All AI news

AIAustralian cloud-computing firm Firmus Grid scrapped its IPO, which had sought a $30 billion valuation. The company had only two operational data centers at the time.

AIOpenAI defended its decision to fire three safety researchers, Jasmine Wang, Tomek Korbak and Mikita Balesni, saying they committed a "significant breach of trust." The company said the dismissals were not about the researchers raising safety concerns, though it agreed with the letter they sent to board members and safety committees about preserving the monitorability of frontier models.

AIHarrison Chase says trajectory labeling should be split into several questions rather than one pass/fail judgment. LangChain's Jev judge in LangSmith Evals answers difficulty and correctness as separate typed outputs in a single pass on every trace.

AIOpenAI says it parted ways with researchers Jasmine, Mikita, and Tomek after an internal investigation found they violated policies on handling sensitive information. The company says the decisions were not about raising safety concerns, which it says it encourages, and that it has not terminated any employee for raising concerns. OpenAI also says it is finalizing contracts with third-party safety assessors and will announce details in the coming weeks.

AIMarkTechPost publishes a hands-on tutorial implementing RRSI (Regularized Recursive Self-Improvement), a method that lets an LLM agent revise its own harness around a frozen model. The full loop drafts edits with Claude Opus on Vertex AI and scores them in Docker benchmarks, but the edit-selection rules are plain Python that the tutorial runs in a simulated environment with a calibrated noise band.

AIA LoRA for MiniMax H3 removes a person from video by tracking them with SAM 3.1 and generating the replacement background in overlapping windows. Users supply the original video and a clean version of its first frame.

AIDoubao Work has added a Canvas feature for complex creative tasks, placing materials, design plans and outputs on one infinite canvas where users can keep editing text, colors and layout after images are generated. The update also integrates the lightweight Doubao 2.1 Lite model, aimed at everyday Q&A, document writing, spreadsheets and PPT creation, with optimized response speed and usage consumption.

AITRAE announced on October 9 that it has merged TraeWork and TraeCode into a single platform offering Agent mode and IDE mode with seamless switching between them. The upgraded product covers desktop, web, and mobile, letting users start tasks on a computer, check progress on mobile, and continue development back on desktop.

AIStar Motion Era's VPP2, a world action model, ranked first on the RoboDojo simulation leaderboard with a 32.26% average success rate and 39.26 average score. The article attributes gains to staged training that separates video prediction from action learning, and reports a 58.5% zero-shot success rate on a real ALOHA dual-arm robot versus 40% for π0.5. The code is open source on GitHub.

AISoftBank Group Corp. is seeking to raise as much as $100 billion from Gulf investors to expand its AI investments, the Financial Times reported, citing people familiar with the matter. The report did not specify the timing or terms of the proposed fundraising.

AIopenJiuwen, an open-source AI Agent platform built by Huawei teams with universities and enterprises, released and open-sourced an enterprise-grade AgentOS. It integrates multi-agent collaboration, self-evolution, compute affinity, multi-tenant isolation and security sandboxing, and a plugin architecture with the Agentic Hub ecosystem.

AIXPENG has named its robotaxi business XPENG YOYO and launched a ride-hailing mini program that lets invited members of the public test the service. The company says its first production robotaxi, based on the flagship GX model, rolled off the line in May 2026, and it has completed more than 2,000 internal test rides in Guangzhou. XPENG says YOYO uses four in-house Turing AI chips delivering 3,000 TOPS, a second-generation VLA model, and a vision-based approach without high-definition maps or LiDAR.

AIQwen3.8-Max, Qwen3.8-Flash, and Wan3.0 are available free for a week on GMI Cloud, which is extending the offer by seven days and raising rate limits across all three models. GMI Cloud is also running a contest where three winners each receive $200 cash plus $200 in GMI credits for the most creative, most challenging, or most effort-driven projects built with Qwen or Wan.

AIAdo Kukic posted a salute emoji reacting to Anthropic's launch of OSS Scanner, which uses frontier models to periodically scan opted-in open-source projects for vulnerabilities at no cost. The reports include a proof-of-concept, explanation, and suggested fix.

AIAnthropic has paused the Claude Team plan and $1,000 API credit offers for its Claude Startups program after underestimating demand, with hundreds of thousands of applicants. Claimed offers will remain in accounts, but some approved applicants who had not yet claimed their offers will lose access as applications are re-reviewed. Startup Stack and Applied AI office hours remain available to accepted members.

