Legal Storm Brewing: Zoom Faces Backlash for AI Training on Video Calls Without Opt-Out

18 Mar 2024 · 9 min

Ask about this episode

Ask anything about it. ChatGPT or Claude reads this page and answers with the times it was said.

Connect VO and ask about every podcast you hear, including the moments you saved. Add to ChatGPT · Add to Claude

In short

AI Today Podcast Episode Notes

Episode Title

Legal Storm Brewing: Zoom Faces Backlash for AI Training on Video Calls Without Opt-Out

Episode Overview In this episode, the hosts discuss the backlash against Zoom for using users' video calls to train AI models without providing an opt-out option. The conversation highlights the implications for privacy rights and the ethical considerations in an era of pervasive digital surveillance.

---

Key Topics Discussed

  1. Background of the Controversy
  2. March 2023 update to Zoom's terms and conditions sparked concerns regarding data usage.
  3. Users have voiced concerns over their data being used for AI training without explicit consent.
  1. Opt-Out Issues
  2. Users found that the only way to opt-out of this data usage was to switch to a different platform (e.g., Google Meet).
  3. The lack of a clear opt-out option led to significant pushback on social media.
  1. Zoom's Terms and Conditions
  2. Key clauses grant Zoom extensive rights to use non-content data for AI training.
  3. There are mixed signals regarding consent for customer content, with a focus on account admins being able to provide consent on behalf of all users.
  1. Legal Implications
  2. EU laws require individual consent for data processing, which Zoom's practices may violate.
  3. The process for consent is questionable, as participants are only given a notice option to leave if they do not wish to participate in AI data sharing.
  1. Concerns with Consent Mechanisms
  2. The opt-out process is viewed as misleading; users are pressured to either accept AI data sharing or leave the meeting.
  3. The notice process may not meet the requirements for informed and freely given consent as mandated by EU regulations.
  1. Regulatory Scrutiny
  2. Zoom faces scrutiny from EU regulators due to its status as a non-EU company, risking enforcement actions.
  3. Past privacy issues and FTC settlements may affect public trust and regulatory perceptions.
  1. Comparison with Other Companies
  2. EU regulators recently required OpenAI to provide opt-out options, indicating a trend towards stricter consent requirements for AI training data.
  1. Future Implications for Zoom
  2. If Zoom fails to comply with consent requirements, it may face severe penalties under GDPR laws.
  3. The importance of clearer communication and control over user data is highlighted as a means to avoid future privacy issues.

---

Conclusion The episode concludes with a call for companies like Zoom to be vigilant in their data practices, especially as they develop AI capabilities. The hosts emphasize the necessity for transparency and adherence to privacy regulations to maintain user trust and avoid regulatory backlash.

---

Additional Resources

  • Invest in AI Box: [AI Box on Republic](https://republic.com/ai-box)
  • AI Box Waitlist: [AI Box](https://aibox.ai/)
  • AI Facebook Community: [Join Here](https://www.facebook.com/groups/739308654562189)
  • Learn about AI in Music: [Musical AI](https://musicalai.pro/)
  • Learn about AI Models: [AI Models Pro](https://aimodelspro.com/)

---

Privacy Notice For privacy policies and further information, visit [art19.com/privacy](https://art19.com/privacy).

Written by AI. May contain mistakes. Listen to the episode to check what was said.

Hear the part that matters, and keep it.Open this episode in VO. Double tap your headphones to save a moment as you listen.
Get VO free

Transcript

Automatic transcript. May contain errors.

0:00The wait is over. Dive into Audible's most anticipated collection, The Best of 2025. featuring top audiobooks, podcasts, and originals across all genres. Our editors have carefully curated this year's must-listens from brilliant hidden gems to the buzziest new releases. Every title in this collection has earned its spot. This is your go-to for the absolute best in 2025 audio entertainment. Whether you love thrillers, romance, or nonfiction, your next favorite listen awaits. Discover why there's more to imagine when you listen at audible.com slash best of the year.

1:03time it's related to a clause that allows the video conferencing platform to use customer data to train AI models and they're doing this without an opt-out which a lot of people are obviously concerned about right like a lot of people when they find out their data is about to be used ingested by AI models they would just you know simply like to say no thank you I don't want this and it appears the only way to opt out of this is to switch over to a different video conferencing platform like Google Meet or something else. And, you know, that's left to be said. People have to scour through their privacy policy to see what Google Meet plans on doing it with, you know, everything in AI that Google is doing.

1:41So this whole controversy began when a March 2023 update to Zoom's terms and conditions was highlighted for essentially allowing Zoom expansive rights to usage. So, you know, the product they said was going to be used for analytics or for any purpose including trading its AI so we got a lot of backlash pushback on social media over its lack of you know control that users have on how their data and their video calls are being used but on closer inspection some users actually suggested the no opt-out applied only to non-content data zoom generates from usage so not necessarily your private meeting contents but you know still a lot of people are uncomfortable with Zoom repurposing their information to develop AI that could automate their role, right?

