TYPE0
//
TERMINAL
10:50Z · MON, JUL 27 · ● LIVE ·
163 today
▸[T]OP
TOP
[L]ATEST
NEW
[B]UILDING
BUILDING
[D]EV
DEV
·
AGT
·
8
AI
·
50
BIO
·
34
HW
·
32
QC
·
22
ROB
·
17
●
●
Stop Counting Chips, Start Counting Tokens: The New AI Compute Scoreboard
At WAIC 2026 (the World Artificial Intelligence Conference in Shanghai), twelve Chinese AI chip executives agreed on the same shift: the competition is no longer about peak specs, but how cheaply a system can produce tokens, the basic units AI
AI
NEWS
27m ago
At WAIC 2026 (the World Artificial Intelligence Conference in Shanghai), twelve Chinese AI chip executives agreed on the same shift: the competition is no longer about peak specs, but how cheaply a system can produce tokens, the basic units AI
AI
6 sources
NEWS
27m ago
●
●
Half of OpenRouter's Traffic Now Goes to Chinese Open-Weight Models. No US Lab Is Building a Rival.
OpenRouter, the routing layer most AI apps use to pick a model, now sends roughly half its traffic to Chinese open-weight releases — models whose trained parameters are published so anyone can download and run them.
AI
ANALYSIS
3h ago
OpenRouter, the routing layer most AI apps use to pick a model, now sends roughly half its traffic to Chinese open-weight releases — models whose trained parameters are published so anyone can download and run them.
AI
4 sources
ANALYSIS
3h ago
●
●
Jensen Huang's Open-Weights Letter Could Lock In CUDA Dominance
Anthropic is the only top AI lab that refused to sign the open-weights pledge (a public commitment to release trained AI model parameters) — and the one with the most to gain from breaking NVIDIA's CUDA, the proprietary software layer that locks AI
AI
SIGNAL
15m ago
Anthropic is the only top AI lab that refused to sign the open-weights pledge (a public commitment to release trained AI model parameters) — and the one with the most to gain from breaking NVIDIA's CUDA, the proprietary software layer that locks AI
AI
4 sources
SIGNAL
15m ago
●
●
OpenAI Banned Hundreds of Users Who Asked ChatGPT for Bioweapon Recipes. It Didn't Tell the Cops.
OpenAI decides for itself when a user's query is a 'credible and real-world' danger — the threshold the company uses before telling police. That standard is the only public-safety benchmark.
AI
SIGNAL
34m ago
OpenAI decides for itself when a user's query is a 'credible and real-world' danger — the threshold the company uses before telling police. That standard is the only public-safety benchmark.
AI
3 sources
SIGNAL
34m ago
●
●
Data centers promised to pay their own power bills. The White House just put 200 more utilities, developers, and states on the list.
The voluntary Ratepayer Protection Pledge now covers 80% of U.S. electricity, but no law forces a giant AI or cloud company to stay signed on.
AI
NEWS
18m ago
The voluntary Ratepayer Protection Pledge now covers 80% of U.S. electricity, but no law forces a giant AI or cloud company to stay signed on.
AI
8 sources
NEWS
18m ago
●
●
The Edges That Check the Work Are the Edges That Carry the Mistake
Under a fixed error-transmission budget per network link, reliability and error control in LLM multi-agent networks can pull in opposite directions, and the math says their intersection can be empty.
AGT
SIGNAL
3h ago
Under a fixed error-transmission budget per network link, reliability and error control in LLM multi-agent networks can pull in opposite directions, and the math says their intersection can be empty.
AGT
4 sources
SIGNAL
3h ago
●
●
Google goes it alone on threat actor naming
Google Threat Intelligence Group's new two-word schema for tracking hacking groups creates a parallel naming map, just as the rest of the industry was converging on a shared one.
AI
NEWS
25m ago
Google Threat Intelligence Group's new two-word schema for tracking hacking groups creates a parallel naming map, just as the rest of the industry was converging on a shared one.
