All AI news

AIPerplexity has released pplx-embed-v2-late, a pair of ColBERT-style multimodal embedding models in 0.6B and 9B sizes that retrieve text, images and rendered PDF pages in a shared embedding space. Both are available on Hugging Face under the MIT license, while a hosted API endpoint is planned but not yet live.

AIHowie Serious (@howie_serious) says his Grok bot has found its first powerful use case: a personal agent that collects Twitter information. He argues this beats humans scrolling phones, manually gathering posts, or getting absorbed in endless feeds.

AIThe post announces DevDay 2026 and frames its theme as a shift from single-player to multiplayer agents. It gives no further details about specific products, features, or dates.

AIAnthropic is halving the price of cache reads on Claude Sonnet 5.5, to $0.10 per million tokens. The company says this makes Sonnet 5.5 about 20% cheaper to run on most long-running work, where agents rely heavily on cached reads.

AIClaude Code v2.1.294 fixes prompt and agent hooks written as instructions, which had allowed actions they should block. It also improves how prompt hooks on Stop and SubagentStop are judged, making Claude less likely to stop early.

AIMicrosoft announced that its Surface Laptop Ultra will launch October 18 with a starting price of $2,599 for the lowest-tier configuration, rising to $6,000. The 15-inch laptop runs Nvidia's RTX Spark processor on Windows on ARM, with up to 128GB of unified memory and a 2,000-nit HDR display. The article notes that five RTX Spark laptops from Asus, Dell, HP, Lenovo, and MSI are also available for preorder starting at $2,599.

AIDeveloper Dex Horthy highlights Fold, a Effect-based agent harness by @0xblacklight that stores state as a portable append-only log. It runs on Cloudflare Durable Objects out of the box and includes a retro OpenTUI interface with 10 themes.

AIA South China Morning Post reporter spent a week using Huawei's Mate XT 2, a US$3,500 trifold phone with a 10.2-inch unfolded display. The source excerpt focuses on the device drawing attention at a family dinner during China's National Day "golden week" holiday in early October, with no further specifications or verdict provided in the available text.

AIEvery moved its Slack-based Every Agent from its own OpenClaw fleet to Anthropic's Claude Managed Agents, letting the team focus on building a shared coworker. The newsletter, written by engineer Paridhi Agarwal, covers the gains and trade-offs of stopping self-run agent infrastructure.

AIMastra launched the Mastra Agency Program, a network of certified agencies and consultancies that build Mastra agents for clients. The launch includes five partners: Deerfield Group, Blue Drop Labs, Frontleap, Handpicked, and Young Security. Every partner has been vetted by Mastra's FDE team and receives direct access to Mastra's leadership and regular roadmap updates.

AIAnthropic has launched the Anthropic Cyber Mission, which starts with the Critical Infrastructure Defense Program for operational technology and OSS Scanner for open-source projects. The defense program brings frontier Claude models, on-site engineers and threat research to trusted providers such as Accenture, CrowdStrike and Palo Alto Networks. OSS Scanner gives enrolled open-source projects periodic free scans from its strongest models, with reports sent without human review and an expected true-positive rate above 90%.

Why it matters: The announcement shows how a frontier AI lab is packaging cyber defense around critical infrastructure and open-source maintainers, including the program's partners and access routes.

AIAnthropic has published a 2026 update to its Usage Policy, taking effect November 12, mostly to clarify existing rules for longer, more autonomous Claude work. The changes consolidate deceptive-campaign prohibitions into a new section, narrow the elections rules to voter deception and disruption, and explicitly ban weapons-related software and surveillance tools. Requirements for high-risk uses and for models connected to autonomous physical hardware were also tightened.

AIAnthropic is launching OSS Scanner, an opt-in service that runs periodic security scans of enrolled open-source projects using its strongest models at no cost. Its outputs are fully model-generated without human review, so some reports may be incorrect or invalid, though a pilot found 85 of 97 checked critical and high-severity findings met Anthropic's disclosure bar. Core maintainers of eligible projects can enroll through a GitHub pull request.

AIClaude now turns company data into dashboards that stay current, and it can build animated explainers from a prompt. Dashboards connect to BigQuery, Databricks, Snowflake, and Salesforce in beta on paid plans, while Motion is in beta on Team and Enterprise. Docs, Slides, and Design are out of beta and available on every plan, including Free.

Why it matters: The post specifies which data platforms connect, which features move out of beta, and where admins control access, clarifying what changes for enterprise workflows.

