Zoom’s Privacy Misstep: Is their AI Being Trained with Your Information?
By Betsy Burton
Zoom’s Privacy Misstep: Is Their AI Being Trained with Your Information?
In early 2023, Zoom changed its terms of service so that its statement about its rights over customer data was, to say the least, confusing.
It led to several high-profile social media postings that charged that Zoom’s new terms of service give the company unfettered permission to use all customer data, including private conversations, for training artificial intelligence, with no ability to opt-out.
This led to a flurry of comments, speculation, and assertions on various social media platforms, including LinkedIn.
Zoom responded with a blog that these assertions were not their intention. And that their effort was to clarify their different treatment of service data and customer content data.
Customer content data is all the data created by users, such as images, collaboration, audio, or chat transcripts. While, service data refers to the aggregate use of the Zoom services like number of Zoom sessions, length of sessions, number of attendees, etc.
But….the problem is the damage was done. The trust Zoom built up so well during the pandemic has been damaged.
Even though they have clarified their terms of service by adding the language “For AI, we do not use audio, video, or chat content for training our models without customer consent,” many organizations and users are thinking twice about their interactions on Zoom.
What Does Zoom Collect for AI Training?
Zoom has clearly told users that it will not use customer data without consent. For most regular meetings data sharing is by default “off.”
However, if you use Zoom IQ Meeting Summary and Zoom IQ Team Chat Compose the default is that data sharing is “on.” This means Zoom can use (e.g., distribute, copy, etc,) your meeting content (images, audio, creation, etc.), including using this content to train AI systems.
If someone is attending a meeting, and the host decides to use Zoom’s IQ meeting or Chat Compose, a user will get a message saying you either approve of this usage or choose to leave the meeting.
Great….my boss asked for this meeting….am I really going to say no to attending the meeting?
As Zoom navigates this privacy issue, we all need to clearly define our privacy strategy, and communicate it to our peers and employers, so they know what we are and are not comfortable sharing.
How Will This Affect Zoom and Other Service Providers?
For Zoom, this was clearly a significant PR misstep. As mentioned above, Zoom garnered so much goodwill during the pandemic. This privacy issue will certainly damage its brand. This isn’t fatal, but it was a significant misstep.
Note to service providers, don’t let your legal team set your terms of service alone. Make sure to involve product, customer service, and sales leaders. Also, terms of service should be tested with users before deploying to make sure you don’t damage your brand.
However, we need to be realistic, we live in an information-capitalist economic model. Most businesses are a hybrid mix of revenue from products and services and revenue from selling customer information, both aggregate and specific. Service providers will increasingly want to use both service data and customer content data about you and your business.
This is not the last time we have seen a provider stumble over this issue. And people’s willingness to forgo information in exchange for services and convenience will continue to evolve.
Users and businesses are becoming increasingly aware of what they are giving away in terms of their personal information, image, and generated content. The increased use of AI is just making this issue more acute. Some people and businesses really care about managing that information; while others are willing to give away that information.
Product and service providers must ensure that they have an information usage strategy that supports the business strategy and brand. This must not be just a tactical decision, or just left to the lawyers.
End users and end-user organizations must pay attention to the information that providers are gathering and ensure that this fits into their information-sharing strategy and plans.
Last, every one of us must clearly define now what information we are willing to share and what not. And ensure you are managing the usage of your services.
Does the evolution of artificial intelligence (AI) into the workplace mean a job desert or a gold rush? The answer, in my view, is neither of these extremes. But it will absolutely change the workplace landscape and we must work on understanding and planning for these changes.
During this webinar, we will be exploring the potential impact artificial intelligence will have on different jobs and on the workforce, in general. In addition, we will be introducing Aragon Research’s new AI Technology Arc.
- What jobs will be the most impacted by AI technologies?
- How will AI change the workforce?
- What technologies do leaders need to track over the next 3 years?