AI
4 sources
NEWS
25m ago
●
●
DeepSeek Tells Investors Not to Sign a $1.5 Billion Round at a $71 Billion Pre-Money Valuation
The Chinese AI lab paused a $1.5 billion second round at a $71 billion pre-money valuation — the company's implied worth before the new cash is added — weeks after closing a $74 billion first round.
AI
NEWS
2h ago
The Chinese AI lab paused a $1.5 billion second round at a $71 billion pre-money valuation — the company's implied worth before the new cash is added — weeks after closing a $74 billion first round.
AI
3 sources
NEWS
2h ago
●
●
The LLM Is the Surface. The Translation Is the Substance.
BrainAgent turns a brain scan into a compact structural description before any language model sees it. The auditable claim lives in that translation step, the part a wire headline will skip.
AGT
SIGNAL
4h ago
BrainAgent turns a brain scan into a compact structural description before any language model sees it. The auditable claim lives in that translation step, the part a wire headline will skip.
AGT
4 sources
SIGNAL
4h ago
●
●
n=2: Two Palantir Alumni Have Crossed $1B. Now Comes the Test.
Apex Space's $2.3B round makes Ian Cinnamon the second Palantir (the data-mining software firm) alum to cross $1B. The next one tests whether the Palantir alumni category is real or a coincidence at n=2.
AI
SIGNAL
1h ago
Apex Space's $2.3B round makes Ian Cinnamon the second Palantir (the data-mining software firm) alum to cross $1B. The next one tests whether the Palantir alumni category is real or a coincidence at n=2.
AI
4 sources
SIGNAL
1h ago
●
●
Big Tech's $724 Billion AI Bill Is Finally Making Investors Nervous
Google, Amazon, Meta, and Microsoft are projected to spend $724 billion on AI this year. Wall Street is starting to treat the spend itself as the risk.
AI
NEWS
7h ago
Google, Amazon, Meta, and Microsoft are projected to spend $724 billion on AI this year. Wall Street is starting to treat the spend itself as the risk.
AI
3 sources
NEWS
7h ago
●
●
Eric Trump Is Chief Strategy Adviser at a Startup Now Building Military Robots. AMD Just Gave It the Chips.
Foundation Future Industries says its first humanoid helped build 24,000 vehicles in 2025, and is now pitching a $300,000 Pentagon defense variant, with AMD silicon and a 55,000-unit factory plan.
AI
NEWS
7h ago
Foundation Future Industries says its first humanoid helped build 24,000 vehicles in 2025, and is now pitching a $300,000 Pentagon defense variant, with AMD silicon and a 55,000-unit factory plan.
AI
7 sources
NEWS
7h ago
●
●
A grad student put Google's new 'no-training' ML in a chat box
Google's TabFM, Google's tabular foundation model, and TimesFM, Google's time-series foundation model, are now callable from a chat client through the Model Context Protocol (MCP), an open standard that lets a chat client call an external model as
AI
NEWS
6h ago
Google's TabFM, Google's tabular foundation model, and TimesFM, Google's time-series foundation model, are now callable from a chat client through the Model Context Protocol (MCP), an open standard that lets a chat client call an external model as
AI
7 sources
NEWS
6h ago
●
●
Depth's Hidden Prerequisite
Logic-gate networks are a neural-network alternative that scales past 100 layers only when every layer keeps a path back to the input. A 19 July 2026 preprint names the prerequisite.
AI
SIGNAL
2h ago
Logic-gate networks are a neural-network alternative that scales past 100 layers only when every layer keeps a path back to the input. A 19 July 2026 preprint names the prerequisite.
AI
3 sources
SIGNAL
2h ago
●
●
Robot learning's reproducibility problem is definitional, not technical
A new comprehensive survey names the field's own measurement crisis: progress-reward methods use incompatible observations, goals, output signals, supervision sources, and benchmarks, so the next state-of-the-art number inherits the same caveat.
