Why AI is taking over social media (and how to opt-out)
From AI-generated posts to large language models collecting your profile’s data - artificial intelligence is in every facet of social media. But why is this happening, and can you opt-out?
Facebook, LinkedIn, Reddit, you name it: It seems that every large online platform has implemented some kind of AI. One can’t help but notice it’s taking over and invading every aspect online, notably on social media. While ChatGPT or DeepSeek might be your go-to tools to boost productivity, the implications of these large language models (LLMs) invading your life on social media is far reaching.
When you think about why AI is taking over social platforms there are multiple reasons: 1. Your online data is a gold mine used to train these artificial intelligence tools, from LLMs to image creators – they all depend on large amounts of data for training. 2. Big Tech has to implement AI to survive. Why? Because everyone seems to be doing it, so the AI race is on. 3. AI creates social media content much faster than humans.
While AI helps with productivity, it is a nightmare for your data privacy, particularly if your social media data is scraped. Let’s review why this AI takeover is happening, what popular platforms use it, if and how you can opt out, and lastly let’s look at better privacy-focused alternatives – which have no business in AI, like Tuta Mail.
Why is AI taking over social media?
Your social media data is a gold mine for AI
We already know that our data is used for targeted ads, and every Big Tech profits from our data. That’s why Big Tech platforms offer us services for free – the worst offenders are Google and Facebook. Through collecting our personal data, third parties and ad agencies can curate advertising that targets us. But now, our precious data is not only valuable for advertising, but it’s also for artificial intelligence training.
AI needs extensive training, and it’s only able to reproduce great results because it gets trained from content created by humans – but how do they get the data? Scrapers and web crawlers, crawl the web collecting your data and use it to train their data sets.
If you’ve had a Facebook account for 10 years, or even LinkedIn just for two years everything you have ever posted, shared, clicked, liked, or watched is highly valuable data that companies use and sell for training large language models. And this is not a secret, Meta has even confirmed that its AI was partially trained on public Facebook and Instagram posts.
AI is faster: AI-generated content saves time
Another way AI is taking over social media is through artificial intelligence generated text (AIGT), images, artworks, articles, and even through AI generated spam. AI content on social media is out of control and it’s constantly increasing. It’s scary to think that each time we scroll on Facebook, Instagram, or any other social media platforms, a huge proportion of the texts, images, and even videos we watch are likely created artificially.
There are many reasons AI-generated content is taking over: for businesses and marketing agencies, using AI to create and share contents faster and saves time. Others might use AI content because some algorithms prefer it and boost this type of generated content.
Social platforms using AI
We already know that artificially generated content is everywhere, but which popular social media platforms are using user data for AI purposes? We explain below!
Meta
Tech giant Meta, has openly said that it may use public Instagram and Facebook user’s data to train its AI models.
Meta has publicly said that it may train its AI with public Facebook and Instagram users’ content such as posts, profile pictures, and comments. Luckily, users within the EU and UK are exempt from having their data scraped for AI purposes for now, as Meta paused its rollout of using user data and introducing Meta AI within Europe after it faced massive backlash.
By default, users of the job networking platform LinkedIn were opted into having their data used for AI purposes.
LinkedIn opted all its users in by default – without asking - when it started using their data to train its generative AI models. Instead of opting in actively, users need to do the opposite: go to the settings and choose to opt-out. It’s also worth noting that anything LinkedIn has already scraped, cannot be undone. If you’re situated in the EU, UK, or Switzerland, for now this does not apply due to stricter data protection laws.
X
X is another Big Tech hungry for your data! It also uses your data to train its AI chatbot known as Grok.
X (formerly Twitter) came under fire for quietly updating its privacy policy to give itself permission to use user data to train its AI chatbot, Grok. So like LinkedIn, users of X must actively opt-out if they do not want their posts used to train the AI chatbot.
Anything shared publicly on Reddit can be used by third-parties for AI purposes.
