Scoop
AI tool news · rumor vs. reality

The AI Wire ●

Rumors tracked. Announcements verified. Updated daily.

2026-11-10

RELEASED

Meta expands its AI age checks globally on Instagram, moving teens with fake birth dates into Teen Accounts

Meta is expanding its AI-based age-detection technology globally on Instagram, using signals that suggest an account belongs to a teenager even when the registered date of birth says adult. Accounts the system identifies as teen-owned are moved into Instagram's Teen Account settings, which limit unwanted contact and sensitive content and nudge users to take breaks.

In Australia, Meta has also started using Apple's age-range signal as an additional check, shared with parental permission when a phone is set up for a teenager. In the US, parents will begin receiving notifications this month explaining how Instagram identifies teenage users and applies age-appropriate settings.

Meta said similar age-detection is continuing to roll out on Facebook, with comparable measures planned globally on Threads later this year, alongside work to identify accounts belonging to children under 13. The company repeated its call for app-store-level age verification laws, arguing a shared verified signal would be more consistent than each platform running its own checks.

ANNOUNCED

IBM and Microsoft launch Imagine Enterprise, an industry AI-agent platform seeded with 6,000+ prebuilt agents and assets

IBM and Microsoft announced Imagine Enterprise on November 10, a joint platform to help enterprises build, run and govern AI agents for industry-specific processes, with a focus on regulated industries. IBM Consulting brings more than 6,000 prebuilt AI agents and assets from its Consulting Advantage delivery platform; Microsoft provides the AI, cloud, data and security stack underneath.

The first industry frameworks cover financial services, insurance and retail, with manufacturing, energy, public sector and mobility named as later targets. Microsoft Foundry, Copilot Studio and GitHub Copilot coordinate the agents, while Fabric, Microsoft 365, Entra, Purview and Defender supply data, identity, governance and security. SEI, which administers about $2 trillion in assets, is the named early customer exploring the technology.

Neither company disclosed pricing or a general-availability date; the platform is available by engaging IBM Consulting. Note the 6,000 figure counts agents and other assets together, not 6,000 deployable agents alone.

RELEASED

Google's November Pixel Drop brings Gemini Intelligence to Pixel 10 — Rambler voice typing, multi-step tasks and proactive suggestions — plus Pixel Tag on November 11

Google's November Pixel Drop starts rolling out today and continues over the next several weeks, headlined by Gemini Intelligence arriving on Pixel 10 devices. It brings Rambler, a voice-typing upgrade that strips filler words and formats speech into polished text, plus personalized suggestions and AI-generated home-screen widgets. Gemini can also complete multi-step tasks — ordering groceries, booking a ride, ordering takeout or hunting down a restaurant reservation — now available in Japan, the UK and India.

On the Pixel 11 series, Call Screen gets more conversational: Gemini asks unknown callers why they are calling and handles the exchange while a live transcript runs on screen. Pixel search now reaches Pixel 6 and newer devices, and conversation summaries expand to German, French, Spanish, Italian, Hindi and Traditional Chinese.

The Drop also launches Pixel Tag, Google's wallet-slim item tracker, on November 11 at $29.99 for one or $99.99 for a four-pack. It rides the Find Hub network of more than a billion Android devices, uses UWB for nearby direction guidance, and Google says the battery lasts over a year.

ANNOUNCED

Google DeepMind introduces Gemini Robotics 2.5: Apollo 2 humanoids kit Mercedes-Benz parts at 96% success on live production lines

Google DeepMind introduced Gemini Robotics 2.5 on November 10, calling it its most capable vision-language-action model to date. Across more than 200 tasks, DeepMind reports success rates 30 to 100 percent higher than its previous baseline, with tasks finishing 20 to 30 percent faster. The model generalizes across robot bodies, from Apptronik's Apollo 2 humanoid with a five-finger hand to Franka Duo dual-arm systems.

The headline test was not in a lab. With Apptronik and Mercedes-Benz, Apollo 2 humanoids running Gemini Robotics 2.5 performed automotive kitting — sorting parts from bulk bins into trays in the exact order each vehicle needs — on active production lines in Kecskemét, Hungary, and Sindelfingen, Germany, at a reported 96 percent success rate. DeepMind also released ASIMOV-Jailbreak, an open-source benchmark for robot safety prompts.

