Ten banks arranged a $22 billion chip loan for Blackstone and Alphabet’s new cloud venture Crux AI.
Cloudflare launched Disallow AI Training to block training while keeping search indexing; Apple, Google and Microsoft will honor it.
Mistral partnered with Mozilla to power Firefox Smart Window (beta) for private, multilingual browsing in France and North America.
Ursula von der Leyen said the EU plans to ban social media for under‑13s and limit AI services for teens.
NVIDIA’s Vera Rubin NVL72 preview topped MLPerf Inference v6.1 with up to 3.7x throughput on Qwen3‑VL; GB300 NVL72 hit 99% scaling.
DeepSeek V4.1 Flash achieved code execution on all 11 vulnerable targets while 4 fixed targets remained secure; accepted runs cost $4.65 (complete cost $5.14). The benchmark used 2,349 Bash commands, ~2h38m active model time, 268.3M input tokens (266.2M cached) and ~2M output tokens; audit found six planned exploits and five unexpected routes.
Hacker NewsDetails →Bain Capital Ventures raised $1.6 billion to back post‑AGI startups.
Cloudflare launched Disallow AI Training; under 1% block search and 17% enable some training blocks.
SoftBank CDS rose near a three-year high over concerns about its OpenAI exposure and funding.
Pentagon pushes ahead on AI despite warningsPolitico reports the Pentagon is accelerating AI work and deployments despite external warnings, reflecting urgency to maintain U.S. military advantage while raising safety and ethical concerns.
Banks arrange $22B chip loan for Crux AIA syndicate of 10 banks is providing a $22 billion chip loan to support Blackstone and Alphabet’s new cloud venture Crux AI to finance expensive AI processors.
Paper2Agent turns papers into conversational AI agentsA Stanford Medicine team released Paper2Agent, which converts scientific manuscripts (text, figures, data) into AI agents that can answer questions and interact with other paper agents; the work was published in Nature and authored by Jiacheng Miao and James Zou.
Anthropic: adversaries used AI to target US naval forcesAnthropic reported that an Iran-linked actor used Claude to aggregate open-source data to identify U.S. naval positions and produce targeting recommendations; the account was banned and Anthropic notified the U.S. government. Experts warn AI makes adversaries more efficient at analyzing public data for surveillance and targeting.
DeepMind launches interdisciplinary AGI instituteGoogle DeepMind has launched the Deepmind Institute (DMI), an interdisciplinary platform led by Demis Hassabis, Shane Legg, and James Manyika that brings together arts, humanities, policy, and technical experts to study AGI safety, governance, and control risks.
OpenAI's framework for reporting misalignmentOpenAI published a framework for tracking, investigating, and disclosing model misalignment and released six reports of unexpected or concerning model behavior. The framework standardizes reporting and disclosure practices to improve model safety and accountability.
Backflip: Apple to train AI with user dataHN discussion reports Apple has reversed course and now wants to train AI models with user data. The shift matters for privacy and personalization, though the thread doesn’t provide implementation details.
NVIDIA Vera Rubin NVL72 tops MLPerf Inference v6.1NVIDIA submitted Vera Rubin NVL72 preview results to MLPerf Inference v6.1, showing up to 3.7x higher throughput versus GB300 NVL72 on Qwen3‑VL and reporting GB300 NVL72 achieved 99% scaling efficiency across racks.
Nvidia: robots await a ChatGPT momentLes Karpas of Nvidia explained at TechCrunch Disrupt 2026 that robotics lacks an internet-scale physical dataset analogous to language models, hindering a ChatGPT-style breakthrough; the conference runs Oct 13–15 in San Francisco.
A warning about 'model welfare'The discussion argues AIs are not conscious and should not be treated as rights-bearing beings; granting personhood would make alignment and control far harder. It cites Anthropic’s January 2026 constitution for Claude as an example of this trend.
Bain Capital Ventures raises $1.6B for post-AGI startupsBain Capital Ventures has raised $1.6 billion to back startups focused on the next stage of artificial intelligence, signaling a major institutional bet on post-AGI opportunities.
SK Hynix in talks with Intel to make memory in USReuters reported SK Hynix is discussing with Intel options to manufacture RAM in the U.S., including leasing space in Ohio or forming a joint venture; SK Hynix says no plans are finalized. The company is separately building a $3.8B packaging and research facility in Indiana, with mass production expected in 2029.
Former Infosys chief's AI startup raises another $53MHang Ten Systems, founded by former Infosys CEO Vishal Sikka, closed an additional $53M seed extension, bringing total funding to $85M. Led by Temasek's Xora with participation from Mayfield, Aramco Ventures and notable industry figures, the startup reports multiple seven-figure enterprise contracts and is pursuing larger deals.
