All AI news

AIAi2 reports that over a 30-day test of its new scheduler, teams received 98% of the GPU hours they were owed, based on actual demand. Cluster occupancy stayed at 98%, and spare capacity went to interruptible work without drawing down team budgets.

AIAi2 has published a blog post on the design decisions behind its scheduler, along with rollout lessons and next priorities. The post links to The text provided does not include further details of the scheduler's features or results.

AIMeta spent heavily on Anthropic's Claude models, with internal use reaching up to 60,000 employees and a projected $10 billion yearly spend, according to The Algorithmic Bridge. The author says Meta then released Muse Spark, which scored 52 on the Artificial Analysis intelligence benchmark, on par with Claude Opus 4.6.

AIAi2's AI Infrastructure team replaced its priority-based scheduler for GPU clusters with GPU time budgets, hierarchical fair-share allocation, and a time-slicing contract. The team says the change moved debates over how much GPU time each research project deserves from case-by-case operational decisions into a transparent budgeting process. The clusters range from 88 to 1024 GPUs across NVIDIA H100, B200, and B300 hardware, and serve about 150 internal researchers.

AIAmazon Bedrock Managed Agents, powered by OpenAI, entered public preview, and OpenAI's GPT-6 Astra, GPT-6.1 Sol, and GPT-6.1 Luna became generally available on Amazon Bedrock. AWS also released Strands Decider 2B, a 2B-parameter open source decision model that answers in about 115ms locally, and said the Strands harness uses 28 percent fewer tokens than popular harnesses while matching their accuracy.

AIAi2's AI Infrastructure team replaced its priority-based GPU cluster scheduler with a system using GPU time budgets, hierarchical fair-share allocation, and a time-slicing contract. The team says the change turns decisions about how much GPU time each research project receives into a transparent administrative budgeting process. Its clusters, which range from 88 to 1024 GPUs including H100, B200, and B300 units, serve about 150 researchers facing demand two to three times available capacity.

AIOpenAI's research leaders say they parted ways with Jasmine, Mikita, and Tomek after an investigation found they violated policies on handling sensitive information. The company says the decision was not about raising safety concerns and that it is finalizing contracts with third-party safety assessors, with details to follow in the coming weeks.

AIPixVerse says each session includes a platform walkthrough, a live OpenAI demo showing ChatGPT generating a creative brief and finished video via the PixVerse Plugin, a creator sharing their workflow, and live Q&A. The post presents these as recurring sessions rather than a new product launch.

AIPixVerse says every attendee receives free credits to try its workflow. Those who complete the post-session survey may be randomly selected, with 100 winners per session, to win a month of ChatGPT Pro.

AIOpenAI says Codex on Windows now has a new sandbox mode built on Microsoft's Execution Containers (MXC), offering faster setup, stronger network enforcement, and granular file access controls. The mode requires a compatible Windows 11 device. Background from Microsoft's announcement says MXC is now generally available on Windows 11, keeping agents within boundaries the operating system enforces.

AIBaseten explains how to decide which layers of a model can run in 4-bit NVFP4 without losing needed information. The post compares architecture-based heuristics, isolated-layer sensitivity scoring, and SaturationQuant, which accounts for other quantized layers. It also covers calibration with representative data and block-level scales of 16 values.

Why it matters: The post explains how to choose which layers run at NVFP4 precision using heuristics, sensitivity scoring, and saturation-aware scoring, with clear calibration steps.

AIInstinct, a startup AI agent that works through text messages, handled online tasks such as swim lesson searches, an eye doctor appointment email, and an Ikea return about as well as Muse and Dots, according to The Verge's testing. The startup was valued at $10 billion in late September, and it has no app or subscription fee for now, with access by invitation or waitlist. Its founder, Noah Shinn, says its focus on a personal assistant sets it apart from OpenAI and Meta.

AIGuizang says he turned his workflow into a Grok bot that users can install with one click. The bot runs on Grok's cloud virtual machine to collect content, write code, and render a daily morning AI news video without using a local computer.

AIArena's weekly update says Nano Banana 2.1 ranked in the top six across three Image Arena modes, with #4 in Multi-Image Edit at 1431 points. Mistral Large 4 placed #43 overall in Agent Arena, 11 spots above Mistral Medium 3.5, and Claude Haiku 5.5 (High) landed #30 in Code Arena WebDev at 1587 points, priced at $0.10/$0.50 per 1M input/output tokens. The post also introduces Arena's Alignment Index and announces a $200M Series B at a $3.1B valuation.

