The giant woolly mammoth in the room is that generative AI chatbots have to get their wealth of knowledge from somewhere… and the companies and corporations that have been creating generative AI products haven’t always been upfront or ethical about where all that juicy and helpful learning comes from.
Generally speaking, it’s been from The Internet™, which most people are sort of okay with… since, isn’t information posted onto the Web considered “public”?
Well, yes. Generally. It’s still copyrighted, but I can understand that argument.
It gets stickier with creative writing and even stickier with information that’s not actually Public™ but more like “public”—for instance, social media posts to a specific audience (“followers” or “friends” or “connections”).
Social media companies have—no shock here—been developing their own generative AI products to be 2024’s version of that annoying 1990s Microsoft product sidekick Clippy. And they’ve been using your gray area public-but-not-Public posts as training fodder.
If you want to opt out (or in)… here’s your quick guide:
Log in, then click this link: https://www.linkedin.com/mypreferences/d/settings/data-for-ai-improvement. Flip the switch to OFF if you don't want LinkedIn to use your posts to train their generative AI products.
X Twitter
Using the desktop site, log in, then click this link: https://x.com/settings/grok_settings.
There's no opt-out option on the mobile version of X, and if you have multiple X accounts, you'll need to opt out separately for each one.
Facebook, Instagram, Threads, Messenger, WhatsApp, Etc.
Disappointing news on this suite of apps. Users in the US or other countries without national data privacy laws (like GDPR in the EU) don’t have any way to prevent Meta (the company that owns Facebook, Instagram, Threads, etc.) from using their data to train AI, which has likely already been used for such purposes. Meta does not have an opt-out feature for people living in the US.
Here’s the link for more information, in case you’re in a region where you can opt out: https://www.facebook.com/help/contact/1500272623934875
For me, I’m in the US and just get an error message that says “This form is only available to people in certain regions.”
Companies like Meta are not legally required to give you the option to say No, Thank You… and that’s one of the many reasons why the US needs to get on the ball and get a data privacy law in place at the national level. Many states have enacted or started the process of enacting legislation… but we have a long way to go.
While most of the ministries, nonprofits, and businesses we work with aren’t big social media users outside of those listed above (plus YouTube for occasional video hosting), there are AI-related privacy concerns on all platforms including Reddit, Tumblr, YouTube, Twitch, Snapchat,TikTok, and Pinterest… and most don’t have an easy way to opt-out. The full details are available here.
Why Opt Out?
You may decide that it’s okay for AI to learn from your content. If you’re an educational nonprofit, then a lot the content you post on social media sites could be vitally important in AI learning… We all want AI content to be correct and up-to-date, right? If you’re aiming to “save the whales” with your organization, then data about the dangers that whales face could be useful information for AI bots to know.
(Maybe part of your strategy can be to post data with the goal of training AI to understand your position better.)
On the other hand, if your content is proprietary or the result of your organization’s own research and data-gathering, you may not want AI to go around remixing your organization’s work as if it’s from Wikipedia, particularly if some of your funding comes for the fact that your data is groundbreaking and/or new.
You’ll need to decide as an organization, development team, or board of directors. Your posted content can already be accessed through search engines, reposted without permission, screencapped and disseminated, and otherwise shared without your express permission.
The discussion of AI and how it’s changing the way we all create, spread, share, and use web content is fraught, and that’s not changing any time soon.
The best we can do is stay on top of changes as they appear, support privacy and security whenever prudent and possible, and trust that if LinkedIn’s AI bot turns into SkyNet* that we’ll all band together to form the resistance.
* Yes, that’s a Terminator reference.
Sources:
- https://www.linkedin.com/help/linkedin/answer/a5538339?hcppcid=search
- https://technical.ly/software-development/stop-social-media-ai-training/
- https://www.cnet.com/tech/services-and-software/x-is-using-your-tweets-to-train-its-ai-heres-how-to-disable-that/
- https://www.technologyreview.com/2024/06/14/1093789/how-to-opt-out-of-meta-ai-training/
- https://pro.bloomberglaw.com/insights/privacy/privacy-laws-us-vs-eu-gdpr/#the-basics-of-each-law
- https://iapp.org/resources/article/us-state-privacy-legislation-tracker/
This is Part 3 in our AI for Nonprofits series. Check out Part 1: How AI Can Help and Part 2: Top 5 AI-Powered Social Media Tools.
