LLMs and Data Poisoning Are Weaponized to Manufacture Consensus(medium.com)
17 points by marketinganai | 7 comments
Medium essay arguing that marketing, power, and data poisoning shape what LLM assistants present as truth, manufacturing consensus.
#misinformation on Hacker News
17 stories tagged misinformation, newest first.
'Pink Slime' Is Infecting AI Chatbots Ahead of the Midterms(politico.com)
7 points by JumpCrisscross | 1 comments
Politico Magazine article on how partisan sites posing as independent local news outlets are shaping AI chatbot answers about midterm races.
What is the size of Yemen? (2024)(theborys.substack.com)
220 points by kspacewalk2 | 72 comments
A Substack post investigating why Yemen's commonly cited area of 555,000 sq km differs from map measurements, tracing a century-old misinformation.
Disagreement among frontier LLMs on real-world fact-checks(lenz.io)
457 points by kostaj | 314 comments
Lenz research report finding that five frontier LLMs disagree on the verdict for 67% of 1,000 real-world fact-check claims.
Rumors of my death are slightly exaggerated
1556 points by CliffStoll | 240 comments
Cliff Stoll's Hacker News post saying an AI-generated Facebook review of The Cuckoo's Egg falsely claimed he died in May 2024.
Knitting bullshit(katedaviesdesigns.com)
419 points by ColinEberhardt | 177 comments
A blog post by knitwear designer Kate Davies criticizing AI-generated knitting images and content, illustrated with an implausible AI image of a woman knitting.
AI assistants misrepresent news content 45% of the time(bbc.co.uk)
369 points by sohkamyung | 263 comments
BBC Media Centre press release on an EBU-coordinated, BBC-led international study finding AI assistants misrepresent news content 45% of the time.
Wikipedia survives while the rest of the internet breaks(theverge.com)
534 points by leotravis10 | 396 comments
A Verge feature on how Wikipedia has endured attacks and remained resilient through its neutrality policies and history, with Jimmy Wales.
Google AI Overview made up an elaborate story about me(bsky.app)
566 points by jsheard | 230 comments
Bluesky post by Benn Jordan showing Google's AI Overview fabricating a story that he released a video titled 'I Was Wrong About Israel' after a trip there.
Learning not to trust the All-In podcast(passingtime.substack.com)
348 points by paulpauper | 170 comments
A Substack essay by Michael Bateman arguing the All-In Podcast made misleading economic claims, questioning whether it reflects incompetence or deception.
Google’s AI thinks I left a Gatorade bottle on the moon(edwardbenson.com)
325 points by gwintrob | 167 comments
Ted Benson's blog post describing how he tricked Google's NotebookLM AI podcast generator by modifying his website to feed it false claims.
How big is YouTube?(ethanzuckerman.com)
693 points by MBCook | 227 comments
Ethan Zuckerman blog post exploring how to estimate the total size of YouTube to provide a denominator for social media research on misinformation.
The Carrot Problem(atvbt.com)
492 points by deadcoder0904 | 267 comments
Essay on Atoms vs Bits about the 'carrot problem': using a plausible cover story, like WWII British claims that carrots improved night vision, to hide a real advantage.
Journalists should be skeptical of all sources including scientists(natesilver.substack.com)
424 points by amadeuspagel | 299 comments
Nate Silver's Silver Bulletin Substack essay arguing journalists should scrutinize scientists as sources, citing prominent scientists' misinformation on COVID's origins.
PoisonGPT: We hid a lobotomized LLM on Hugging Face to spread fake news(blog.mithrilsecurity.io)
369 points by DanyWin | 184 comments
Mithril Security blog post demonstrating how they surgically modified an LLM to spread false facts and uploaded it to Hugging Face to show supply chain poisoning risks.
Wikipedia had wrong Vatican flag for years – now incorrect flags are everywhere(catholicnewsagency.com)
327 points by axelfontaine | 135 comments
Catholic News Agency article on how Wikipedia displayed an incorrect Vatican City flag for years, leading to inaccurate flags spreading widely.
India ruling party's IT cell used AI to show smile on arrested protesters' faces(altnews.in)
653 points by throwaway384629 | 324 comments
Alt News fact-check showing that a viral image of detained Indian wrestlers Vinesh and Sangeeta Phogat smiling was digitally altered with AI.