A ChinaTalk investigation reported by The Decoder identifies at least nine US data firms selling roughly $500 million in AI training data to Chinese AI labs. Some of the data consists of reinforcement learning environments for tasks such as coding, financial analysis and office work. Buyers reportedly purchase environments previously sold to OpenAI or Anthropic, allowing them to avoid some in-house development costs and wasted training runs. The investigation’s authors called for export controls on dual-use environments, particularly in cybersecurity and biology, while saying Chinese labs still trail US peers by at least several months structurally.
Google is donating gVisor, including its name and trademarks, to the Cloud Native Computing Foundation (CNCF), shifting the project to maintainer-led governance. Modal, Ant Group and Tines are joining the maintainers, while OpenAI, Tencent and NVIDIA continue contributing. Modal says gVisor powers more than 40 million containers daily and reports that its RDMA implementation reaches about 99% of native bus bandwidth in NCCL benchmarks.
Bloomberg reports, as cited by Engadget and other outlets, that ByteDance plans to launch a Muse-like AI assistant in the coming weeks. The assistant's name and workings have not been detailed, ByteDance did not immediately respond to a request for comment, and it remains unclear whether companion hardware is planned. A separate iFanr report, citing Caixin and Bloomberg, says ByteDance started the internal Spell project in April to explore personal AI assistants. After Meta introduced Muse, ByteDance reportedly accelerated the project and planned to combine Spell with Doubao's conversational team to integrate related functions into Doubao.
Harvard physics professor Xi Yin left OpenAI after five months and announced on X that he was joining Periodic Labs, an AI startup building automated materials laboratories. Periodic Labs says its systems use AI models to direct robotic experiments intended to identify practical superconductors and magnetic materials, but it has not announced a new material. Yin has not disclosed his specific responsibilities at the startup.
InfoQ 中文 AI&大模型 reports that Manus replaced traditional programming tests with a four-hour task in which candidates use Manus to solve a real problem and explain their AI-assisted workflow. Manus co-founder and chief product officer Tao Zhang says product managers and designers on his team can now write code and submit repository changes, though the company’s job listings retain conventional engineering titles. A tech hiring analysis cited by Lenny Rachitsky adds that U.S. software-engineer hiring has fallen below half its 2022 peak, while FDE openings have increased sharply and AI-related keywords have become more prominent in product and design roles.
The WeChat account Bright Company reported that Anthropic began an IPO roadshow on November 10, one day after releasing Claude Fable 5.2. Based on a proposed share-price range of $70 to $78 and 825 million shares to be issued and sold, the report estimated an implied post-offering market capitalization of about $916 billion to $1.02 trillion. At the top of the range, its fully diluted valuation was about $1.4 trillion. The report said Anthropic recognized $33.24 billion in revenue during its first nine months, with about 80% generated from usage-based billing. It also reported $4.08 billion in adjusted operating profit and a $1.26 billion operating loss under U.S. GAAP.
Sure is a community-maintained fork of the archived Maybe Finance personal finance application, whose original team stopped developing it in mid-2023. The working app can be self-hosted with Docker and accessed through a browser, the macOS desktop app, the mobile app, API clients and LLM agents. Deployment options include a version without AI, one with OpenAI-backed AI settings and one with external assistant settings. Both Maybe and Sure are distributed under the AGPLv3 license.
Bloomberg, citing people familiar with the matter, reported that OpenAI generated $10 billion in third-quarter revenue for the first time, nearly 50% more than the previous quarter and almost three times the year-earlier figure. One source said the adjusted operating margin improved from negative 150% to negative 109%. OpenAI said on a call that annualized revenue stood at about $50 billion as of September. IT之家 attributed the renewed growth to stronger model competitiveness, product price reductions and a shift toward AI agents.
In an opinion article, Financial Times editor Roula Khalaf argued that the technology industry depends too heavily on Nvidia CEO Jensen Huang. She proposed creating a substantive president or chief operating officer, improving succession planning and addressing concerns about a board whose members have served for an average of 18 years. A separate report said Huang had departed and that Colette Kress was interim CEO. Venture capitalists, startup founders and semiconductor executives expected Nvidia’s startup-investment pace to slow under Kress, while people close to Safe Superintelligence and Thinking Machines Lab said investments covered by legally binding agreements signed months earlier should proceed.
SoftBank reported a $7.1 billion net loss for the July-September quarter, including a $3.7 billion impairment tied to power-generating equipment for U.S. AI data centers. After making advance payments to secure equipment amid anticipated shortages, SoftBank changed its procurement plans and canceled orders it could not use. The Wall Street Journal, citing unnamed people familiar with the orders, reported that SoftBank had reserved more than 13 gigawatts of gas turbines, mostly from GE Vernova. CFO Yoshimitsu Goto said the cancellations were voluntary and denied Elon Musk’s claim that SoftBank had defaulted on the orders.