AIPerplexity says its Computer tool created Alexandria, a free, source-backed encyclopedia with 66,574 entries, for $25,000 in credits (2.5 million). Computer scanned and synthesized 348,980 sources to build it, and will maintain the site at alexandria.pplx.app.

AIBoris Power, who owns the OpenAI account, posted only the word "Wow!" with no details. Background from a separate post says the integer multiplication problem #109 witness value κ rose to 2⁻¹⁰·⁵⁴⁷ (about 6.6857 × 10⁻⁴), past the 2⁻¹¹ threshold. The author notes gains are now fractional and a major breakthrough is still needed.

AIAnthropic claims Chinese AI developers used fraudulent accounts and proxy networks to extract reasoning data from its flagship model, Claude. The company says some firms used Claude as a covert back end for their own apps. A joint advisory from the NSA, FBI and CISA last month, and US Treasury Secretary Scott Bessent's July warning about large-scale distillation, add to the allegations.

AISailPoint's Navigate conference in Austin, Texas, featured executives arguing that enterprises must secure AI agent identities at machine speed through just-in-time access and enforcement outside the agent. Mark McClain, SailPoint's founder and chief executive, said real-time decision-making is needed because manual administration cannot keep up. The event also covered the Entro Security acquisition and a partnership with AWS on Amazon Bedrock AgentCore, which grew 15-fold in the first six months of the year.

AIAmazon's Alexa Plus, powered by generative AI, now responds in three to five seconds and handles multistep smart home commands, cooking questions, and calendar imports more reliably than the original Alexa, according to a year-long test by The Verge. The reviewer says its personal assistant features remain underbaked and frustrating, and that ads on Echo Show displays are excessive. Alexa Plus costs $19.99 a month in the U.S. unless users have an Amazon Prime membership, and the Echo Dot Max is recommended as the ad-free option.

AIQwen has published Qwen-Image-2.1-Turbo on Hugging Face, an accelerated checkpoint of Qwen-Image-2.1 for text-to-image generation and image editing with 8 denoising steps. The checkpoint uses the same 7B visual generation architecture, loads directly with QwenImage21Pipeline in Diffusers, and includes its recommended sampling schedule. It defaults to CFG=1 and uses prefix KV caching to reuse text and reference-image context across steps.

AIThe Claude Blog published a guide to building scheduled agent automations with Claude Managed Agents (beta) that reads custom sources such as Slack and GitHub and posts a daily brief. The guide covers scoped vault credentials, per-source bookmarks so no window is lost or repeated, and confirming each Slack post before updating records. It also covers read-only access, a per-run spending cap, and a reference implementation with a Claude Code setup command.

AIZvi Mowshowitz reports that OpenAI released 722 math manuscripts from an internal frontier model on GitHub, later reduced to 719 after three withdrawals, covering 90 of the top 500 open problems. He says the work came mostly from a single prompt, with an average of three hours of compute per solution. Mathematicians reacted with mixed feelings, and the post highlights concerns about unread papers, cryptography implications, and the role of Lean verification.

Why it matters: The post traces how OpenAI's release of 719 math manuscripts divided mathematicians and reshaped verification, credit, and publication norms in the field.

AISimon Willison says he built a Newsletters index for his blog almost entirely by voice, using the ChatGPT desktop app's Codex voice mode while cooking dinner. The feature imports weekly Substack posts via RSS and undocumented API, monthly newsletters from a GitHub archive repository, and a private sponsors-only newsletter. He says he switched back to typing for review and fixes before deploying the pull request.

AIGoogle AI announces that SynthID.com is now available globally in English for verifying AI-generated images, video, and audio. The post also lists Nano Banana 2.1, EmbeddingGemma 2, Gemma 4 with BOTANIC-1, a Gemini business agent, and Guided Vision in Gemini Live.

AIMerve Noyan, of Hugging Face, says a new guide on speculative decoding is in progress, followed by one on quantization. She points readers to the Llama App docs page comparing prefill and decode.

AIAlibaba's Qwen team says Qwen-Image-2.1-Turbo is an accelerated checkpoint of Qwen-Image-2.1 on the same 7B architecture, now with open weights. It generates 2K images from text in 8 denoising steps and supports natural-language edits, with Pro and Turbo APIs also live.