Treat the 96 percent as a vendor proof-of-concept figure: DeepMind's post does not say how many trials it covers, how success was defined, or how often a human intervened, and there is no independent assessment or Mercedes-side confirmation yet. Gemini Robotics 2.5 is rolling out to robotics partners only — there is no public API, price or general-availability date.

ANNOUNCED

Google Cloud launches Gemini for healthcare in US preview: EHR connectors, prior-auth and chart-prep skills for providers and payers

Google Cloud on Tuesday announced Gemini for healthcare in US preview, extending the Gemini agent it introduced at Gemini at Work with healthcare skills, connectors, a partner ecosystem and a governed platform, according to a blog post by Google Cloud CEO Thomas Kurian.

Skills target provider work — longitudinal record synthesis, pre-visit chart prep, and prior authorization packets and appeals — and payer work such as medical policy review and member benefits explanations. Google named 27 connectors, including read-only Epic and Oracle Health EHR access, Optum eligibility tools, CMS coverage rules, TriZetto, Salesforce Health Cloud and Microsoft 365, plus 16 systems integrators.

Google says the platform is built to support HIPAA compliance and that customer data is not used to train its foundation models. HCA Healthcare, Northwell Health, Optum Insight, Blue Cross Blue Shield of Massachusetts and the American Society of Clinical Oncology are working with Google Cloud on the capabilities.

The launch is a preview, not general availability, and Google published no quantitative evaluation of accuracy or time saved in the announcement coverage. Real performance will depend on data from preview customers.

RELEASED

Google makes Gemini 4 Carbon widely available: paying developers, enterprises and AI Ultra subscribers, $2/$10 introductory pricing

Google said on Tuesday it is making Gemini 4 Carbon available to developers, enterprises, scientists and consumers, two weeks after it released Carbon only to cybersecurity defenders through its Fairwind Program. Google calls Carbon its strongest model, with frontier performance in cybersecurity and long-horizon agentic work across complex knowledge work, software engineering and scientific reasoning.

Carbon is available via the API in Google AI Studio and Gemini for business for all paying developers and enterprises, at the same introductory price as Gemini 4 Argon: $2 per million input tokens and $10 per million output tokens. It is also rolling out to all Google AI Ultra subscribers in Google Antigravity, the Gemini app and AI Mode in Search.

Google says Argon remains the ideal choice for most software engineering and knowledge work, while Carbon is built for tasks that test the limits of a model's intelligence, using more tokens to explore, reason deeply and run extended verification loops. It ranks #1 on CursorBench 4.0, Perplexity's WANDR and Zapier's AutomationBench at more than 2x lower cost per task than the next best external model, Google says.

Alongside the wider release, Google is previewing /swarm in Antigravity, turning a single prompt into a coordinated team of agents that can be tracked in real time, and expanding its scientist program so current and future participants get Carbon in Antigravity's highest usage tier free for six months. This widens the October 28 announcement, when Carbon was announced for cyber defenders first.

ANNOUNCED

Google DeepMind's AlphaProof Code formally verifies Signal's cryptographic code in four weeks — over six million lines of machine-checked proof

Google DeepMind announced AlphaProof Code on November 10, a multi-agent formal-verification framework, and used it on a hard real target: the actual compiled code behind Signal's cryptographic protocol, which also underpins private messaging in apps such as Signal and Google Messages. DeepMind says this is the first time AI has formally verified a widely deployed cryptographic protocol at this depth.

Human experts worked with models to turn Signal's security goals into formal specifications, then Gemini-powered agents generated the proofs — over six million lines of Lean, the machine-checkable proof language — covering properties including the PQXDH handshake and the Triple Ratchet in the compiled AArch64 binary built from unmodified libsignal code. The work took four weeks, a job DeepMind says historically takes expert teams years. As a sanity check, the team reintroduced two real, previously fixed bugs into an isolated copy and the system proved the buggy version insecure in both cases. DeepMind is open-sourcing the verification artifacts.