2:32If you're doing something on Zoom, a lot of people are concerned that the AI is just going to replace whatever they're doing and essentially they're giving over the data the way they do that for free. So the key clause in Zoom's terms allow it extensive uses rights of all non-content data. However, a bolded line actually emphasizes content is still required for audio, video, or chat customer content in AI training. So this implies that Zoom is treating consent as delegatable and essentially to the account admins that can, quote, provide consent on behalf of all users. Now, that is where a lot of people are a little bit concerned, right?

3:16They're saying, you know, you don't get an opt-out if it's non-content. But essentially, they're saying that their own account admins can provide consent on behalf of all users. And so really, they're saying there's no opt out for anything, including your video content. And it's kind of just a sneaky way of doing that, which a lot of people are not happy about. So EU law actually requires per individual consent if claimed as a legal basis for data processing. So the ePrivacy directive also mandates confidentiality of electronic communications without consent. So in response to the whole uproar and the controversy, Zoom updated its terms to, you know, confirm that no customer content will train AI without consent.

4:01But its blog post gives some mixed signals on this, again, really concerning a lot of people. Essentially, their blog post is, you know, depicting account admin menus as, quote, providing consent. while participants merely get quote notice if AI meeting summaries are enabled and so the the whole you know notice I think is a pop-up informing participants that account owners allowed their input to train Zoom's AI and the two options are got it or leave meeting so it really doesn't give you know like it really forces someone in the meeting to accept the AI data sharing or essentially just withdraw from the meeting like use it and have it added to the ai corpus or withdraw from the meeting those are your two options and essentially you know go find a different video platform so if zoom is claiming consent as its legal basis in the eu i think this notice process likely fails requirements like informed specific freely giving consent with no nudging pre-selections i'm telling users you know you're free to leave doesn't equally uh free doesn't really equal freely given consent to process personal data.

5:11So Zoom avoids directly, you know, confirming its legal basis, but reiterates participants can leave meetings if unhappy with AI data sharing. So essentially they're saying it is how it is and, you know, you gotta just not use the platform. So under EU law, that does not constitute valid consent. Even the admin consent process looks problematic with, you know, pre-checked data sharing boxes. Administrators have to, like, they have to actually notice and unchecked to opt out of right this isn't something that you can opt into this is something that the admins of the meeting have to figure out how to go and opt out of so according to one legal view zoom essentially is relying not on consent but on performance of contract so that's applying u.s law notions of data ownership however eu law recognizes metadata is personal data requiring proper consent and this makes a lot of sense right included in your metadata could be, you know, the GPS location where a photo was taken or a lot of other highly sensitive content.

6:11And so with higher bars for sensitive data in the EU, this could potentially become problematic because, and you know, in light of this, Zoom's communication also suggests it believes that metadata needs no consent. So another area where, you know, there is an issue. So with video and audio content also considered personal data, Zoom likely needs explicit opt-in consent from users for data to train its AI offerings in the EU. And merely, you know, noticing participants, I think, really falls short of this. So without a main EU establishment, Zoom faces, you know, full scrutiny on data practices by any of the bloc's regulators.

6:51And I think that while Zoom may hope that some of the oversight is kind of lax and, and lets it drive the new AI tools that they're trying to create, I think the substantial GDPR fines and additions of e-privacy duties along most places in the EU are really going to be painful for them. So for a company feeling pressure to keep pace with generative AI, ignoring regional safeguards is a high-risk legal strategy. And here's a few other points about this whole Zoom situation that I thought might be relevant. So number one, Zoom's privacy policy controversy comes just a few years after a previous FTC settlement over misleading claims about encryption and security.

7:38And that's, you know, had repeated privacy issues. These sort of repeated privacy issues, I believe may actually hurt consumers and also regulator trust. So the other thing to bring up, I think, is that details are still kind of unclear about exactly what types of data Zoom is using. its AI for its AI systems. So metadata only, or, you know, also submitting content. So its communications have also been vague on that. EU, here's one other thing also that I think to take to note, is that EU regulators recently required OpenAI to offer opt-out for chat GPT data processing. So, you know, though for chatbots, consent may be the only legal basis.

8:19I think This suggests EU authorities see limits to using data for AI training. One other thing to bring up is that with video and voice data considered highly sensitive, EU DPAs may take a strict view that Zoom needs opt-in consent for its meeting summaries or transcription AI. And then the other thing is just the fact that Zoom is, you know, it's a non-EU company. And so Zoom risks enforcement by any of the bloc's regulators, unlike big tech firms with the lead authority like ireland's dpa and so i think that is just more exposure um it all in all i think you know while trying to really quickly match ai competition zoom could draw some pretty swift regulatory action in the eu if it handles consent carelessly or if it doesn't tightly limit data usage to just what's needed for core services and i definitely think clearer communication controls is going to help zoom to avoid this whole privacy storm in the future so this is going to be a very interesting area to watch not just for zoom but for other companies trying to do similar things in the future definitely something i'll keep you updated on

From the publisher

In this episode, we unpack the controversy surrounding Zoom as it comes under fire for training AI on users' video calls without providing an opt-out option, exploring the implications for privacy rights and ethical considerations in the age of pervasive digital surveillance.

See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.

More from AI Today

All 897 episodes
Legal Storm Brewing: Zoom Faces Backlash for AI Training on Video Calls Without Opt-OutAI Today · 9 min
Listen in VO