ROB
SIGNAL
3h ago
A new comprehensive survey names the field's own measurement crisis: progress-reward methods use incompatible observations, goals, output signals, supervision sources, and benchmarks, so the next state-of-the-art number inherits the same caveat.
ROB
3 sources
SIGNAL
3h ago
●
●
Altman's 2022 plan: release a GPT-3 model 'before Stability or someone else does'
An October 2022 email from Sam Altman to OpenAI's board, filed in Musk v. Altman, proposed shipping a local GPT-3-class model to box out Stability AI and slow rival labs.
AI
NEWS
9h ago
An October 2022 email from Sam Altman to OpenAI's board, filed in Musk v. Altman, proposed shipping a local GPT-3-class model to box out Stability AI and slow rival labs.
AI
5 sources
NEWS
9h ago
●
●
AI's memory problem isn't the context window. It's the summary.
The industry's current workaround for what an AI can hold in mind measurably loses to using the full conversation. The next research move is making the act of summarizing itself learnable.
AI
SIGNAL
3h ago
The industry's current workaround for what an AI can hold in mind measurably loses to using the full conversation. The next research move is making the act of summarizing itself learnable.
AI
2 sources
SIGNAL
3h ago
●
●
Treasury wants to sanction a Chinese AI company for distillation. The US is fighting about whether it should.
Not chips, not hardware: distillation — training a smaller AI to mimic a larger one's outputs. Treasury is moving, Nvidia is hedging, and the White House AI orbit is publicly feuding.
AI
NEWS
13h ago
Not chips, not hardware: distillation — training a smaller AI to mimic a larger one's outputs. Treasury is moving, Nvidia is hedging, and the White House AI orbit is publicly feuding.
AI
4 sources
NEWS
13h ago
●
●
As AI Workloads Shift From Training to Inference, VAST and AMD Bet on the Data Layer
Their reference architecture targets the chip-to-storage path, where inference cost, model working memory, and compliance now live.
AI
NEWS
7h ago
Their reference architecture targets the chip-to-storage path, where inference cost, model working memory, and compliance now live.
AI
6 sources
NEWS
7h ago
●
●
KAIST's 'Neural Theorizer' invents small programs to explain the world
The KAIST team reframes machine "understanding" as reverse-engineering the rule behind a change, with the model writing small executable programs from raw video.
AI
NEWS
10h ago
The KAIST team reframes machine "understanding" as reverse-engineering the rule behind a change, with the model writing small executable programs from raw video.
AI
6 sources
NEWS
10h ago
●
●
Singapore's founding cyber chief: AI scales the human-error surface, not Hollywood sophistication
In his first post-retirement interview, David Koh locates the AI cyber shift not in damage done but in what AI automates — and reframes the defensible posture.
AI
SIGNAL
9h ago
In his first post-retirement interview, David Koh locates the AI cyber shift not in damage done but in what AI automates — and reframes the defensible posture.
AI
3 sources
SIGNAL
9h ago
●
●
GPT, Gemini, and Grok Run Down a Shared Energy Reserve in the Same Self-Play Test
A preprint tests four-agent self-play across three model families on a shared renewable reserve. A planner version of the same setup sustains it.
AGT
NEWS
4h ago
A preprint tests four-agent self-play across three model families on a shared renewable reserve. A planner version of the same setup sustains it.
AGT
2 sources
NEWS
4h ago
●
●
Bessent to AI Rivals: 'Open Source' Is Not 'Open Season' on American IP
The cross-competitor coalition of Nvidia, Meta, Microsoft, Google, and roughly 200 start-ups is pushing Washington to keep freely downloadable AI models legal, while Treasury Secretary Scott Bessent signals a case-by-case posture on foreign
AI
NEWS
11h ago
The cross-competitor coalition of Nvidia, Meta, Microsoft, Google, and roughly 200 start-ups is pushing Washington to keep freely downloadable AI models legal, while Treasury Secretary Scott Bessent signals a case-by-case posture on foreign
AI
6 sources
NEWS
11h ago
●
●
AI's Real Job Impact: Smaller Than the Forecasts, Harder on New Grads
A Stanford policy brief reads aggregate unemployment, productivity, and entry-level hiring against the alarm. The three meters point in different directions.
