Newsletter

Editorial

Flock slashes plate-camera data retention to 7 days after abuse reports

The Everyday

Flock slashes plate-camera data retention to 7 days after abuse reports

America's largest license-plate surveillance network is tightening its own rules after a string of reports that police officers used the system to stalk people — and its CEO is conceding the company got the balance wrong. Flock Safety is making its anti-abuse safeguards mandatory and cutting default data retention from 30 days to seven, after reports of cops across the country using its license-plate cameras to stalk ex-romantic partners.

DeepSeek hikes API prices up to 4.7x with new peak-hour billing

The Arena

DeepSeek hikes API prices up to 4.7x with new peak-hour billing

The price war's last holdout just posted its new rate card. DeepSeek has turned its promised "significant" API increase into hard numbers — and, for the first time, into a peak/off-peak billing model that charges more than double for busy-hour inference. DeepSeek has finalized its first major API price increase, with output rates rising roughly 2.4x to 4.7x and a new peak/off-peak billing structure taking effect at 16:00 UTC on August 16.

Claude settles every open Hadamard matrix below order 2000

The Frontier

Claude settles every open Hadamard matrix below order 2000

A morning of frontier math and agent reality checks: Claude just closed out every open Hadamard matrix below order 2000, Anthropic published its first deep study of what happens when agents interact at scale, and Qualcomm's $3.9 billion bet on AI software is officially complete. Anthropic mathematician Levent Alpöge and two collaborators used Claude to construct every previously unknown Hadamard matrix below order 2000 — twelve in all, led by the order-668 case that had been the smallest…

Anthropic's Fable 5 stalls as corporate AI spending hits a ceiling

The Arena

Anthropic's Fable 5 stalls as corporate AI spending hits a ceiling

Spending data is questioning the core bet of the frontier labs — that each new, pricier model sells itself — while Morgan Stanley warns compute stays scarce for years and the milestones of automated AI research keep falling. Anthropic's most capable model is its slowest seller: Fable 5 accounted for just 6 percent of the tokens companies purchased from Anthropic in its first month, according to spending data from fintech Ramp.

Claude Cowork arrives in Chrome with skills and plugins

The Arena

Claude Cowork arrives in Chrome with skills and plugins

Anthropic's agentic workspace just got a browser home: the Chrome extension that used to read pages for you can now run a full Cowork session in its side panel, skills and plugins included. Anthropic's Cowork agent now runs inside the Chrome extension's side panel, bringing skills, plugins, and connectors into the browser for the first time. Until now, Claude in Chrome could read the page you were on, click, type, and fill forms — fine for research, but the work stopped at the tab.

Open Source Radar — August 13: decks, video & a 14MB agent

The Stack

Open Source Radar — August 13: decks, video & a 14MB agent

The coding-agent parade paused and the daily board filled with makers instead: a deck generator that ships real PowerPoint files, a financial-market foundation model, Lightricks' one-model audio-video release, and a tool-calling agent small enough to live on a thermostat. Four projects worth your attention. PPT Master (Python, ~46k stars) — An AI workflow that turns PDFs, documents, or topics into native PowerPoint decks: real slide masters, native shapes with working adjustment handles…

SSI's first model reportedly trains itself while it thinks

The Frontier

SSI's first model reportedly trains itself while it thinks

The most mysterious lab in AI may finally be about to show its hand. On August 13, a post on X from an account called "Three Strawberries" (三只草莓) claimed Safe Superintelligence Inc. — Ilya Sutskever's famously silent startup — has been building a small reasoning engine around test-time training (TTT): a model that keeps updating its own weights while it works, rather than arriving fully formed from pretraining. SSI has not commented, and the claim is unverified.

Ex-Qwen lead's Pragmatik Labs raises $220M at a $2B valuation

The Arena

Ex-Qwen lead's Pragmatik Labs raises $220M at a $2B valuation

Lin Junyang, the researcher who built Alibaba's Qwen into one of the world's most-used open model families, came out of stealth early Wednesday with a company that has no product, no demo, and a $2 billion price tag. That gap is the story. Lin Junyang, the former technical lead of Alibaba's Qwen family, announced the founding of Pragmatik Labs (语用科技) in Shanghai — backed by a $220 million angel round at a $2 billion valuation, closed before the company shipped anything.

AI 101 — What is a context window?

The Frontier

AI 101 — What is a context window?

A context window is the amount of text a language model can look at at one time — its working memory. It is measured in tokens (roughly three-quarters of an English word each), and it caps how much of your prompt, your conversation, or your document the model can actually see when it writes a reply. Everything you want the model to consider must fit inside that window, or it gets cut off, summarized, or ignored. Why it's in the news Context size has quietly become one of the biggest selling…

NAND flash: AI inference boom lifts eSSDs to 48% of shipments

The Arena

NAND flash: AI inference boom lifts eSSDs to 48% of shipments

The AI hardware story keeps moving down the stack. This week's data from Counterpoint Research shows the shift from training to inference is now reshaping the memory market — and NAND flash makers are posting record revenue because of it. Server-grade enterprise SSDs hit 48% of all NAND flash bits shipped in Q2 2026 — nearly double their 26% share a year earlier — and the AI-driven demand surge pushed industry revenue to a record, roughly five times its level in Q2 2025. According to…

See all