fal announced a new partnership with Microsoft that brings fal to Microsoft Foundry. The companies said the partnership will add video models including Kling, LTX, and MiniMax, giving developers more options for video and multimodal applications.
A Microsoft patent application, US 2026/0341309 A1, publicly disclosed on October 29, 2026, describes a near-real-time AI system that could replace, delete or rewrite words in livestream audio and adjust the speaker’s lip and facial movements to match the altered speech. TechSpot, citing Patentlyze, reports that proposed uses include translation, removing pauses and filler words, and filtering profanity. The reports say separate configurations could retain an unmodified recording as a “ground truth,” while some Chinese coverage says the patent proposes preserving original audio and video records and, in some configurations, digital markers of edits. The filing does not state whether viewers would be notified when the first claim’s edits occur. It is a patent application, not a product release, and does not establish that the technology will be commercialized.
TechRadar's reviewer tried Snap's sixth-generation Snap Specs AR glasses ahead of their UK launch, reporting that the glasses do not need to be tethered to a smartphone and can top up via a portable charging pad. Setup took only seconds, with on-glasses arm buttons for recording and resetting, and the reviewer typed in Google Docs using a virtual keyboard that can also pair with a Bluetooth keyboard. Other highlights included capture of up to 30-second video clips and stills, a built-in browser and media viewer, a fast voice AI assistant, Google Maps in a heads-up display, and a live speech-translation overlay that the reviewer called a possible killer app for business meetings. The reviewer was less convinced about the workplace case overall, noting that many demos were entertainment or education oriented, and priced the glasses at £1,999.
Microsoft co-founder Bill Gates said taxation based on wealth levels, including measures similar to California's billionaire tax, may be appropriate for the AI era. He argued that governments should bear primary responsibility for addressing inequality while charitable giving remains voluntary, and also proposed taxes on AI tokens and robots. IT之家 cited Fortune for the policy views.
Anthropic reportedly filed for an IPO that could value the Claude developer near $2 trillion, while the filing disclosed rapid revenue growth alongside nearly $81 billion in first-half net losses. Bank of America, Citi, Morgan Stanley and Blackstone also announced a $63 billion special-purpose vehicle to buy Broadcom AI chips and networking equipment and lease them to Anthropic.
Censys found more than 800 Artex deployments accessible from the internet without passwords, according to TechCrunch; TechRadar reported an updated scan of 1,465 exposed hosts. Artex is an open-source autonomous penetration-testing tool that combines AI agents with reconnaissance, and some exposed instances also ran offensive tools such as Cobalt Strike, Vshell, or Adaptix C2. Microsoft and CrowdStrike have linked Artex use to attacks involving South Korean financial institutions and ConnectWise Cloud Backup customer environments, while Censys said not every exposed instance was malicious.
Microsoft plans to reduce the number of neoclouds it counts as partners, select key suppliers and focus more on its own data centers as its AI infrastructure buildout shows signs of strain. The report does not identify which partners will be retained or discontinued, or specify the timing and scope of the reductions.
Bloomberg reported that Microsoft is winding down a 2023 arrangement to rent AI computing capacity from Oracle after declining to renew the contract. People familiar with the matter said Microsoft has been switching off the rented capacity in recent months because the servers used older NVIDIA chips that no longer matched its changing technical requirements. The arrangement provided Oracle servers for AI-powered Bing, and its full financial size is unknown. Oracle said demand for its AI infrastructure, including older chips, remains strong, while Microsoft declined to comment.
Push Security reported a phishing campaign using fraudulent invitations for nonexistent OpenAI Academy livestreams, a private SpaceX AI-marketing briefing, and an Anthropic presentation at AWS re:Invent. The pages promised ChatGPT Pro, Grok Heavy, or SpaceX shares, then used Kali365 and VoidProxy to capture Microsoft or Google credentials, MFA responses, and authenticated sessions.
Voters opposed 40 of 41 state and local data-center measures in the U.S. midterms, while county commissioners in Georgia postponed a vote on a new data-center site and a Microsoft project in New Jersey faced major permitting delays. The report also cites GlobalData, which said public opposition and structural issues such as grid-connection delays have made the average data-center project take 33% longer to complete. AI infrastructure and chip stocks fell as investors reassessed data-center costs amid political opposition and elevated interest rates.