AIOpenAI's most senior Asia-Pacific public policy executive is leaving after six months in the role, according to a person familiar with the matter. The departure comes as the ChatGPT maker faces regulatory scrutiny in the region.

AIByteDance Seed researchers found that language models compressing their KV cache in fixed-size chunks retrieve the same information unevenly depending on token position. In a 128K-token needle-in-a-haystack test, base DeepSeek-V4 checkpoints differed by up to 40.2 percentage points by phase, and post-training narrowed but did not eliminate the gaps. The authors urge evaluating such models across positional phases, since high average accuracy can hide systematic failures.

AIHuawei has opened its DevEco Studio for HarmonyOS PCs to public beta, alongside first public betas of the AI tool DevEco Code and the agent toolkit DevEco CLI. The beta requires HarmonyOS 7.0.0.107 or later, at least 16 GB of memory and 100 GB of storage, and runs on several MateBook models and the MatePad Edge. DevEco Code ships with Zhipu AI's GLM-5.3 and GLM-5.1 models and supports third-party model connections.

AITencent Hunyuan, with Fudan and Tsinghua researchers, released ExplorationBench, a benchmark testing how AI systems explore through verifiable "Alien Worlds" with executable rules that conflict with familiar knowledge. Across 10 frontier systems, feedback mattered most: the best AlienCode run reached 89.0% after four rounds of probing, versus 0.5–11.0% without feedback. Answers are graded by an interpreter or proof checker rather than an LLM judge.

AICoreWeave is layering managed services over its infrastructure to address AI inference bottlenecks, including a preview capability called CoreWeave RL Rollouts that improved model reload latency by 15x versus a baseline configuration in testing. The capability is built on Nvidia's Dynamo framework, and the features are packaged into CoreWeave Forge, a platform that is free to start with paid tiers offering additional capabilities.

AIOpenAI expects to reach or exceed $70 billion in annualized revenue by the end of 2026, according to people familiar with the matter. The growth is driven largely by its enterprise business.

AITeknium called a Hermes Desktop demo "pretty sick" after Jonathan Bylos reported that Hermes Agent produced an intelligent UI embed during a design discussion without being asked. Bylos said the feature has been running in Hermes Desktop for a few days.

AIOpenAI published 719 AI-generated math proofs covering 372 result families, after withdrawing 3 for a symbol error. Reports say the release falls short of the AGMAI advisory group's standards, since it uses proprietary models, includes reasoning chains for only 10 manuscripts, and leaves about 42% unformalized. Terence Tao argues that rapidly solving famous problems harms the mathematical community's understanding and collaboration.

AIGoogle Cloud introduced the Gemini agent, a general office agent that can search, write emails, build slides, analyze data, run code, and coordinate sub-agents. It can take on an enterprise identity with email, calendar, and account, and it selects underlying models automatically, including Anthropic's Claude. The article presents this alongside OpenAI's Dots and Meta's Muse as competing office and personal agents.

AIJeremy Howard says Jasmine Wang accessed an executive's email that OpenAI had given her access to, despite her request that the access be removed and IT's failure to remove it. He implies she was fired because of that IT failure, not misconduct on her part.

AISnyk moved its internal support agent, Snyk Assist, into the core Snyk product in September 2026, giving every paying customer access. Built on LangChain and LangGraph with observability in LangSmith, the agent answers questions in plain language and can open support cases or log feature requests. It runs as a single agent behind Slack, web and API surfaces, with tools attached per user permissions.

AIAnthropic has barred users from exhibiting "sustained and needless abusive or cruel behavior" toward its models, according to a policy change first reported by The Verge. The San Francisco-based company says the ban does not apply to common user frustrations, model testing, or "dark creative themes." The change follows an August feature that lets Claude end conversations when a user is persistently harmful, which Anthropic framed as a safeguard for AI welfare.

AIAustralian AI data center operator Firmus, backed by Nvidia, has withdrawn its planned initial public offering, citing market volatility and conditions. Its board concluded the proposed terms did not adequately reflect the company's business strength and long-term growth outlook. Firmus said it will now pursue private market capital and consider other public and private options.

AIOpenAI's DevDay 2026 session demonstrates Codex shifting from a single-user tool to a team-oriented agent. The session shows a persistent personal agent investigating a 2am outage, from the first Slack message through a reviewed fix, using voice, Appshots, plugins, and meeting notes to keep the team informed.