Will Zoom eavesdrop on your video calls to train its AI model? Here is the answer
Popular video conferencing platform Zoom is back in the news for privacy concerns. Zoom updated its terms of service in March this year, which has now triggered concerns that the company might use user-generated content on the platform, such as conversations and chats, to train its Artificial Intelligence (AI) model, that too without the users’ consent.
However, the company has tried to put the rumours to rest by releasing a statement on Monday (August 7). In a statement, it said the company “will not use audio, video, or chat customer content to train our artificial intelligence models without your consent.”
Will Zoom tap your devices now?
According to Zoom’s latest promises, it would be incorrect to say it will eavesdrop on your video calls and virtual meetings. The current terms don’t allow user-generated content to be accessed by the company.
If the company is ever found to be collecting user-generated content, irrespective of consent, it would be violating its own terms, thus paving the way for litigation.
Are privacy concerns raised by experts illegitimate?
Not exactly. The company can still access “service-generated data” for machine learning purposes, provided the required consent is given. Service-generated data refers to data regarding customer behaviour such as what features do customers use more.
According to the new terms of service, service-generated data can be used for “machine learning or artificial intelligence (including the purposes of training and tuning algorithms and models.”
However, to allay the privacy concerns, the company says customers will have to enable the feature themselves and then separately give consent to share the data with the company.
What if the host of a meeting enables this feature?
According to experts, the functioning of this feature could still present challenges for certain participants during Zoom calls, particularly when the host chooses to activate the generative AI capabilities.
Zoom has clarified that when a meeting organiser opts for the meeting summary function, participants will receive a notification informing them about the activation of an AI feature. This notification will also indicate the potential sharing of their data for AI training purposes. Participants will then have the option to proceed with the meeting or to exit.
Watch:Zoom orders partial work-from-office policy for workers
While this approach theoretically empowers all participants to manage their data usage, it might not be feasible for someone to decline a meeting or entirely avoid using Zoom if they disagree.
Katharine Trendacosta, the director of policy and advocacy at the Electronic Frontier Foundation, raised a pertinent question: “If the administrator consents and it’s your boss at your work who requires you to use Zoom, how is that really consent?”
WATCH WION LIVE HERE