AIJohns Hopkins astrophysicist Brice Ménard, working as an Anthropic researcher, used Claude Science to produce the first complete map of the sky in ultraviolet light. Claude orchestrated agents to merge GALEX, Swift, and FIMS/SPEAR data, then predicted roughly a third of the sky that no UV telescope had observed, using relationships to visible, infrared, and radio data. Hidden test regions were reconstructed to within about 10% of real measurements, and each pixel is labeled measured or predicted with uncertainty estimates.

Why it matters: The post shows how an astrophysicist used Claude Science agents to merge UV surveys and predict missing sky regions, with a validation step that makes the method reusable.

AIAnthropic is committing $150 million over three years to the Genesis Mission, a federal initiative to accelerate scientific and technological discovery through AI. The funding will make Claude available to more than 15 participating agencies, including NASA, the National Institutes of Health, and the National Science Foundation. Over the next three years, Anthropic plans to provide Claude, Claude Code, and API credits to several hundred Genesis Mission research projects.

AIArtificial Analysis is adding trusted-access models to its Cyber Index, starting with GPT-6 Sol (Daybreak Blue, max), which is available only through OpenAI's Daybreak program. The model hits no safety blocks across the Index and scores 32 points higher overall than the publicly available GPT-6 Sol (max), with its largest gains on CyberGym-E2E.

Why it matters: The source shows how safety refusals shape cyber benchmark scores, with the trusted-access model's gains concentrated on CyberGym-E2E, useful for comparing guarded and unguarded models.

AIBlock's AI capabilities lead describes using Claude Fable to plan large code migrations and direct smaller models like Opus and Sonnet on individual tasks. He says Block routes frontier and smaller models by task and keeps merges and production deploys behind human dual approval.

Why it matters: Block's engineering lead describes how frontier models orchestrate large migrations and how access, effort levels, and safeguards are managed across an organization.

AILuma announced that Claude Motion animations can now open directly in Luma through an MCP connection, letting creators restyle them and reframe them to 9:16, 1:1, 4:3, or 21:9. Claude Motion, currently in beta on Claude Team and Enterprise plans, generates animated explainers from prompts, while Luma's Ray and Uni video models produce final video files.

AIMIT associate professor Christina Delimitrou is applying machine learning to make large-scale data centers more efficient, secure, and reliable, rethinking how servers and networking equipment operate. Her group redesigns outdated cloud systems, manages shared hardware resources, and creates streamlined server architectures so operators can extract more computing power from existing hardware. She also uses AI to help programmers find and fix problems in cloud-based applications, reducing downtime that hampers performance and drains resources.

AIHarvey LAB-AA v1.1 adds hallucination checks that audit every model deliverable against task source documents, with material hallucinations zeroing a task's score. GPT-6 Astra averaged 0.03 material hallucinations per task across 120 tasks, while Gemini 3.8 Flash averaged 13.96. Harvey uses GPT-6 Sol (high) as the hallucination checker, separate from its three-judge rubric panel.

AILangChain built Restock, a sample office-supply agent that runs in Slack on Managed Deep Agents and pays through Stripe's Link wallet. The agent searches products, builds a cart, and pays over the Machine Payments Protocol, with the user approving the purchase in Slack and the payment in Link. The post uses a pens order at $22.18 to show the flow from request to confirmed order.

Why it matters: The post walks through how an agent handles search, budget limits, Slack review, and Link approval, showing where each control sits outside the model.

AIOpenAI's Tibo confirmed that the GPT-6 rollout has landed across all accounts and asked how things are going so far. In a previous post on Day 3, he said GPT-6 is coming to Chat, and that Codex and ChatGPT Work reached 40M active users, a new high.

AIOpenAI published 722 math manuscripts covering 372 result groups in a new GitHub repository, openai/math, all produced by an unreleased internal model. The author describes the results as including a near-Riemann hypothesis claim pushed to 0.875, and notes that 25 Fields Medal winners criticized the company's approach to AI math research.

Why it matters: The piece traces how AI math results moved from benchmarks to open problems, offering context on verification and the mathematicians' pushback.

AIWindows sparked a platform shift that created a new industry for NVIDIA. Then we invented programmable shading GPUs for DirectX, which led to CUDA. Then we partnered to bring GPU supercomputers to Azure, which helped OpenAI train GPT. That collaboration inspired us to reinvent Windows for the age of personal agents. 4 years. Thousands of engineering years between us. So proud of what we built together.

AIScott Aaronson reports, based on his sources, that some AI companies have begun discreetly investigating whether their latest internal models can break important cryptographic protocols and primitives. He notes that cryptography is conspicuously absent from OpenAI's list of 376 papers, and the quoted post adds that the US government has censored academic quantum cryptanalysis results.

