What We’re Reading

We keep an eye out for the most interesting stories about Labby subjects: digital media, startups, the web, journalism, strategy, and more.

TechCrunch / Amanda Silberling

Why is AI so bad at spelling? →

“If you ask text-to-image generators like DALL-E to create a menu for a Mexican restaurant, you might spot some appetizing items like ‘taao,’ ‘burto,’ and ‘enchida’ amid a sea of other gibberish.”

What else we were reading on March 21

Financial Times / Daniel Thomas

BBC develops AI plans and talks to Big Tech over archives access →

“The BBC is making plans to build its own artificial intelligence models, while holding talks over selling access to the broadcaster’s vast archives to Big Tech groups developing the cutting-edge technology.”

NBC News / Daniel Arkin

Crime stories drove readers to GoFundMe campaigns, only the victims didn’t exist →

“The articles published by Blast News 365 would most likely raise eyebrows among anyone who regularly reads professionally produced news content. The articles were riddled with typos, grammatical errors, off-kilter formatting and other red flags. They were also filled with discrepancies. The father killed in the hit-and-run is alternately identified as ‘Herman Cruz’ and ‘Henry Cruz’ in the span of just two paragraphs, for example.” —LO

Substack / Richard J. Tofel

What went wrong at the Center for Public Integrity? →

“What to make of this, beyond mourning the loss of CPI itself, and its good journalism? A great deal of the focus, as I hope the narrative above reflects, should be on a Board that seems to me, fairly consistently, not to have provided the necessary leadership.” —LO

404 Media / Jason Koebler

Deadspin is becoming a gambling referral site →

“A poorly protected IP address suggested that the new owner had ties to the online gambling industry.” —LO

WIRED / Kate Knibbs

Here’s proof you can train an AI model without copyrighted content →

“OpenAI claimed it’s ‘impossible’ to build good AI models without using copyrighted data. An ‘ethically created’ large language model and a giant AI dataset of public domain text suggest otherwise.” —SS

The Verge / Ariel Shapiro

TikTok’s podcast boom might be a bust →

“Even when a video gets multiple tens of thousands, I would be lucky if I got more than single-digit numbers of clicks in a day.”

WIRED / Steven Levy

8 Google employees invented modern AI. Here’s the inside story. →

“They met by chance, got hooked on an idea, and wrote the ‘Transformers’ paper — the most consequential tech breakthrough in recent history.” —SS

The New York Times / Michael M. Grynbaum and Mike Isaac

Condé Nast’s owners are set to reap a $1.4 billion windfall from Reddit →

“Reddit is a far cry from the meticulously curated guides to haute living in the Condé Nast stable. But its public offering will reward an early and prescient bet on the company by the Newhouses, who own roughly one-third of the outstanding shares.” —AD