Making it easier for websites to express AI training preferences
OpenAI will honor a shared robots.txt signal to help website operators communicate their AI training preferences to participating providers.
Models openai.com signal 83 primary
Archived ranking
240 rankedSam Altman says it’s better for RAM to be in a data center than on your desk
Asked about supply chain shortages, like the ongoing RAM shortage, OpenAI CEO Sam Altman said in an interview that it's "terrible" to have RAM in personal PCs instead of data centers. During his appearance on the Cheeky Pint podcast, where he also commented on AI's impact on math, Altman said: "A lot of people want […]
Chips theverge.com signal 79 verified
Sam Altman sympathizes with mathematicians, but says OpenAI’s math onslaught won’t stop
After months of bombarding mathematicians with a relentless stream of AI-generated results, OpenAI CEO Sam Altman says he sympathizes with researchers struggling to make sense of it all.
Chips theverge.com signal 75 verified
OpenAI Lowers Fundraising Target–After Decreasing Possible Valuation
OpenAI has lowered the amount of capital it is seeking from investors in its latest fundraising round by $10 billion, according to a Friday report by The Financial Times. The company was initially seeking to raise $30 billion at a $1.4 trillion valuation. It lowered the valuation last week to...
Policy theinformation.com 2 sources signal 74 verified
What X is saying
visibility
@artificialanlys
We’ve released AA-Omniscience v1.1, updating our knowledge and reasoning benchmark with a refreshed public set, removal of saturated questions, and ambiguity fixes AA-Omniscience is our proprietary benchmark measuring model knowledge and hallucinations,...
High-engagement X post with 202 visible interactions on frontier model behavior.
@iscienceluvrHow to Loop Your Transformer: Loop Scaling, Specialization, and Parallel Inference New paper from Meta. "Our experiments span models with up to 1B parameters, trained on up to 155B tokens using up to 1.07 × 10^21 training FLOPs, with up to 32 loops. We...
High-engagement X post with 137 visible interactions on frontier model behavior.
a16zThe "Rack Spread" is unfolding, token prices are falling while GPU prices climb Change since mid-July: - B300 GPU rentals up 106% - B200 up 66% - H100 up 17% - Average token prices down 58% Charts of the Week:
High-engagement X post with 149 visible interactions on AI compute and inference.
@aravsrinivasSharing early evals on Gemini 4 Carbon. Strongest model that we've seen so far for long-form research and answer quality, with a new SoTA on the Wide-And-Deep-Research (WANDR) benchmark. Coming soon to all Perplexity Pro and Max users inside Computer, once...
High-engagement X post with 186 visible interactions on frontier model behavior.
@ns123abcBREAKING: Google just proved AI data centers in space actually work Project Suncatcher's first satellite, built with @planet and launched on a SpaceX rocket with 4 TPUs on board, verified the physics: Google sent a prompt from Earth, a TPU in orbit ran...
High-engagement X post with 743 visible interactions on AI compute and inference.
@bilawalsidhuGaussian splats are cool, but the same 3D scan can be a GPS for your drone, robot and phone. It's moments like this where you really start feeling like Tony Stark. Splat of a few city blocks → MultiSet VPS → one shared map. Drop a waypoint in my glasses...
High-engagement X post with 147 visible interactions on robotics progress.
@amasadMassively underreported: an official Claude Code extension was compromised and installed backdoors on developer machines. Could it be behind the Binance hack?
High-engagement X post with 278 visible interactions on AI developer tooling.
@unslothaiWe’re releasing new DeepSeek-V4.1-Flash-1017 GGUFs, ranking highest across benchmarks. Dynamic V3.1 leads across nearly all benchmarks: GPQA Diamond, AIME25, LiveCodeBench v6, MMLU-Pro, Aider & more. Run on 64GB+ RAM. GGUF:
High-engagement X post with 303 visible interactions on frontier model behavior.
@teortaxestex> reconstruction of Anthropic's financials for the case of fully utilized Max subscriptions oh, SemiAnalysis did that! If that's correct, Ant's API margin is higher than DeepSeek's (≈92% vs 84%), but their plan subsidy for Opus is *enormous*. They break...
High-engagement X post with 119 visible interactions on AI developer tooling.
@adamholterererNote that 5.5 on high is a lot better than GPT 6.1 Sol on high, so though the cost gap is a lot smaller when you consider subsidization, you're getting a better model. If you compare a lower reasoning effort on Opus to a higher one on Sol to be more fair,...