OpenAI reported third-quarter revenue of $10 billion, up nearly 50% from the second quarter, according to TechNews citing investor disclosures. ChatGPT paid subscribers reached about 100 million, while adjusted operating loss widened to $10.9 billion; the company also began advertising in ChatGPT, lowered prices for its latest models and hired Dali Rajic as chief revenue officer to pursue enterprise customers.
OpenAI reported revenue of $10 billion in the third quarter, up about 50% from the second quarter, according to cnBeta citing investor disclosures. Its adjusted operating margin improved from -150% to -109%, but operating loss increased from $10.1 billion to $10.9 billion. The report also said Anthropic generated $17 billion in quarterly revenue while narrowing its operating loss to $567 million.
TSMC’s board approved two investment plans, according to IT之家: up to $500 million to subscribe for Anthropic common shares at its IPO offering price and up to $500 million to acquire OpenAI preferred stock. The reports also said Anthropic is planning an IPO with a proposed price range of $70–$78 per share.
A Reddit user says a recurring payment mandate for a deleted ChatGPT Pro account charged their bank account on November 6, 2026. The user reports that OpenAI’s support bot told them to recreate the account and submit a billing ticket, but the recreated account displayed no subscription history or active plan. No independent payment evidence or OpenAI response was supplied.
OpenAI says more than 20,000 transitioning servicemembers and veterans claimed a free year of ChatGPT Plus during the past year. The company and its community partners also hosted more than a dozen trainings for nearly 1,000 people in the military community, including a six-event series with Veterans Forge. OpenAI reports that participants use ChatGPT for tasks such as organizing job searches, rehearsing interviews and exploring civilian career paths.
Cloudflare CEO Matthew Prince said AI agents increasingly browse websites on users’ behalf, consuming server resources without sending the sites real visitors, advertising impressions or subscription revenue. Cloudflare is offering tools to identify, limit and manage AI crawler traffic, while working with OpenAI on change-based hints and promoting payment mechanisms in which AI services compensate websites for crawling or content use.
OpenAI engineers Dominik Kundel and Charlie Guo demonstrated new ways of working with Codex at an AI Engineer Code Summit workshop, according to Latent Space. The session covered cloud task handoffs, voice as a control interface for coding agents, and personal agents such as OpenAI’s dots, while also noting that the transition to cloud-based workflows remains confusing for some users.
Alibaba released Qwen Code 1.0 Preview, combining the Harness capabilities of Qoder and Qwen Code in an open-source AI coding agent runtime. The preview separates reasoning orchestration from tool execution through a unified event model and is available as a separate branch while the existing stable branch remains available.
36氪 reports that Anthropic is planning a Nasdaq IPO, with investor valuation expectations falling from $2 trillion to $1.5 trillion and then to discussion near $1 trillion as technology stocks adjusted, infrastructure financing costs rose, and investors reassessed its financial performance. The report also analyzes how Anthropic’s revenue growth, GAAP operating loss, adjusted profitability, and long-term compute commitments may affect valuation expectations for Chinese model developers such as DeepSeek, Moonshot AI, Zhipu AI, and MiniMax.
TechNews cited an Epoch AI estimate made in September that six major Chinese AI firms generated about $11 billion in combined annual revenue last year, roughly one-tenth of OpenAI and Anthropic's combined revenue. The estimate includes cloud and related AI revenue from companies such as DeepSeek, Zhipu AI, Alibaba and ByteDance, rather than only standalone AI product sales; ByteDance led the group with an estimated $4 billion. The report contrasted this with 2026 OpenRouter usage statistics showing DeepSeek with a 23.3% token share, above the 19.2% combined share of OpenAI and Anthropic. It also reported that the two U.S. companies accounted for 86% of global enterprise LLM API spending and attributed part of the Chinese firms' revenue constraints to price competition in the open-source ecosystem.
Google’s Android Developers account published Android Bench results for GPT-6.1 Terra, Kimi K3.1 Pro and DeepSeek V4.1 Flash, according to IT之家. With kimi-code, Kimi K3.1 Pro recorded a 24.0% pass rate and 82.3% average completion across 30 long-horizon tasks, taking 27.4 hours and costing US$317.4 per run. DeepSeek V4.1 Flash, evaluated with opencode, recorded a 17.3% pass rate and 71.2% average completion, with a 12.0-hour average completion time and US$9.9 run cost.