CBC News reports that university libraries across North America, Europe and Oceania are facing persistent AI-related scraping traffic that can slow or crash public-access sites. The University of North Carolina at Chapel Hill reported more than 11,000 searches for Finnish music in one day, while its catalogue periodically blocked students, faculty and staff because of bot traffic. Libraries, museums and other cultural heritage institutions are developing ways to preserve access for people and benign services such as search engines and web archives.
NavInfo held its 2026 EVOLUTION conference in Shanghai on November 6, where CEO Cheng Peng announced a strategic expansion from automotive intelligence into mobile agents, positioning the company as a "super partner" for mobile agents. The company also released an AI-native smart cockpit solution for global markets and discussed a partnership with Microsoft, while citing figures such as 270万 units of cumulative designated orders for its sub-1,000-yuan parking-and-driving solution and more than 100 million cumulative shipments of AutoChips SoCs. Other items included the sixth-generation AC8035 SoC planned for 2027 Q1 and the ASIL-D AC7870 MCU, with performance and shipment numbers coming from the company itself.
Protesters entered the construction site of a Microsoft data center in Amsterdam on November 6, with some climbing a crane and displaying anti-AI banners. A Reuters witness estimated that around 100 people entered, while activist group Geef Tegengas claimed about 500; Dutch police reportedly detained 150 to 200 people suspected of trespassing. The protest formed part of coordinated European demonstrations against large data centers and included objections over energy use, environmental impact, housing and Microsoft's role in technology used in war.
TechRadar interviewed four people with different levels of technical expertise who switched to Linux within six months to regain control over their privacy and data. They cited frustration with Windows 10 support ending, Microsoft’s promotion of Windows 11, and increasingly intrusive data collection, while friendlier Linux interfaces and LLM assistance helped them configure their systems. The interviewees still advised checking potentially incorrect or destructive AI instructions and using privacy-focused or local models once personal files are stored on a device.
Microsoft released VibeVoice-ASR-Streaming, a unified streaming automatic speech recognition model that transcribes speech as audio arrives. According to Microsoft’s VibeVoice GitHub repository, it identifies who said what, supports customized hotwords and 10 languages, and emits text for each audio chunk.
IT之家 reported that Microsoft Windows Services and Delivery partner director Nuno Costa said Windows 11 security had not deteriorated because AI-assisted, broader code scanning found more vulnerabilities that might otherwise have been overlooked. Microsoft said it fixed 1,413 CVEs in October 2026, including 1,123 Windows-specific patches, compared with 134 Windows-specific patches in October 2025. The company warned that attackers could use AI to analyze published fixes and locate exploitable weaknesses within hours, urging users not to delay updates.
Microsoft plans to add an opt-in Copilot integration to the redesigned Windows 11 Search interface before the end of 2026, according to IT之家, Wccftech, TechRadar, and DoNews. The feature is expected to be disabled by default, provide inline quick answers in the taskbar search box, and offer access to the full Copilot app for deeper conversations; rollout will begin in selected markets before expanding further.
Scott Hanselman released gputaskman, an open-source GPU monitoring tool that shows per-process GPU usage, VRAM consumption, temperatures, power, frequencies, codec activity, and PCIe traffic across Windows, Linux, WSL, and Apple Silicon macOS. It also exposes AI/LLM details such as the model, runtime, weights, quantization, parameters, and memory use; Tom’s Hardware tested version 0.3.0, while other reports describe the project as experimental and note occasional stale temperature readings.
TechNews reported that Citi maintained a buy rating on GUC and raised its target price to NT$8,000, citing growth from CPU business involving Google and Microsoft. The report also said Citi raised its EPS forecasts for 2026, 2027 and 2028 by 7%, 11% and 61%, respectively, while expecting stable order visibility across GUC’s AI ASIC projects and further N3 product ramp in 2027.
KYEC reported third-quarter revenue of NT$12.142 billion, up 9% sequentially and 31% year over year, with operating performance matching market expectations. The company reduced its fourth-quarter sequential revenue-growth guidance to approximately 7%–10% and expects gross margin of 36%–37%, citing HBM4 yield-related delays for some new GPU production schedules, weaker smartphone and consumer-electronics demand, and higher depreciation and labor costs. Strong AI ASIC and high-performance CPU orders, including Google v8t/i TPUs and Nvidia, Google and Microsoft CPUs, were reported to offset part of the GPU delay.