AIElevenLabs is officially launching in Singapore, opening a new Southeast Asia hub that is already working with businesses including Funding Societies and Atome. The company is also hiring in Singapore, with open positions listed on its careers page.

AIByteDance ranked first in the debut SCMP Plus Going Global 100 ranking of mainland companies' overseas expansion. Foxconn Industrial Internet, Shein and BYD placed second through fourth, according to the databank, which assessed more than 2,000 companies using weighted scores.

AIA NVIDIA study accepted at NeurIPS 2026 reports that multimodal models refuse harmful requests less reliably when they call tools. Refusal failures rise by up to 68.7% relative and by 17.7% on average across the models tested, including Claude Opus 4.6 and 4.7 and Gemini Agentic Vision. The authors attribute this to tool outputs crowding out the original harmful intent and to attention shifting toward describing tool results. Re-inserting the original request and image before the final response restores part of the lost refusals.

AIMax Zeff highlights new data showing OpenAI's growing momentum with business customers, though the post itself provides no specific figures. The post refers to a quoted post from Angel Au-Yeung, which supplies the underlying context but is not reproduced here.

AIStanford's CS 329Z: Engineering AI Agents course has published slides for lectures 4 and 5, following the earlier release of lectures 1–3. Lecture 4 covers tool use and is taught by Diyi Yang, while lecture 5 covers frameworks and orchestration, taught by Diyi Yang, Michael Ryan, and John Yang.

AIThe vLLM team released a technical report on vLLM-Omni, a unified serving runtime for omni-modality generation spanning multi-stage autoregressive pipelines, iterative diffusion, and stateful sessions. Current LLM servers and diffusion stacks each cover only one of these patterns, pushing deployments to stitch disjoint runtimes together. vLLM-Omni offers a shared control plane in which an orchestrator advances requests across stages, specialized engines handle compute, and a connector carries payloads.

AIAndrew Curran posted a brief "Updated again." with no further details. The post links to background from @0xdoug on OpenAI problem #109 (integer multiplication), which reports tightening the bound from κ = 2⁻¹⁸² to κ = 2⁻³⁴, a roughly 48-million-fold improvement.

AICodex is giving paid accounts banked reset cards today, according to Guizang's post, and the main post's author refers to it as a celebration day. The quoted post from Tibo Sottiaux says the day's headline is GPT-6 in Chat, alongside a new high of 40 million active users across Codex and ChatGPT Work.

AIMicrosoft has repositioned Windows as the home for hybrid intelligence, where AI agents can run locally or in the cloud. The announcement covers four layers: MXC reaching general availability for agent isolation, local models such as MAI Code 1.1 Flash, Copilot on Copilot+ PCs gaining local context and actions in coming months, and new hardware including RTX Spark PCs and DGX Station for Windows.

AIMemory has overtaken compute as the central control point in IT, according to The Next Platform, as generative and agentic AI drive demand for DRAM, HBM, and flash. Server DDR5 memory now sells for roughly 9X to 13X its November 2022 street price, while a 30 TB enterprise SSD costs 6X to 7X more. HBM pricing has risen only about 1.6X since the GenAI boom began, the article says.

AIHewlett Packard Enterprise unveiled the first of its ProLiant Gen 13 systems, built for enterprise AI inferencing and agentic workloads, with AMD 6th Gen Epyc 9006 "Venice" CPUs in common. The ProLiant DL585a, a 10U server holding up to eight double-wide GPUs and two Epyc CPUs with up to 256 cores each, will be available in March 2027. The air-cooled ProLiant DL525, a single-socket 1U system with a 256-core AMD chip, becomes available next month.

AIThariq, of Anthropic, says the most common failure in AI coding is users working outside their domain who cannot write precise prompts and plans. As a result, they spend many turns iterating imprecisely. The quoted post contrasts this with programmers' advantage in vibe coding.

AIOpenAI is rolling out GPT-6 to ChatGPT's over 1.2 billion weekly users, adding Intelligent UI, which lets replies include charts, buttons, forms, and interactive tools. The post's image cites tiered access, with Free/Go and Plus/Pro/Business/Enterprise sharing the Sol and Luna model splits, and says the feature is progressively rendered as the model generates it.

AIStep Terminal will unveil its first large-model-native agent smartphone, the STEPX Neo, at a "Ready Builder One" launch event in Shanghai on October 13. The company says the device is built agent-native across its model, system and hardware, and the event will also announce the latest progress in its ecosystem partnerships.