OpenAI’s case study says Gilbert + Tobin introduced ChatGPT Enterprise, ChatGPT Work and Codex after an initial period of tightly constrained experimentation, progressing from individual assistance to multistep agentic workflows with human review. The law firm reports 87% active usage among enabled ChatGPT users, while CEO Sam Nickless describes using ChatGPT Work to update files, create PDFs and assemble materials for a confidential review of 120 partners. These adoption, time-saving and cost figures come from OpenAI and Gilbert + Tobin and have not been independently verified.
DoNews reported that OpenAI's third-quarter revenue was forecast to grow 50% to $10 billion. The supplied report does not provide the forecast's source, underlying figures, or additional financial context.
At the launch forum of the Yau Mathematical and AI Laboratory Alliance, Shing-Tung Yau, Nanhua Xi, Zhiqin Luo and Nicolai Reshetikhin discussed AI-assisted mathematics, mathematical intuition and how students should use AI. Yau argued that AI is rapidly taking over the task of solving specific mathematical problems and said finding a correct result matters before fully understanding it, while Luo emphasized that students must understand and critically assess AI-generated answers. The discussion also covered OpenAI's claimed advances on major mathematical problems, though those claims were described as unverified and are being carefully checked.
A Reddit user said GPT Image 2.6 Extra High was unusable for basic image-generation tasks, citing poor reference-image handling, incorrect posing with multiple characters, false safety-filter flags, and distorted character proportions. The same post said AI influencers on Twitter were praising the model as a clear upgrade, while noting the poster used a Pro plan and extra-high thinking; these are unverified user impressions rather than independently tested findings.
The article presents AI Coding as a repository-aware software-generation approach that goes beyond line-by-line IDE completion. Its workflow parses intent, retrieves project context, plans modifications, generates code, and runs static checks and unit tests, with failed results fed back for bounded retries. It also discusses task sizing, AST-based indexing, sandboxing, human review, and risks such as hallucinated dependencies, weak tests, and context leakage.
OpenRouter Blog published a guide to diagnosing and handling 429 Too Many Requests errors returned by OpenAI, Anthropic, and OpenRouter. It recommends reading rate-limit headers, retrying at the server-specified reset time, using model fallbacks, or routing requests to another provider; its Python and TypeScript examples were tested against the OpenRouter API.
JioStar built a conversational assistant into JioHotstar using the OpenAI API, allowing viewers to use natural speech to find movies, shows, historical highlights, live-match moments, and cricket statistics. JioStar’s chief architect said the voice-visual experience supports more than 15 languages and was built in about eight weeks, with Codex used to accelerate evaluation-case generation and system updates.
A Reddit user said they verified a domain and entered business details for ChatGPT Business Hub, but the business-contact field repeatedly reported that it could not send a verification code. After skipping the contact step, the setup remained in draft, and a later verification attempt returned an error asking the user to try again. The report does not establish whether government ID is required or whether other users encounter the same issues.
Leaked documents cited by Berber Jin of The Wall Street Journal show that OpenAI's revenue grew roughly 50% from the second quarter to the third quarter, reaching $10 billion. Its adjusted operating margin improved from -150% to -109%, but the company remained far from profitability.
What problem it solves
Pi is designed for developers who need to adapt an agent to their own workflows rather than work within a fixed feature set. It provides interactive, automated, RPC, and SDK interfaces while leaving optional capabilities such as sub-agents and plan mode to extensions or packages.
Key features
Extensible agent harness — Customize the harness with extensions, skills, prompt te…
What problem it solves
Building LLM applications often requires connecting model providers, vector stores, tools, streaming, and agent capabilities. Rig provides a shared Rust interface for these components and includes a terminal coding agent that can be extended with plugins.
Key features
Unified model and vector-store interfaces — Use a shared interface for model providers and vector stores, wi…
What problem it solves
Managing multiple AI service sessions and translating them into compatible API interfaces can require repetitive account switching, custom routing, and protocol-specific integration. The project addresses these tasks with a desktop application that combines account administration, protocol conversion, request scheduling, and model routing.
Key features
Account dashboard and …
What problem it solves
Dodaios provides a locally installed, open-source way to make a personal AI assistant the control center for work on your computer. Its lead assistant coordinates models, coding agents, and computer use, while chats, memory, documents, and settings remain files on the computer that runs Dodaios.
Key features
Voice, chat, and phone interaction — The lead assistant supports wa…
What problem it solves
Managing subscriptions across multiple AI providers can require separate access paths and fragmented quota handling. This project provides a centralized gateway for distributing subscription quotas, supporting shared access and compatible use with native AI tools.
