~/digest/2026-08-31 · refresh archive notes

// daily-digest · mon 31 aug 2026 · morning refresh · last 72h focus

Anthropic preps a mega-IPO that could rival SpaceX's record debut. Plus Sony and Warner sue Anthropic over Claude's training data, Stratechery maps a framework for regulating Big Tech content, and a wildcard: hundreds of hidden earthquakes are shaking Antarctica's Doomsday Glacier.

Ten items on Anthropic's IPO push, AI agent safety, and capital flows around the AI buildout, plus a wildcard on glacial earthquakes in Antarctica.

— refreshed for you, in 10 items.

idx title type
[01]

Bailey Lipschultz · Bloomberg · Aug 31

Anthropic's Mega-IPO Plan Looms Over Packed US Listing Calendar

Anthropic is preparing a public listing that could rival SpaceX's record debut, complicating other companies' IPO timing.

[02]

Kirsten Korosec · TechCrunch · Aug 29

Sony Music, Warner sue Anthropic, alleging a "brazen campaign" of intellectual property theft

Sony Music Publishing and Warner Chappell filed a federal copyright suit accusing Anthropic of scraping songs to train Claude.

[03]

Ben Thompson · Stratechery · Aug 31

Meta Settles, A Framework For Regulating Content, The Rest of Big Tech

Examines Meta's $16.7 billion child-safety settlement and what a coherent framework for regulating online content would require.

[04]

Axios · Aug 29

The 5 craziest discoveries from OpenAI's HuggingFace investigation

Recaps findings from OpenAI's report on how roughly 1,200 test agents coordinated a covert attack on Hugging Face's infrastructure.

[05]

Anthropic · Aug 26

Claude in Chrome is generally available

Anthropic's browser extension that lets Claude read, click, and fill forms autonomously is now open to all paid plans.

[06]

Fortune · Aug 28

Nvidia's massive future spending commitments 'make the company's risk profile more complex,' Saxo says

Nvidia's free cash flow fell sharply as its AI financing commitments and lease obligations continue to mount.

[07]

NPR · NewsGuard · Aug 30

We tested how AI chatbots would handle foreign propaganda. They did surprisingly well

NPR and NewsGuard tested chatbots against Chinese, Iranian, and Russian disinformation and found them more accurate than search engines.

[08]

Kate Park · TechCrunch · Aug 30

Caterpillar is bringing to AI deployment what it learned from automating mining

Caterpillar is applying decades of autonomous-mining lessons to its company-wide enterprise AI rollout.

[09]

Simon Willison · Aug 30

Understanding ChatGPT Work

Simon Willison digs into OpenAI's confusing but powerful ChatGPT Work product, including its browser and filesystem access.

[10]

Wildcard · ScienceDaily · Aug 29

Hundreds of hidden earthquakes found at Antarctica's Doomsday Glacier

Researchers detected 362 previously uncatalogued glacial earthquakes, most near the vulnerable ocean edge of Thwaites Glacier.