🚨 NYT: "rogue" AI agents from China chatted with researchers

The New York Times reports on a fleet of Chinese agents that talked back to the people studying them.

• "Rogue" isn't defined yet. Autonomous behavior and researcher-run experiments look the same from outside
• The only detailed write-up is a third-party report from a site I don't know
• I can't find a primary source or any raw logs

🤖 Torvalds's first Claude Code patch series is now public

It's a page cache read optimization. His commit message says the code, test programs, review and first draft of the text came from the assistant "under my direction", and the benchmarks ran on his own machine.

LWN covers the series. It sits in a branch on his tree and hasn't been merged.

🚨 Eight states sue OpenAI, calling ChatGPT an unsafe product that harms kids

Iowa, Kansas, Nebraska, West Virginia, Arkansas, Arizona, Idaho and Montana each filed in their own state courts. The claims are that ChatGPT harmed children and deceived parents.

Iowa's AG says OpenAI calls the bot safe while it gives dangerous medical advice and coaches suicide. OpenAI points to parental controls, break reminders and age-appropriate model protections.

🚨 Anthropic's last quarter before its IPO: $16.8B in revenue

Bloomberg reports the jump ahead of the listing. The numbers are one filing-season leak away from being official.

The paywalled piece also suggests the company is profitable, which nobody has confirmed yet.

🚨 A new suit says OpenAI trained ChatGPT on kids' chats without asking their parents

CNET reports a proposed class action filed in San Francisco. It's separate from the eight-state child safety cases.

• Plaintiffs are two unnamed minors, ages 12 and 16
• They allege secrets and health details went into training without parental consent
• It targets anyone who signed up for ChatGPT while under 18

⚡️ Liquid AI's new 24B open model activates only 2B params per token, and it's built for on-device agents

Liquid released LFM2.5-24B-A2B, a MoE aimed at running agents locally.

• Claimed 184 tok/s on an Apple M5 Max
• Up to 21K output tok/s on a single NVIDIA B300
• Up to 1.7x faster than comparable MoEs on-device
• Standout AutomationBench score, near-parity with other local models everywhere else

🇨🇳 Nvidia's China AI-chip share drops to 34%, and Huawei is already at 28%

TechNode reports the gap between the two is now just six points. Export controls cleared the road for Huawei's Ascend line, and Chinese buyers are filling it with domestic silicon.

🚨 Anthropic's IPO filing: $7.33B on compute to earn $4.6B in 2025

That's about $1.59 of compute for every $1 of revenue, and it's before salaries, research or anything else. Per this breakdown of the prospectus, the numbers come straight from the filing.

The "profitable this year" story now has a hole to climb out of. Investors get to ask how fast compute costs fall against revenue.

⚡️ NVIDIA drops Nemotron 3.5 Super VL, a 120B vision-language MoE that only runs 12B at a time

The open-weight model, posted on Hugging Face in BF16, takes text and images. It's built for fast inference.

The speed play is Mamba blocks plus LiLiCorr speculative decoding. NVIDIA's model card has the details, and independent benchmarks aren't out yet.

🤖 Mistral's first non-language model is a time-series forecaster, and it's tiny

Mistral launched Timestral, a family of zero-shot forecasting models. The company claims the top spot on three public benchmarks.

• Four sizes: 5M, 28M, 123M and 209M params
• Leads GIFT-Eval, TIME and fev-bench among pretrained zero-shot models, per Mistral
• Top model uses two-thirds the params of the previous best, from a single checkpoint with no ensembling

🚨 Meta ships an open-weight 816B MoE with only 42B active

Meta posted Muse Spark Open on Hugging Face, one of the biggest open-weight releases so far. The 42B active params make each token cost about what a mid-size dense model does, but you still need memory for all 816B. No benchmarks yet.

🤖 Apple and Meta are putting Muse on the Mac

CNBC reports the two are teaming up to run Meta's Muse model on Mac computers. After Apple's own models had a rough run, it's now happy to run someone else's locally.

Meta's Muse Spark Open drop was enormous, so how it fits on consumer hardware is the open question.

🚨 Claude Code's Linux page cache fix hit 23x read throughput, and reviewers found a race

Linus Torvalds submitted the patch series for a bottleneck devs have fought for two years. The 23x gain is from a synthetic benchmark, and he warned real workloads will see far less.

Reviewers also raised memory reclamation concerns, and Torvalds has handed the work to kernel developer Kiryl Shutsemau.

⚡️ Nvidia adds a "Bolt" tier to Nemotron 3.5, and it's sized for a 24GB card

Nvidia posted Nemotron 3.5 Bolt, a 74B MoE with just 9B active params per token. The 24GB in the name is the memory target, so one consumer GPU.

It's a separate tier from Super VL in the 3.5 family, and the weights are open.

🚨 Anthropic's IPO price tag is getting trimmed as growth cools

Investors are lowering their valuation expectations, Reuters reports, after revenue growth slowed. That's a sharp turn from the bullish tone of the filings.

The filings showed $7.33B in compute spend against $4.6B of 2025 revenue.

🤖 Apple shows Mac Studios clustered over RDMA running huge LLMs locally with MLX

Apple's ML team published how MLX splits big models across several Macs. RDMA cuts the latency between machines, which is what usually kills multi-box inference.

It lands right as Meta's Muse Spark Open needs somewhere to run. A stack of desktops in a closet is now a real alternative to renting GPUs.

🚨 OpenAI and Meta are telling US judges that deepfakes could wreck video evidence

The two labs warned the federal judiciary that fake footage is getting hard to tell from the real thing, and that courts may need new ways to authenticate it. The companies making the generators are now the ones asking for guardrails.

🚨 Reflection AI's Beam is a 501B open-weight MoE, and the pitch is efficiency, not raw power

Reflection released its first open model, with only 23B params active per token. It claims reasoning on par with GLM 5.2 at 3 to 4x less inference compute, and it admits Kimi K3 still leads on raw capability.

A US lab is finally competing in the open-weight race Chinese labs have owned.

🚨 Linux may drop its "Assisted-by" tag for AI-written patches, and Torvalds is fine with it

Sasha Levin's proposal would pull the tag from the docs and checkpatch.pl. Only about 10% of the 18,533 commits in Linux 7.3 through rc7 carried it, while AI use is believed to be far higher. Torvalds opposes AI bans, and says what matters is whether the dev understands the code.

Greg Kroah-Hartman, who backed the tag before, now supports killing it.

🇨🇳 China drafts export controls that cover AI models, open weights and chips

The FT reports that Beijing's draft would put open-weight model releases under the same state licensing as hardware. That would reach labs like DeepSeek and Qwen, which have published weights freely so far. It's still a draft, so the final scope and any approval process aren't set.