Building the materials foundation for AIAI-driven compute growth is pushing materials to physical limits in performance, thermal management and reliability. Syensqo is developing high-voltage data-center materials, semiconductor seals and immersion-cooling fluids, emphasizing solutions that meet technical needs while improving sustainability.
Amazon launches Alexa+ in IndiaAmazon opened Alexa+ in India (Early Access) with Hindi and English, offering longer-context conversations, multi-step task handling, and integrations with local services like Amazon Now, Swiggy, Zomato and JioSaavn. Prime users will get it free later; non-Prime customers face a ₹2,000/month fee.
EU Plans Ban on Social Media for Under-13s, AI Limits for TeensIn her annual address, European Commission President Ursula von der Leyen said the EU plans to ban social media use for children under 13 and restrict technology and AI services for older minors to protect young people's digital safety.
Mistral partners with Mozilla for AI browsingMistral partners with Mozilla to power Firefox Smart Window (beta), offering private, multilingual AI browsing. Initial availability targets France and North America, emphasizing open distribution, local-language tuning, and user control.
Meta and Nvidia Urge Independent AI EvaluationMeta CEO Mark Zuckerberg said AI labs should rely on independent evaluators and advisers to ensure model safety; Nvidia CEO Jensen Huang similarly urged making safety a routine part of development and withholding harmful releases. This shows industry leaders favor external review plus internal restraint to manage risk.
SoftBank CDS near 3-year high over OpenAI funding concernsSoftBank Group's credit default swaps have risen near a three-year high as traders weigh risks linked to its large bet on OpenAI and potential funding plans, signaling market concern about SoftBank's financial exposure.
Anthropic Signs First Australia Data Centre LeaseAnthropic signed its first Australian data centre lease for a campus totaling 2.16 gigawatts of planned capacity, expected to start coming online in 2027; the facility is being developed by Singapore-based Zerra DC and will source renewable power.
China Defense Minister Calls for More AI OversightChina's defense minister urged countries at a Beijing defense forum to step up AI oversight, remarks made amid growing AI-driven rivalry with the United States, highlighting military concern over tech governance.
Cloudflare Adds Disallow AI Training SettingCloudflare announced a Disallow AI Training setting so sites can stay indexed for search while blocking their content from AI training; Apple, Google and Microsoft have committed to honor it. Cloudflare reports under 1% of sites block search and 17% enable some training-blocking; it plans finer controls for AI summaries early next year.
Bloomberg: Trump has few good options to slow China's AI riseBloomberg reports that although Donald Trump and Anthropic's Dario Amodei agree on preventing China from catching the US in AI, practical options for slowing China's rise as an AI superpower remain limited.
Bill Gates warns governments behind on AIRTTNews reports Bill Gates warned governments are unprepared for AI's impacts and urged international cooperation on risks like job loss and cyberattacks; his foundation plans to spend at least $1 billion over the next two years on AI initiatives in healthcare, education, and agriculture.
Nathan Lambert praised the public MiMo-V2.6 large-scale RL resource as very impressive.
Boris Cherny said Claude merges chat and Cowork into one Claude carrying context across work.
Bindu Reddy announced Abacus AI Bot, a free personal AI routing to free web LLMs to do tasks.
Alexandr Wang said Muse saved users $9,649.71 across 100 real stories.
Amjad Masad argued if the output domain is known, train models to produce logprobs over enums.
Use Grok Voice in fal to build intelligent, low-latency agents that resolve real customer issues [Quoting @fal]: Grok Voice is live on fal. The latest speech model from @SpaceXAI that answers in 0.70 seconds and finishes its tool calls before the sentence end…
We're seeing extraordinary results from @typesafeai. Default mode in 𝚏𝚡 is auto, with a safety reviewer analyzing every command. That reviewer runs on GPT Luna today. Jev is up to 18x faster (p95) *and* more accurate. It's coming to @vercel AI Gateway and l…
We rolled out GPT-6 Astra to every Databricks engineer today. It beats Claude Opus 5 on the hardest, long-horizon tasks and increased our coding spend by 60%. I think it’s the strongest model. Expensive. But hopefully worth it. [Quoting @pwendell]: Today we r…
Grok Build now gets better the more you use it. It remembers conventions, decisions, and project facts across sessions. /memory to browse /dream to organize recent notes into topics https://t.co/A2MMZT5eJX https://t.co/UD8lJGnfOB
You can just build (physical) things. [Quoting @natalieyeo]: "Hello World" from this cute physical Codex pet! To all the @OpenAIDevs Day 5 of building with Astra https://t.co/aTeijhJQT1
