AI Forensics (previously https://nitter.cf/t.co/gtSmmruTMC) is a European non-profit that investigates influential and opaque algorithms.
Europe
Joined November 2016
- Tweets485
- Following751
- Followers2.4K
- Likes299
Pinned Tweet
In light of recent developments, we will be increasing our presence on Bluesky and can be found there via @aiforensics.bsky.social
Stay tuned.
We’ve received many requests in recent days for updates on our Grok report. We’re currently conducting additional monitoring and analysis and will have more to share shortly. 🔍
aiforensics.org/work/grok-un…
⚠️ AI chatbots are reshaping how we search. But EU rules haven’t caught up.
Our new policy report digs into the risks of LLM-powered search and the gap in Europe’s regulatory approach.
So how do we fix it?
👇 Full report + recommendations: tinyurl.com/2p2h77et
🚨354 automated AI accounts generated 4.5 billion views on TikTok through more than 43,000 AI-made posts.
Our investigation finds Agentic AI Accounts spreading harmful content. Yet TikTok labels only around 1.38% of these posts. ⚠️
Full report: tinyurl.com/4kkfm4nn
Last month, w/ @mozilla @algorithmwatch and others, we asked the top platforms to share their 1,000 most-viewed posts.
Not one agreed.
If we can’t see what goes viral, we can’t know how misinfo spreads.
Sign the petition for basic transparency.👇 mzl.la/49HTh8C
📢 New: TikTok and YouTube failed to address dangerous climate dis/misinformation during Spain’s 2024 floods
Our study with @maldita reveals that such videos remain online a year after the event and appear to have been amplified, receiving over 21 million views. 👇
These findings reinforce what we already know: dis- and misinformation are profitable commodities for platforms built around engagement. While YouTube must adopt a policy to tackle climate disinfo, TikTok must do more to enforce its existing one. 👇
It’s also time that platforms offer researchers meaningful access to algorithmic data to study amplification and monetization — both still opaque, and in need of greater scrutiny.
Read the full report: tinyurl.com/yf53jhn6
Europe is rapidly rolling out age-verification — in the name of “child safety.” 🚷 But is it working?
Our first audit found AgeGO’s selfie feature streamed biometric data straight to Amazon. 🪪
Our new report examines AgeVerif: What we found raises serious concerns. 👇
Key findings on AgeVerif:
• Only ~7% minors in the training data → ~1 in 10 minors still get through
• Black representation is extremely low (~6%) → higher misclassification for darker-skinned minors
• Age thresholds shift arbitrarily: 20 🇫🇷 / 25 🇩🇪 / 23 elsewhere 👇
AI Forensics retweeted
📣7amleh has released a new report titled “Digital Rights Under Threat: The Impact of LinkedIn’s Biased Moderation Amid Genocide.” The report documents, through fifteen user testimonies and interviews with LinkedIn and Microsoft employees, the violations resulting from biased content moderation practices and their impact on Palestinian human rights defenders amid Israel’s ongoing war and genocide in Gaza.
The report reveals serious gaps in LinkedIn’s moderation policies and an internal work environment marked by institutional bias in favor of Israel and anti-Palestinian racism. This environment has enabled violations of standard moderation procedures and the restriction of human rights and humanitarian content supportive of Palestinians.
🔗Read Report: 7amleh.org/post/digital-righ…
🚨 NEW RESEARCH! Could your TikTok’s For You Page be amplifying a pro-military narrative?
That’s the question we ask in our new report “From FYP to WW3.”
🧵👇
As debates over NATO’s role and military spending continue — and TikTok becomes a space for political engagement — users should ask:
👉 Which narratives are platforms amplifying?
👉 And which are they downplaying?
Have you been targeted on X based on your sensitive characteristics?
We built haveibeentargeted.online so you can upload your X data and see whether advertisers targeted you based on your sensitive characteristics on X.
📊 The scale is massive: These ads generated hundreds of millions of impressions across the EU, creating a system where your most personal characteristics determine what information you see. It's algorithmic discrimination at industrial scale.