US AI, semiconductor and data-center stocks sold off sharply on November 5, 2026, reversing gains from the day after the midterm election. The Nasdaq Composite fell 5.5% and the Philadelphia Semiconductor Index about 12%. Nvidia dropped roughly 9%, while AMD and Micron fell more than 17%. Investors weighed tighter scrutiny of data-center permits and resource use, higher financing costs and uncertain returns on AI infrastructure spending. Cited analysts said the decline did not establish that the AI bubble had burst or that demand had entered a systemic decline.
Google is rolling out a voluntary severance package as Microsoft, Oracle and Meta also reduce their workforces, according to 36Kr citing Sina Finance. The report says layoffs and buyout departures at large technology companies have exceeded 100,000 this year. It cites AI-driven efficiency gains and heavy data-center spending as possible reasons for the layoffs, but does not establish either as a confirmed cause.
Microsoft is reportedly increasing oversight of data center projects involving Nebius Group, IREN and Nscale, exiting projects it considers high-risk and discussing whether to sign additional new-cloud deals. Citing people familiar with the matter, the report says some Microsoft executives have expressed regret over commitments as new-cloud providers face project delays, regulatory obstacles and high costs. Microsoft previously signed tens of billions of dollars in contracts with emerging data center operators to meet demand for cloud and AI services.
Bloomberg sources report that Microsoft is tightening oversight of some data center projects, stepping away from projects it considers risky, and debating whether to sign new neocloud deals. The report says Microsoft has helped create the neocloud boom by signing tens of billions of dollars in contracts with emerging data center operators.
Microsoft is increasing oversight of data-center projects being handled by IREN, Nebius and Nscale amid delays and rising costs, Bloomberg reported, and is reportedly exiting some higher-risk projects. IREN said Microsoft-requested network-design changes and supplier delivery timing delayed infrastructure in Texas, pushing its US$4 billion-plus annualized revenue target from year-end to January, with a possible delay to late February. IREN also plans about US$15 billion in capital expenditure for computing equipment and infrastructure during the rest of its fiscal year.
A Reddit user said they used Claude to inspect a large batch of spam and phishing emails, extracting each link’s real destination without opening it. The user reported identifying deceptive addresses, naming suspicious cloud infrastructure, and filing abuse reports; one week later, they said no new scam emails had reached their mailbox.
A commentary from Miaotou APP says investors interpreted the U.S. midterm election results as increasing scrutiny of data-center construction, contributing to selling in neocloud, optical-networking and data-center stocks. The report cited gains in Apple, Amazon and Microsoft and declines in CoreWeave, Micron Technology, Nvidia and Oracle. It characterized this divergence as market pricing for concern that AI capital spending may slow, while emphasizing that major technology companies had not already cut AI investments across the board.
At TrendForce’s 2027 technology industry outlook forum with Capital Securities, senior analyst Chung Ying-ting said the global foundry industry could grow about 30% in 2027, extending the strong momentum of 2026. AI and general-purpose server demand was cited as a driver. Eight-inch wafer capacity is expected to remain tight, while TSMC and Samsung have full utilization and advanced-node order visibility extending through the end of 2027. Twelve-inch mature-node utilization is forecast to rise only modestly because of new capacity. Senior analyst Hsu Chia-yuan said AI server demand could lift 2027 server CPU shipments by at least 30% and potentially 40%-45%. DRAM bit supply growth is expected to slow to 24%, while servers already account for more than 60% of DRAM bits consumed; Microsoft’s server DRAM purchases alone exceed total global PC DRAM demand. The forum also identified wafer supply, lasers, packaging and testing as four obstacles to volume CPO production in 2028. Capital Securities separately presented NT$3,200 as a possible TSMC price scenario, not a realized result.
The Verge AI reports that Amazon, Microsoft, Alphabet, and Meta plan to spend at least $720 billion on capital expenditures this year, mostly on AI infrastructure, alongside at least $1 trillion in future commitments. Revenue is growing at Amazon, Meta, and Google Cloud, but The Verge AI says these gains may not yet justify the scale of AI spending. All four companies continue to support their data-center plans, although regulatory changes could affect them.