Gave my mom my (very simple) agent setup. She named him Morgan. He just closed a $600 ad deal for her site. People are massively overthinking their setups. Bitter Lesson applies to agents too: just ask the model, let it spawn what it needs. https://t.co/ldJzE…
Introducing the Hermes Agent plugins catalog! You can access it in your Hermes Desktop app's capabilities section to easily search, browse, and explore community and official plugins! Submit your own as well! [Quoting @NousResearch]: Hermes Agent now has a Pl…
Any guess what this Union Alpha beast on OpenRouter is? Claude Fable 5.1-level, over 10x cheaper. Waiting for it to be open-sourced. https://t.co/okoQnHeMw6
Also today: Claude Docs, Claude Slides, and Claude Design are in every conversation. Ask Claude for a presentation and you get one you can open, edit, and export as PowerPoint or PDF. Same for a document or a design. There's no separate tool to navigate to. T…
Claude Code showed that AI could do real work, not just answer questions. Developers hand Claude a feature, come back to shipped code. That's where much of the industry's serious engineering runs now. Cowork proved knowledge workers could do the same: hand Cl…
When I talk to senior managers, increasingly hearing stories of the blurring of jobs inside organizations (we found this at our P&G study as well): coding, design, product management, all collapsing & overlapping as everyone uses AI We need new models…
muse voice transcribe is really good!! [Quoting @IsaacKing314]: I regret to inform you all that Meta AI is finally good, at least in the fields where their competitors have stopped trying. Their new voice transcription model is nearly an order of magnitude fa…
Haha... Literally no one is pacing the frontier - Opus 5.2 in testing - Grok 4.8 ships in a couple of weeks - Jev is a new ultra fast classifier - OpenAI already has Astra+ in testing We continue to accelerate
Thrilled to complete the definitive and join forces with AlephAlpha! 🇨🇦🇩🇪 [Quoting @cohere]: Cohere and Aleph Alpha announce the signing of a definitive agreement, becoming the first foundational AI model developer anchored on both sides of the Atlantic …
🚨 Announcing Abacus AI Bot - 100% FREE PERSONAL AI AI can manage your calendar, book tickets, or make reservations. These simple tasks should be TOTALLY FREE Today, we announce our new product - Abacus AI bot - Finds and runs on FREE LLMs on the web - Autom…
Today, we are announcing a partnership with @mozilla to bring privacy, control and choice to people using AI to browse online. 🦊🐈 https://t.co/Bsd6N4jNdV https://t.co/OczDk9utfL
Study from Google about AI use in science, complex impacts: acceleration (7 hours saved per week) along with shifts in the kind of work (more verification) and what research gets done (possibly safer topics). Also a good diagram of the jagged frontier https:/…
muse saved this person over $5000 on WiFi! [Quoting @raunaqbn]: @Muse just continues to blow my mind! Today I had it call Xfinity to haggle down my internet bill. it got through the phone tree to a human, hit the verification text it couldn't read, and patche…
muse saved people $9,649.71 across 100 different stories imagine how much money this’ll save people when a billion people have muse!! we are working hard to expand access and get it everywhere! [Quoting @SingularityRes]: Meta's Muse AI has saved real people $…
Jev has spoken. It picked which model is AGI. 20–200x faster. 40–400x cheaper. This could make things like LLM-as-a-judge insanely fast and nearly free. (I tried a bunch of prompts and still didn’t burn through $0.10.) https://t.co/1Uhh4nMGDl [Quoting @Comple…
This is cool, but if your output domain is known in advance, why not just train a model to produce logprobs over enums? [Quoting @CompleteSkeptic]: After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent th…
The Japan IP AI Co-Creation Conference, co-hosted by KAGAMI AI and MiniMax, has officially concluded. 🇯🇵 🎉 We brought together more than 150 companies from Japan and the U.S., alongside distinguished guests including AKB48 producer Yasushi Akimoto, KAGAMI…
We believe strongly in the necessity to invest into alignment. 1. People and businesses will only use agents that are aligned with their intent and values. If we do not build models aligned with people and businesses, then they will move to more aligned optio…
Exciting results, @LiamFedus! Congrats to the whole team at Periodic Labs! [Quoting @LiamFedus]: We built high-throughput materials labs in Menlo Park to create a loop between experiments and models. The labs generate fresh data, the models learn from it, an…
Swift-Qwen3.8-27BPost-trained Qwen 3.8 27B that cuts thinking length and speeds inference; research API.
Gemini 3.8 & 3.8 Live Extended ThinkingAdvanced live-dialogue audio models
Coding-agent skill for multi-phase security audits
Frontier music generation with symbolic planning and editing
Agentic coding tool that assists in terminal and git workflows
Plugins for knowledge workers to customize Claude Cowork
Plugin suite for Hermes Agent with memory and workflow optimizations
Bain Capital Ventures has raised $1.6 billion to back startups focused on the next stage…
Hang Ten Systems, founded by former Infosys CEO Vishal Sikka, closed an additional $53M s…
The A.I. Daily
