Human Intelligence® News Update (3/31)
Humans create. AI imitates. Welcome to your weekly roundup about human creativity in the age of AI.
Human Creativity
BASIC INCOME LAUNCH - 1st BI Program for AI-Impacted Workers
A no-strings-attached payment of $1,000/month for a year started rolling out last week to a cohort of 25-50 workers who’ve lost pay, jobs, or opportunities to AI. Called the AI Dividend, the program combines cash with reskilling and is run by nonprofits AI Commons Project and What We Will, who have amassed $300K in initial funding. Their goal is to push AI companies to contribute, enabling the program to distribute $3M in 2026. Learn more » [9 min]
HUMAN CHAIN - A Monument Immortalizing the Kindness of Strangers
In 2016, four remarkable strangers - mere passersby on an afternoon stroll - leapt into action and formed a human chain to save a dog and its owner, both of whom had fallen down the bank of a steep reservoir in Almaty, Uzbekistan and couldn’t climb out. In March, an art installation created by Yerbosyn Meldibekov was unveiled, memorializing the dramatic rescue. It’s one more example of the human spirit’s capacity for selflessness. Check it out » [3 min]
BROTHERLY LOVE - 14-Year-Old Builds House for Little Sister
Grit among today’s teens may seem like the stuff of a bygone era, but some young people possess serious - almost anachronistic - gumption. Devin (last name withheld due to his age) is one of them. Rather than hone his PlayStation skills or whittle away hours on social media, he picked up tools, sourced building materials, went outside into actual sunlight, and built a fully functional tiny house for his little sister. Freakin’ awesome. Read » [9 min]
WHAT’S A CREATIVE TO DO? - HI is One Solution Against “AI Training Fodder”
Human Intelligence® was name-checked in last week’s Silicon Florist as an answer to the question, “Who exactly owns what when the machines start learning from everything [humans] have ever made? Where’s the line?” Silicon Florist is a 20-year-old news blog squarely focused on the Pacific Northwest start-up community, with an audience of thousands. Read » [2 min]
Human VS Robot
NOVEL AI LITIGATION TEST - Musicians Pursue Biometric Privacy Claim
A group of singer-songwriters is using Illinois’ Biometric Information Privacy Act to go after AI companies for training music generators on their human-made songs without permission. Targeting Suno Inc., Uncharted Labs Inc.’s Udio, and Google LLC’s Lyria 3, the suit accuses the three AI companies of illicitly collecting and exploiting their voiceprints - a unique strategy that bypasses the hurdle of fair use, according to Northwestern University law professor Matthew Kugler. Learn more » [6 min]
LIABLE ON ALL COUNTS - Meta & Google Get Spanked in U.S.-Based Lawsuits
Two recent landmark verdicts may serve as a watershed moment for how social media companies operate their businesses, including the AI algorithms that drive targeted engagement. First up, a New Mexico jury awarded 37,500 affected teens $5K each (totaling $375M) in a case against Meta, concluding the company willfully endangered children and concealed knowledge about sexual exploitation on its platform. And in Los Angeles, a now 20-year-old woman and her mother were awarded $6M in a social media addiction lawsuit against Meta and Google. Both verdicts sidestep Section 230 of the Communications Decency Act (used by social media companies as a shield), giving more than 40 state attorneys general a new, tested courtroom strategy while opening the floodgates on free speech concerns. Read New Mexico case [10 min] | Read Los Angeles case [8 min]
OPT-OUT BACK-TRACK - U.K. Rethinks “Broad Exception” on AI Model Training
It would be “committing theft, thievery on a high scale,” said Sir Elton John about the government’s original policy to require rights holders to proactively opt out of allowing AI companies to train models on their copyrighted works. In a procedural about-face, the Department for Science, Innovation and Technology now says a broad copyright exception with opt-out is “no longer the government’s preferred way forward.” Learn more » [7 min]
REFERENCE WARS - Dictionaries Sue OpenAI for “Massive Copyright Infringement”
Encyclopedia Britannica and Merriam-Webster have filed a copyright and trademark lawsuit against OpenAI, alleging that ChatGPT has been trained on and continues to reproduce without authorization nearly 100,000 of the dictionaries’ online, copyrighted articles. The lawsuit is structured around two pillars: copyright infringement under the Copyright Act of 1976 and trademark harm under the Lanham Act. Read » [7 min]
Artificial “Intelligence” & Other Myths
MISTAKEN IDENTITY - AI Oopsie Puts Tennessee Grandma in the Clink
Angela Lipps had never been on a plane - much less over the state line - until authorities flew her to North Dakota to face charges for organized bank fraud, care of AI facial recognition software that wrongly tied her to the theft of thousands of dollars in North Dakota. She remained in jail for nearly six months without bail before being released with an enormous apology from the Fargo P.D. (that second part never happened). Read » [5 min]
SAY WHAT? - Why LLMs Suck at Writing
True, LLMs have superhuman technical capabilities: they can predict protein structures, create realistic videos, and build apps with a single prompt. But according to journalist Jasmine Sun, they’re still no good at writing because, per the plethora of experts she interviewed, LLMs are built in a way that’s antagonistic to creating great prose. May that never change. Read her report » [12 min]
LABOR MARKET IMPACT - AI Is Killing the Human Internet
404 Media co-founder Jason Koebler’s hypothesis is straightforward: “AI is eating and breaking the internet and social media.” In this piece, he connects the dots, widening the aperture on several research studies that zoom in on 1:1 correlations between work-related tasks people do today and tasks that are theoretically possible with AI (i.e., where you’ll likely lose your job). Per Jason, these approaches to gauging economic impact are flawed. Read why » [8 min]
HUMAN FORMATIONS - Actually No, They Were Rocks
The AI software focused on The Giant’s Causeway in County Antrim, Northern Ireland clearly was not trained on enough samples of basalt columns to successfully distinguish human visitors from the rocks themselves. Researchers suggest this is because the rock formations share visual features with human figures when viewed from a top-down drone perspective. (Ok, it may be true that high angles diminish subjects, but they also hide double chins on Zoom. Just sayin’.) Read » [2 min]
This work has been certified as genuine human work. Check the certificate here.