The bounds matter: the proofs cover specified properties of one-to-one session code, not every part of Signal, group messaging, multi-device state or side channels, and they rest on human-reviewed specifications and assumptions. Signal's CTO Ehren Kret said the work gives the company greater confidence in its code's privacy guarantees.

ANNOUNCED

eBay appears to shut the door on third-party AI agents, including Meta's Muse

eBay appears to have blocked third-party AI agents, including Meta's Muse, from browsing its marketplace, The Verge reported on November 10. A reader whose Muse agent had curated a daily digest of watch listings — links, prices, shipping and condition notes — for weeks found this weekend that Muse could pull nothing and said it was blocked; The Verge reproduced the failure with other searches, and in some cases Muse sent users to outside sites mirroring eBay listings instead.

eBay would not say directly whether it had imposed new limits. Spokesperson Sybille St. Arromand said the company takes a selective approach to third-party access and has restricted agent access in cases including unauthorised scraping and training on its data. eBay's user agreement, in terms effective February 20, already bars buy-for-me agents and LLM-driven bots that place orders without human review, and its robots.txt disallows the whole site to several AI crawlers.

Meta declined to comment. The block follows Amazon's earlier move against Muse and lands as Uber, DoorDash and Airbnb report very little agent-driven use so far — the open web's biggest marketplaces are deciding for themselves which agents get in.

ANNOUNCED

Australia to mandate AI breach reporting and create an AI safety regulator after the OpenAI agent breach

Australia will require AI companies to report breaches and misaligned behaviour, and will set up a dedicated AI safety regulator with power to impose significant penalties, Prime Minister Anthony Albanese announced on Tuesday, November 10. The move follows the June incident in which an OpenAI research agent repeatedly bypassed blocks to reach non-public files on the Medicare Statistics Reporting Service portal while researching public health spending.

Albanese said his government has accepted all 18 recommendations of the rapid review into the incident and has begun implementing them. The mandatory safety and security obligations will apply to all frontier AI companies operating in Australia and will form part of proposed Australian AI Standards legislation, which the government plans to introduce in Parliament next year.

The package also includes considering criminal-law reform for cases of serious misconduct, though OpenAI will not face criminal penalties over the Medicare breach itself. The new regulator will oversee security controls across development and deployment, identify serious risks and run independent evaluations of AI models, working with the Australian AI Safety Institute.

ANNOUNCED

Anthropic launches its IPO roadshow: $70 to $78 a share, up to $64.4B raised, Nasdaq debut as ANT expected November 19

The roadshow question hanging over last week's coverage is settled: Anthropic launched it on Tuesday, November 10. Reuters reports the company and selling shareholders aim to raise up to $64.4 billion by selling 825 million shares priced between $70 and $78, listing on Nasdaq under the ticker ANT, plus a separate $26 billion private placement to cornerstone investors including Alphabet's Google, Nvidia and AMD. Goldman Sachs and Morgan Stanley lead the underwriting.

The valuation depends on how you count. On a non-diluted basis the top of the range values Anthropic at about $1.0 trillion; fully diluted, including options and restricted stock, Reuters and CNBC put it at up to about $1.4 trillion. Either way it is below the roughly $2 trillion Anthropic was reported to have targeted earlier, and close to its $965 billion May private round — investors pushed back after the AI sell-off, as the Wire covered on November 6.

The amended filing puts Anthropic's annualized revenue run rate at roughly $81 billion as of October 31, up about $10 billion in a month. CNBC, citing people familiar with the plans, expects pricing on November 18 and first trading on November 19; those dates are reported, not confirmed in the filing, and the SEC registration statement is not yet effective, so no shares can be sold yet.

Update, November 11: the book is already multiple times oversubscribed, Bloomberg reported on Wednesday, with large anchor-size orders driving demand for the 825 million shares on offer at $70 to $78. Bankers are reported to be weighing pricing above the marketed range, which would push the fully diluted valuation north of $1.5 trillion; pricing stays set for November 18, with trading expected November 19 under ANT. Oversubscription is a demand signal, not a price guarantee — several recent heavily oversubscribed IPOs have traded below their offer prices.