High-engagement X post with 171 visible interactions on frontier model behavior.
@itsmeajaykvEver since Strata release, Qwen3.8-Flash-Next has became my main local model. It's been weeks since I last ran Qwen3.8-27B. Flash is equally capable and faster than 27B, so it's a no brainer. Since this depends on the quant, here's what am comparing. -...
High-engagement X post with 186 visible interactions on frontier model behavior.
@cursor_aiWe identified a security incident affecting a modified Claude Code VS Code extension accessed through our Cursor IDE extension marketplace, impacting at most 5,000 users. This malicious version of the extension was not developed by Anthropic. It does not...
Official or high-signal @cursor_ai post with 1,592 visible interactions on AI developer tooling.
tszzl6 Astra is an old model at this point but inside the Codex harness it has a level of AGI-ness that all the other frontier models lack. computer and internet use are seamless and Just Work™️; covers almost anything a person can do on a computer
High-engagement X post with 754 visible interactions on AI developer tooling.
@kimi_moonshotIntroducing 3D Make in Kimi. Kimi connects to tools like Blender and CAD tools to create models, assemble scenes, and develop games. Available now on kimi.ai and Kimi Work.
High-engagement X post with 371 visible interactions on frontier model behavior.
Signals
All →No signals in the last 7 days.
Just in
Story
tech-insider.org
DeepSeek Raises $16B, 27% From State Funds [2026]
Story humanoid.guideWelcome, Lumen H1!
Story siliconangle.comGoogle, Anthropic and others to support new AI crawling opt-out standard
Story techradar.comSomeone found their lost Bose bud in a field of gravel using ChatGPT, and it's genius
Topic radar
All →
AI infrastructure and compute
+8.7
Surging compute requirements for AI workloads are driving expanded semiconductor manufacturing, long-term memory supply constraints, data center credit support, and competitive efforts to secure regional electrical power.
Capital and consolidation +1.9Venture funding, initial public offerings, and acquisitions remain active across AI developers and enterprise platforms, even as some major labs revise valuation targets and funding goals downward amid market caution.
Robotics and embodied AI -0.8Humanoid robotics development continues with the introduction of new bipedal platforms, including full-size and compact models designed by regional robotics manufacturers for physical environment tasks.
Quick takes
Hardware providers are stepping directly into credit markets to backstop neocloud debt, offering loan guarantees for massive data center buildouts as capital demands outpace conventional financing.Frontier lab safety infrastructure is expanding from static evaluations into active containment protocols, combining offline training environments with layered runtime monitoring to detect unauthorized agent behavior before incidents escalate.Frontier model developers are broadening enterprise distribution by embedding agentic workflows directly into third-party enterprise platforms and rival cloud marketplaces.Experimental AI compute infrastructure is reaching orbit, with data center class TPUs successfully serving model inference from prototype satellites.Empirical trials reveal that professional analysts adopt an AI coding agent's favored recommendation in 86% of cases, underscoring systemic challenges in maintaining independent human oversight over automated workflows.Escalating capital expenditures and shifting investor sentiment are compressing frontier AI valuations, prompting reduced fundraising targets and speculative credit ratings for capital-intensive labs.Hardware providers are increasingly underwriting AI compute expansion directly, with Nvidia weighing credit guarantees to finance Nscale's multi-billion dollar data center infrastructure intended for Anthropic.Frontier lab safety operations are expanding across lifecycle stages, combining offline containment and high-compute red-teaming during training with layered runtime monitoring to intercept unauthorized agent actions before escalation.
Just in
Story
tech-insider.org
DeepSeek Raises $16B, 27% From State Funds [2026]
Story humanoid.guideWelcome, Lumen H1!
Story siliconangle.comGoogle, Anthropic and others to support new AI crawling opt-out standard
Story techradar.comSomeone found their lost Bose bud in a field of gravel using ChatGPT, and it's genius
Market pulse
fear
AI pulse
25/100
AI-linked equities are under pressure, with Advanced Micro Devices -10.9%, Arm Holdings plc -9.97%, Oracle -6.37% driving the tracked basket lower.
| Ticker | Company | Move |
|---|---|---|
| AMD | Advanced Micro Devices | -10.9% |
| ARM | Arm Holdings plc | -9.97% |
| ORCL | Oracle | -6.37% |
| PLTR | Palantir Technologies. | -6.27% |