Reddit users should be aware that anything you share publicly on the platform is granted under a free, worldwide license and there’s no option to opt-out from AI data scraping. This means that Reddit is free to share any public data with third-parties for AI training purposes. In 2024, Reddit signed two massive deals with OpenAI and Google, allowing these big techs to train their AI models with Reddit users’ data. This highlights how even if the company isn’t using your data for its own AI tools – there’s a high chance it sells your data to others.
AI even crept its way into your Gmail
Alphabet-owned Google also has a range of AI tools. From Google Gemini, Gmail AI features, AI Teammates to Project Astra – Google is developing all kinds of artificial intelligence tools. In 2023, Google updated its privacy policy allowing it to use all publicly available information from the internet for the purpose of training its AI tools Gemini, Cloud AI, and Google Translate.
And while these tools might make life easier, we will never truly know where Big Techs like Google who are sitting on gold mines of personal data are getting their data from to train their AI models. This huge amount of data used for AI training gives them a competitive edge that smaller companies can only dream of – if they want to jump on the AI train.
Google’s Gemini uses publicly available sources and the app’s information to learn and improve its AI. While they have said they use “publicly available sources” users have experienced otherwise like in the case where a Reddit user shared that Gemini was getting its replies from scanning their mailbox without having been given permission!
Google’s Gemini scrapes information from publicly available sources as well as the Gemini App’s information for AI learning and improvement. But when it comes to the question what “publicly available sources” are used specifically, the lines are blurred and the public remains in the dark and uninformed.
In 2024, a user on Reddit even shared their conversation with Google Gemini which produced proof that the data Gemini used came from their private Gmail emails – even though the Reddit user had never granted the LLM access. When incidents like this come to light online, it shows that what we are told by tech corporations versus what they do behind the scenes does not necessarily add up.
The bottom line: The race for Big Techs to become the most advanced in AI is on. Judging from past behavior, how these companies treat user data, and how far they will go to profit, it would be no surprise if in the future it’s exposed that Big Tech has used private data for AI training all along.
How to opt-out of AI invasion
Opting out of AI on social media is tricky. There are so many aspects to this as anything available online can be used – for example, if someone with a public Facebook profile shares a photo that you’re in, it may be used. On top of that, so many platforms are sneakily updating their privacy policies to opt users into their data being used for AI without explicit consent. Ultimately any type of public presence you have online is an invite in the eyes of Big Tech to use the data for AI training. So if you have profiles on any social media platforms, it’s very likely your data has already been scraped.
If you really do want to reduce the amount of your data getting used for AI purposes, we’d recommend reviewing you privacy and data settings on each platform. Be warned, Big Techs are professionals when it comes to ambiguous wording, so read the fine print carefully and opt-out where you can – or consider deleting your social media profiles completely.
The need for privacy-respecting alternatives
Unfortunately, not everyone is treated equally when it comes to how their data is used. Countries within the EU have the strict GDPR data protection regulation in place which actively protects citizen’s data from exploitation by tech companies and their AI systems. Other countries, like the USA do not. And because of this online users are at greater risk of having no choice of how their data is handled: for example, they might not be given the option to choose whether to opt into their data being used for AI purposes. Because there is not sufficient protection of user data, and tech giants will always put profits first, the best thing one can do to better protect their privacy and how their data is handled is to move away from Big Tech, and opt for better privacy respecting alternatives.
On our blog we have many comprehensive guides to help you ditch Big Tech and regain your privacy - whether it be de-Googling, deleting Gmail, or choosing a better Whatsapp alternative.
Tuta’s commitment to privacy: saying no to AI
Tuta is the opposite of Big Tech. As a privacy-focused email and calendar provider offering advanced quantum-proof encryption, Tuta has no business in AI.
When you choose Tuta, you will get email and calendars as they should be – private, secure, no AI scraping, no data collection. Just easy-to-use, private, encrypted email and calendar.
Learn more about Tuta - the most secure email and calendar provider, that has zero intention to use your private data for AI training!
