Zoom AI Tools Trained Using Customer Data; Terms Now Updated

Zoom intends to enhance its artificial intelligence models by utilizing certain customer data, as per the recent updates in its terms of service.

The updated terms of service for the video platform, effective since July 27, focus on sections related to software licensing, beta services, and compliance. However, a closer look at the fine print reveals that Zoom has made a significant decision in its AI strategy. The update grants Zoom the right to use customer data, including product usage information, telemetry and diagnostic data, and similar content collected by the company, for training and tuning its AI and machine learning models.

Although the utilization of “service-generated data” for AI training is not uncommon, these new terms showcase Zoom’s dedication to its own AI ambitions.

This update comes at a time when there is an ongoing public debate regarding the ethical use of individuals’ data for training AI services, regardless of how aggregated or anonymized it is claimed to be. Lawsuits have recently emerged from authors and artists who argue that AI tools are producing outputs that mimic their own work, including chatbots like OpenAI’s ChatGPT, Google’s Bard, Microsoft’s Bing, and image-generation tools like Midjourney and Stable Diffusion, which are all trained on large amounts of internet text or images.

“You consent to Zoom’s access, use, collection, creation, modification, distribution, processing, sharing, maintenance, and storage of Service Generated Data for any purpose, to the extent and in the manner permitted under applicable Law, including for the purpose of … machine learning or artificial intelligence (including for the purposes of training and tuning of algorithms and models),” state Zoom’s terms of service.

It is worth noting that customer content such as messages, files, and documents does not fall under this data category. In a follow-up blog post, Zoom clarified that they do not use audio, video, or chat content for training their models without the consent of the customer. The key phrase being “without customer consent.”

In June, Zoom introduced two new AI features, a meeting summary tool, and a chat message composition tool, on a free trial basis for customers. Users have the choice to enable or disable these features and must sign a consent form if they decide to enable them, granting Zoom permission to train its AI models using their individual customer content.

“Your content is used solely to improve the performance and accuracy of these AI services,” Zoom explained in the blog post.

A spokesperson for the company stated that “Zoom customers have the freedom to enable generative AI features and decide separately whether to share customer content with Zoom for product improvement purposes.”