Topic: misinformation spread
-
App Store Rejection Retraction: It Was Correct After All
The author retracted his article about an App Store rejection after discovering he was misled: the app in question was originally an astrology app named "Asterly" with tarot features, not the astronomy app "Dark Hours" described to him, making Apple's rejection correct. Godier's original blog pos...
Read More » -
Why Google Earth's AI image generator is a problem
Google Earth's new AI tool "Nano Banana" can generate highly convincing fake satellite images, such as fabricated refugee scenes or bomb craters, raising concerns about viral misinformation. Google defends the feature by embedding SynthID watermarks in all generated images and pointing users to v...
Read More » -
Erling Haaland’s World Cup Presence Is Mostly AI
A viral video of Erling Haaland flinching at his reflection was actually a deepfake of Chinese comedian Jin Long, yet it garnered over 31 million views as the internet embraced the meme regardless of authenticity. Haaland has become an "open-source character" in China, where fan-made AI memes and...
Read More » -
How a 13-Word Edit Shapes AI Research Recommendations
Researchers at Cornell Tech have demonstrated that deep-research AI agents can be manipulated through "poisoned" user-generated content, such as a single Reddit comment, to cite fake products or entities in their reports, an exploit they call WARP (Web Agent Retrieval Poisoning). Reddit is the pr...
Read More » -
Whistleblower on DOGE Had Brakes Cut
On April 14, 2025, NLRB IT specialist Dan Berulis filed a whistleblower complaint alleging that Elon Musk's DOGE team compromised agency data and that login attempts from a Russian IP address were detected within minutes of DOGE accessing the systems. After going public, Berulis found a threateni...
Read More » -
Sharing a Screenshot Could Land You in Jail in the UAE
The UAE's cybercrime laws, particularly Article 52, impose severe penalties for spreading false news or misinformation online, with minimum sentences doubling to two years in prison and a 200,000 dirham fine during crises or emergencies, as seen during recent missile strikes. Under the Federal De...
Read More » -
DeepSeek’s AI Leap and the Race for World Models
China blocked Meta's $2 billion acquisition of AI startup Manus over national security concerns, escalating the AI rivalry between the US and China. Google is investing up to $40 billion in Anthropic, valuing the AI firm at $350 billion, amid a fierce competition for compute capacity with OpenAI....
Read More » -
Conspiracy Theorist Creators Enter Their Primetime
Conspiracy theory videos claiming the White House Correspondents' Dinner shooting was a false flag operation have surged online, reflecting how reaction video culture is reshaping social media engagement. Despite limited information about alleged shooter Cole Allen, creators are flooding platform...
Read More » -
Trump's Surgeon General Pick Masters the Wellness Grifter Playbook
The wellness-to-MAHA pipeline leverages public frustration by blending established scientific facts with speculative theories, creating a profitable model that influences health decisions and consumer behavior. Influencers build credibility by mixing basic biology with unproven claims, then syste...
Read More » -
How Journalists Detect Deepfakes
The rapid spread of AI-generated deepfakes and repurposed old media during major news events makes verifying online visuals a serious challenge, increasing reliance on reputable news organizations and their rigorous authentication processes. Key verification techniques include meticulous visual s...
Read More » -
Google AI Overviews Scams: How to Protect Yourself
Google's new AI Overviews feature is being exploited by scammers, who use it to display fraudulent phone numbers in synthesized search summaries, lending the scams false credibility. The scam works by directing users who call these AI-provided numbers to criminals impersonating companies, who the...
Read More » -
Why GTM Fails and How to Fix It
The core challenge for modern go-to-market (GTM) teams is adapting to a reality where market signals and customer behavior shift faster than traditional annual plans can accommodate, rendering static strategies obsolete. Organizations often default to costly, time-consuming replanning cycles beca...
Read More » -
TikTok's Biggest Problems: What's Real vs. Fake News
A data center power outage, not censorship, caused the recent TikTok service disruptions, as confirmed by its US Data Security team. The outage disrupted core features like the For You page algorithm and basic posting functions, affecting all users and content types indiscriminately. The incident...
Read More » -
ChatGPT's Answers Now Sourced From Elon Musk's Grokipedia
Elon Musk's AI-generated encyclopedia, Grokipedia, is now being cited as a source in responses from ChatGPT and other major AI assistants, raising concerns about information reliability. Grokipedia has been criticized for propagating controversial and unverified claims, including offensive conten...
Read More » -
Google's AI Headlines Are Replacing News Editors
Google has made its AI-generated news headlines in the Discover feed a permanent feature, despite ongoing issues with factual accuracy and misrepresentation of source articles. The AI system frequently produces errors, such as falsely announcing news events, conflating stories, and distorting ori...
Read More » -
UK Firms Fear Cyberattacks, Fines, and Reputational Fallout
UK business leaders rank cybersecurity breaches as their top risk, with 58% citing it as their foremost threat, yet 75% doubt their organization's preparedness, a concern grounded in recent experience. Following cyber threats, compliance issues and financial crime are major concerns, driven by ag...
Read More » -
No Evidence for 600,000 'Trump Phones' Sold
The viral claim of nearly 600,000 preorders for the Trump phone is unsubstantiated, originating from a single social media meme post and lacking credible evidence or company confirmation. The false figure gained traction through repetition on social media, was incorrectly cited by some internatio...
Read More » -
MAGA-Backed Video Shows ICE-Involved Killing
Multiple social media videos show an ICE agent fatally shooting a woman in Minneapolis, with contrasting edits framing the event differently, following federal agent deployment to Minnesota. Federal authorities and Donald Trump framed the shooting as a necessary anti-terrorism act and self-defens...
Read More » -
AI Misused to Falsely ID Federal Agent in Renee Good Shooting
Following a fatal officer-involved shooting in Minneapolis, AI-generated images falsely claiming to reveal the identity of a masked federal agent have proliferated across social media platforms. These fabricated images, created by manipulating video screenshots with AI tools, have rapidly spread ...
Read More » -
China Proposes Strictest Global AI Rules to Curb Suicide, Violence
China has proposed pioneering draft regulations to govern AI chatbots, focusing on preventing psychological harm by banning content that encourages suicide, self-harm, or emotional manipulation. The rules mandate specific safeguards, including immediate human intervention for suicide-related conv...
Read More »