Welcome back, digital enthusiasts, to your weekly dose of all things social! With the summer holidays just around the corner and temperatures climbing toward a sizzling 30 degrees, things are naturally getting a bit quieter around the office. It’s the perfect time to settle into a more relaxed routine and enjoy a summer of sport.
But the social media landscape definitely hasn’t taken a vacation. This week brings a heavy dose of Meta updates, public outcries, and regulatory crackdowns. We are unpacking a controversial AI tool that was axed in days, shocking LinkedIn ghostwriting stats, upcoming paywalls for Instagram, massive EU fines, and new transparency tags from Google.
Here we go, yo! ⏬
Subscribe now to our weekly social round-up
Latest News
🛑 Meta Abruptly Pulls Controversial “Muse Image” AI Tool After Privacy Backlash
Just days after launching its new Muse Image AI feature, Meta has completely shut down the tool following a massive wave of public criticism over severe privacy violations. The feature allowed public Instagram photos to be scraped automatically (without explicit user consent) to let others generate lookalike AI images, and users found themselves opted-in by default. Activists spent the week urging users to manually opt-out to protect their likenesses. Meta issued a formal statement admitting that the feature heavily missed the mark, leading them to completely remove it from the platform.
👉 Read the full article on NU.nl
🤖 Report: Over 40% of Long Posts on LinkedIn are Written Entirely by AI
If your LinkedIn feed has been feeling a little robotic lately, you aren’t imagining things. Data from browser extension developer Pangram reveals that an astonishing 41% of long-form posts on LinkedIn are suspected to be written entirely by generative AI. Short LinkedIn updates aren’t far behind at 30%, meaning only about 55% of long posts are purely human-made. By comparison, AI text generation rates hovered around 30% on platforms like X and Medium, while communities on Substack and Reddit saw a significantly higher share of human-classified writing, particularly in Reddit’s comment sections.
💸 Instagram Chief Confirms Users Will Have to Pay for In-App AI Features
In a weekly Q&A session on Instagram Stories, Head of Product Adam Mosseri confirmed that Instagram is shifting toward a consumer subscription model for its upcoming creative AI tools. Pointing out that generative models for image and video are incredibly expensive to run on servers, Mosseri noted that the platform will eventually be forced to either throttle user caps or ask people to pay. Instagram is already implementing usage limits on its Muse-powered image effects, telling capped users to subscribe to continue. This lines up with Meta’s broader push for paid upgrades, including the newly introduced “Meta One” package which grants unlimited Conversation Focus tools to paying Meta AI glasses wearers.
⚖️ Meta Faces Catastrophic European Fine Over “Addictive” App Layouts
An investigation by the European Commission has concluded that Meta is in direct violation of digital consumer laws due to the intentionally addictive designs of Facebook and Instagram. Brussels explicitly called out manipulative mechanics like infinite scroll, auto-playing videos, and hyper-personalized push notifications for trapping brains on “autopilot” and forcing compulsive habits on minors and vulnerable adults. The Commission stated that Meta’s existing screen-time limits are too easily bypassed by teens to be effective. Meta now has a window to legally defend its design, but faces a potential maximum fine of up to 6% of its global annual turnover if it fails.
👉 Read the full article on NU.nl
🔍 Google Deploys “SynthID” Watermarks to Disclose AI-Generated Ads
Google has introduced an automated transparency feature to identify and label promotional content made with generative artificial intelligence. Leveraging its proprietary SynthID process, a digital watermark embedded directly into the media asset data, and C2PA metadata signals, Google’s system will automatically detect when a brand uses AI tools to design an ad. Viewers can click the three-dot menu on any ad to expand a “how this ad was made” information panel within My Ad Center to check its AI status. Google is also launching a manual control toggle so off-platform advertisers can easily self-disclose AI use.
🗓️ Newsjacking opportunities
Opportunities to celebrate this week:
And next…
Anything else you would like to discuss and see for next week’s social newsflash? Just share it with me via e-mail or slide in my LinkedIn comments or DM.
And don’t forget to subscribe to our Weekly Social Round Up newsletter so you get this delivered directly into your inbox each week!
Until the next one. XoXo, Erik 👋
Subscribe to our Weekly Social Newsletter and receive the weekly trend straight to your email.
Hulp nodig bij het bereiken van jouw doelen?
