatic AI applications. Instead, embrace a mix of AI algorithms and techniques, including supervised and unsupervised machine learning and deep learning.

Once the AI is developed, the next challenge is bringing the technology to the market and getting people to trust it. Providing visibility into these applications is one way to build human’s trust in AI. Darktrace has developed techniques—like 3D visualization and human confirmation mode—that work to build this trust. With no existing training dataset in existence, Darktrace created a unique differentiated approach that sets the company apart from the more common training dataset applications that puts AI at the core of its technology and business, not as an add-on feature.