TechRepublic recommends 16GB of RAM for most users, while 32GB may better support demanding applications, heavy multitasking, and local AI models. Microsoft’s Copilot+ PC specifications require at least 16GB of RAM, 256GB of storage, and a compatible processor with a 40-TOPS NPU. A reported Microsoft roadmap links higher memory tiers to more demanding local AI work, but Microsoft has not confirmed the slide or its categories.
What problem it solves
Documents and office files often need to be transformed into a structured, text-friendly format before they can be processed by language models or text analysis tools. MarkItDown addresses this by preserving important document structures such as headings, lists, tables, and links during conversion to Markdown.
Key features
Broad format conversion — MarkItDown converts PDF, P…
What problem it solves
Editing text should not require extensive familiarity with terminal-based tools. The project aims to provide an accessible editor with a modern interface and input controls that are approachable for users who are largely unfamiliar with terminals.
Key features
Cross-platform installation — Installation options are documented for Windows, Linux, and macOS, including package-m…
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
Web automation and end-to-end testing often require coordinating browsers, interactions, assertions, and failure diagnostics. Playwright provides a unified API for driving Chromium, Firefox, and WebKit across tests, scripts, and AI-agent workflows.
Key features
End-to-end test runner — Playwright Test runs end-to-end tests across Chromium, Firefox, and WebKit with browser is…
What problem it solves
TypeScript adds optional types to JavaScript so tools can support large-scale JavaScript applications across browsers, hosts, and operating systems. It also compiles source into readable, standards-based JavaScript.
Key features
Optional types for JavaScript — TypeScript extends JavaScript with optional types, providing a typed language for application-scale development.
Sou…
What problem it solves
Go applications need a way to create, read, and modify spreadsheet documents produced by Microsoft Excel. They also need support for complex spreadsheet components and workflows involving large worksheet datasets.
Key features
Read and write spreadsheet files — Create and modify workbooks, set cell values, save files, and read cell values and worksheet rows through a Go API.…
What problem it solves
Knowledge workers need Claude to follow their organization’s tools, terminology, critical workflows, and preferred ways of working rather than provide only generic assistance. The repository provides customizable role-specific plugins that package this guidance with connectors, skills, commands, and sub-agents.
Key features
Role-specific plugin collection — The collection in…
What problem it solves
Products and agents need to communicate with users across multiple channels without rebuilding Inbox feeds, provider integrations, and channel webhooks. Novu addresses this with one API and a unified conversation model for product notifications and bidirectional agent conversations.
Key features
Unified multi-channel API — Send product notifications through Inbox/In-App, Pus…
What problem it solves
Applications need to run untrusted code while controlling its access to files, networks, user interfaces, and host capabilities. This requires an isolation approach that works across platforms, supports multiple containment mechanisms, and can be integrated without changing application architecture.
Key features
Cross-platform containment — MXC supports Windows, Linux, and m…
What problem it solves
Teams often have to move work into a separate AI tool, then repeatedly transfer context between that tool and the applications where collaboration happens. Switch addresses this by letting people and AI agents participate in the same conversations while keeping shared context with the work.
Key features
Shared rooms for people and agents — Each supported chat channel becomes…
What problem it solves
Speech recognition systems often lose global context when they divide continuous audio into short chunks, while speech generation must maintain speaker consistency, timing, and coherence across long recordings. The project addresses these needs with unified long-form ASR, streaming recognition, and text-to-speech models.
Key features
Long-form speech recognition — Process co…
What problem it solves
Company knowledge is distributed across business systems such as Slack, Google Drive, Jira, GitHub, and Microsoft 365. AI applications need a governed way to search this information, respect each user's access controls, and provide answers that can be traced to their sources.
Key features
Permission-aware enterprise search — Searches connected company systems while applying …
What problem it solves
Developers need support for their core edit-build-debug cycle. The project addresses this need with comprehensive code editing, navigation, and understanding support along with lightweight debugging, a rich extensibility model, and lightweight integration with existing tools.
Key features
Code editing and navigation — The editor provides comprehensive support for editing, na…
AI has already entered code writing, algorithm optimization, and model development. But after a task is completed, what remains may be reusable capability—or merely an answer and scattered logs. The difference becomes apparent only when a similar problem comes up again: can the system reuse what it learned, or must it pay the cost of exploration all over again?
Recursive Self-Improvement (RSI) co…