We have entered an era where the capabilities of large language models (LLM) are rapidly expanding the world of “possible.” However, as LLMs proliferate, concerns surrounding data ownership are mounting, leaving both privacy advocates and data owners uncertain about the extent of their protection. Representative of this rising tension video-conferencing giant Zoom’s recent scandal, known for its prominent role during the remote work boom.

The source of the controversy lies in Section 10.4 of Zoom's updated terms of service, which were rolled out in March 2023. In these terms, users grant Zoom a broad license for various purposes, including machine learning, artificial intelligence, and product improvement. While this in itself is not novel in an era of proliferating large language models, the verbiage of the privacy policy sparked outcry amongst Zoom’s customers.

In the new iteration of the privacy policy, customers granted “Zoom a perpetual, worldwide, non-exclusive, royalty-free, sublicensable, and transferable license” for various purposes, including “machine learning, artificial intelligence, training, testing, improvement of the Services, Software, or Zoom’s other products, services, and software, or any combination thereof.”

The vague use cases for the data and the expansive language in the privacy policy were at odds with the growing public ethos of data accountability. As highlighted in an article in Stack Diary, Zoom’s updated language was likely to ensure flexibility to stay competitive in the AI race.

Indeed, Zoom has rolled out a suite of AI-powered tools called Zoom IQ, allowing users to generate meeting summaries and auto-draft messages to meeting attendees. While the usage of these features may be optional, Zoom still asserted the right to collect comprehensive user meeting data, for seemingly any use case it wanted.

This situation highlights the increasing scrutiny surrounding AI technology, particularly the worries about using individuals' data and content for AI training without their consent or compensation. Zoom has since released a statement assuring users that their meeting data will not be collected and used for AI model-training without explicit consent.

In the end, the Zoom controversy underscores the broader need for transparency and public discourse around AI integration in various products and services. Privacy experts and advocates argue that users often don't pay attention to terms of service changes, leading to complex challenges for consumers in navigating their digital rights.

In a world where technology continually pushes boundaries, the need for robust data protection laws and user-friendly terms of service agreements becomes increasingly apparent. As individuals, it's crucial to remain informed about our rights and advocate for greater transparency in how companies handle our data in the age of AI.