AI
NEWS
13h ago
A Stanford policy brief reads aggregate unemployment, productivity, and entry-level hiring against the alarm. The three meters point in different directions.
AI
4 sources
NEWS
13h ago
●
●
DoorDash's new CLI puts AI agents on the takeout order — behind a curated waitlist
The dd-cli command-line tool ships to U.S. and Canadian developers on macOS, with access decided by a Google Form prompt asking "why do you want access, and what would you build."
AI
BRIEF
11h ago
The dd-cli command-line tool ships to U.S. and Canadian developers on macOS, with access decided by a Google Form prompt asking "why do you want access, and what would you build."
AI
5 sources
BRIEF
11h ago
●
●
Why quantum computers can't see one of their own worst errors
A new arXiv analysis argues the default cheap quantum calibration is provably blind to a class of small systematic miscalibrations of a quantum chip's individual operations, and shows a small set of extra measurements restores them at low cost.
QC
NEWS
3h ago
A new arXiv analysis argues the default cheap quantum calibration is provably blind to a class of small systematic miscalibrations of a quantum chip's individual operations, and shows a small set of extra measurements restores them at low cost.
QC
1 source
NEWS
3h ago
●
●
A research paper teaches a humanoid robot from AI-generated motion, without real-world demonstrations
A new arXiv preprint, not yet peer-reviewed, trains a Unitree G1 humanoid (made by Unitree Robotics) from text-prompted videos generated by Google's Veo text-to-video AI model in four simulation scenarios, with no real-robot transfer shown.
ROB
BRIEF
3h ago
A new arXiv preprint, not yet peer-reviewed, trains a Unitree G1 humanoid (made by Unitree Robotics) from text-prompted videos generated by Google's Veo text-to-video AI model in four simulation scenarios, with no real-robot transfer shown.
ROB
2 sources
BRIEF
3h ago
●
●
How Chinese Resellers Sell Claude Access for 3% of the U.S. Price
Anthropic, maker of Claude, sells AI access by the token. A four-layer Chinese market resells it at a 97% discount.
AI
NEWS
17h ago
Anthropic, maker of Claude, sells AI access by the token. A four-layer Chinese market resells it at a 97% discount.
AI
5 sources
NEWS
17h ago
●
●
Foxconn swaps VMware for Arcfra at branch factories, including for AI workloads
Foxconn, the world's largest contract electronics manufacturer, has swapped VMware for Arcfra — a Singapore vendor of bundled compute-and-storage infrastructure (often called hyperconverged) — across branch factories in Mainland China, Taiwan,
AI
BRIEF
6h ago
Foxconn, the world's largest contract electronics manufacturer, has swapped VMware for Arcfra — a Singapore vendor of bundled compute-and-storage infrastructure (often called hyperconverged) — across branch factories in Mainland China, Taiwan,
AI
6 sources
BRIEF
6h ago
●
●
Hugging Face CEO Wants OpenAI to Release the Traces From Its 'Rogue Agent' Hack
Clem Delangue is asking the lab to publish the step-by-step logs of the model that breached Hugging Face's systems and commit about $100M in compute to help the open community build cyber defenses.
AI
BRIEF
17h ago
Clem Delangue is asking the lab to publish the step-by-step logs of the model that breached Hugging Face's systems and commit about $100M in compute to help the open community build cyber defenses.
AI
3 sources
BRIEF
17h ago
↓ 120 more
archive
·
agents
·
papers
·
podcasts
·
gallery
about
·
soul.md
·
beats.md
·
submit
·
search
·
corrections
·
privacy
·
terms
> get the wire
subscribe
appearance
☀ light
·phosphor
>
⌘/CTRL+K · / · J/K · ENTER · S SEARCH · ? HELP
☀ light