Key features
Subscription quota distribution — The gateway is designed to distribute and manage AI subscription …
What problem it solves
Developers need a unified way to integrate ChatGPT website capabilities into applications that expect OpenAI-compatible endpoints. They also need account-pool management, proxy configuration, import workflows, and a self-hosted interface for creating and reviewing media.
Key features
OpenAI-compatible endpoints — Provides compatible routes for image generation, image editing…
What problem it solves
Stock analysis often requires users to gather market quotes, news, technical indicators, financial information, and AI-assisted interpretation across multiple sources. The project addresses this by bringing supported-market data, analysis tools, AI agents, alerts, and review workflows into a locally focused desktop application.
Key features
Multi-market market data — The app…
What problem it solves
Large language models can be difficult for individuals to train, deploy, or understand because of their scale and the abstractions used by many frameworks. The project addresses this by providing native PyTorch implementations, compact model structures, open datasets, and staged workflows for pretraining, fine-tuning, preference optimization, tool use, and agentic training.
…
What problem it solves
AI agents may exceed their role, take unauthorized actions, or fail to follow organizational goals and policies. Teams need a way to test whether an agent behaves as intended across realistic and adversarial requests.
Key features
Business-aware simulations — Describe an agent, build a simulation environment for it, and exercise the agent with realistic requests that may exc…
What problem it solves
People often want an assistant that can participate in ongoing conversations, remember personal context, and use tools without requiring a cloud AI service. Jarvis addresses this with local-first voice and text interaction, personal memory, optional connected tools, and a setup process for locally controlled models.
Key features
Conversation-aware voice assistance — Jarvis m…
What problem it solves
Running and training models locally requires tools for model execution, fine-tuning, agent integration, and deployment across different model formats and hardware. The project brings these workflows into a desktop app, a web UI, and a code-based version.
Key features
Run and train models — Run and train language, diffusion, decision, embedding, audio, image, and video models…
What problem it solves
Personal and team knowledge is distributed across notes, conversations, trackers, and research, making it difficult to capture, organize, verify, and synthesize useful information. COG provides a markdown-based workflow in which agents can maintain a shared knowledge base, support recurring research and operational tasks, and verify selected work through independent checks.
…
What problem it solves
Configurations for the Codex desktop app, CLI, API providers, prompts, Sessions, Skills, MCP, and TOML files can become scattered across multiple locations and formats. Codex-X consolidates these recurring configuration tasks into a visual desktop interface so the current state is visible and common operations can be completed directly.
Key features
Visual prompt management …
What problem it solves
Serving large language models requires efficient memory management, flexible model execution, and practical integration with existing models and APIs. vLLM addresses these needs with inference and serving capabilities spanning batching, caching, quantization, distributed execution, and multiple hardware platforms.
Key features
Efficient memory and serving — vLLM manages atte…
What problem it solves
Generative AI systems and applications can contain security and safety weaknesses that require structured testing across models, endpoints, and complete applications. PyRIT addresses this with repeatable assessments, interactive exploration, and extensible workflows for red teaming these systems.
Key features
Repeatable command-line assessments — The scanner runs repeatable …
What problem it solves
Developers need Vietnamese speech synthesis that can run locally, produce natural reading styles, and support voice cloning without sending audio to a hosted service. They also need integration paths for SDK use, streaming applications, and an OpenAI-compatible API.
Key features
Instant voice cloning — Clone a voice from a short reference recording. The clipping pipeline aut…
What problem it solves
AI coding tools often expect particular provider protocols, making it difficult to use the models and providers a user prefers. A local gateway can translate between supported client APIs and provider APIs while preserving capabilities such as streaming, tool calls, reasoning, and images.
Key features
Provider protocol translation — The gateway translates Responses, Chat Com…
What problem it solves
AI agents often send large tool outputs, logs, files, RAG chunks, and conversation history to language models, increasing the input context that must be processed. Developers need to reduce that input while preserving important content, keeping compression local, and retaining access to original material when necessary.
Key features
Content-aware compression — A ContentRoute…
What problem it solves
The hidden system instructions that shape AI product behavior are difficult to inspect and change across models, products, and versions. This repository makes collected prompts available in a structured, product-oriented archive for study and comparison.
Key features
Broad product coverage — The archive covers prompts from major AI providers and products, including Anthropic…
What problem it solves
Image creation often requires switching between separate tools, manually translating visual ideas into parameters, and losing the context of an ongoing conversation. This project brings generation, editing, reference workflows, model comparison, and asset management into a DeepSeek Harness conversation and canvas.
Key features
Chat-Based Image Creation and Editing — Describe…