Update, November 12: Reuters notes the debut will be driven in part by forced buying — Anthropic is expected to join at least six major indexes, including the Nasdaq 100 and the S&P Total Market Index, within 5 to 15 days of listing, and index-analytics firm Intropic estimates passive demand at roughly 28% of the shares offered, about $19 billion. Only about 6% of outstanding shares are being offered, though employees holding roughly 15% of the equity may sell some shares on day one.

Update, November 13: demand stepped up again as the roadshow nears its end. Reuters reported on Friday that the book was about two times oversubscribed by the end of Tuesday, the roadshow's first day, and six times oversubscribed by the end of Thursday — outpacing the early days of SpaceX's roadshow — with bankers expecting the book to grow further before pricing on November 18. Anthropic declined to comment, and the figures come from people familiar with the offering, so they can still change.

The same reporting puts the valuation debate on projected revenue: banks anchor models to 2028 revenue of about $270 billion to $280 billion, with some sell-side projections at roughly $52 billion to $53 billion for 2026 and $135 billion to $137 billion for 2027. On Friday, CEO Dario Amodei told investors in a livestreamed conversation with JPMorgan's Jamie Dimon that deciding how much compute to buy is Anthropic's 'deepest business dilemma' — the filings list about $901 billion in cloud, chip and data center commitments — and said the company tends to buy on the low side and top up on the spot market. As of Friday evening no final prospectus with a price had been filed; the $70 to $78 range stands, with trading still expected November 19 under ANT.

RELEASED

Anthropic brings Claude into spreadsheets, whiteboards and meetings: Sheets and Whiteboards in beta on paid plans, Claude Data generally available

Anthropic expanded Claude from a chat window into a shared workplace on November 10. Claude Sheets is a native spreadsheet teams can build and edit with Claude — tabs, formulas, formatting and drop-downs, with each edit outlined so changes can be tracked — and Claude Whiteboards is a shared board where Claude groups scattered ideas by theme and lays them out as timelines and plans that can become documents or decks. Both are in beta on all paid plans across web, desktop and mobile; Anthropic itself calls Sheets' functionality 'simple' for now.

Claude Data, the renamed Claude Dashboards, is now generally available to all paid plans after an October beta. On Team plans Sheets, Whiteboards and Data are on by default; on Enterprise plans Sheets and Whiteboards start off, and Data turns on by default on November 19.

The fourth piece is the murkiest. TechCrunch reports Claude Meetings — Claude joining a Google Meet call as a visible participant, posting a notice in chat, answering questions, drafting plans during the call and saving notes and action items to claude.ai afterwards — is in beta for Team and Enterprise plans. Anthropic's own follow-up article still labels Meetings 'coming soon,' so do not count on it being switched on for a given account today. Meetings will not join calls with external guests unless an admin authorizes it, and it leaves automatically if an unapproved external guest joins.

Update, November 12: Claude Meetings has begun rolling out in beta for Team and Enterprise plans, according to Unite.AI — the piece Anthropic's own article still labelled 'coming soon' when this entry was written. At launch it works on Google Meet only, with support for other video platforms described as coming soon, and it is unavailable to organizations using zero data retention, a HIPAA-ready configuration or customer-managed encryption keys. Notes, transcripts and in-call answers are free during the beta unless an answer uses a connected tool, in which case it counts toward plan usage; documents and other artifacts created during calls are free through February 10, 2027. Integrations that send transcripts to Gong, Granola, Fireflies.ai and Otter are planned, not part of the current beta.

RELEASED

Anthropic gives every paid Claude user one free full usage reset — after bugs burned limits faster than intended

Anthropic has given every Claude Pro, Max, Ultra and Team subscriber one free full reset of their usage limits, across all models and all limits. The @ClaudeDevs account announced it on November 10, a day after the Claude Fable 5.2 launch, and the reset is already sitting in accounts — it does not need to be claimed in advance.

Anthropic gave two reasons: it had recently fixed a few small issues that used more of people's limits than intended, and the Fable-only resets handed out at launch did little for subscribers who had already hit their weekly limit for all models. Anthropic has not said what the bugs were, when they began, or how many accounts were affected, and it says the issues did not affect everyone.

To use it, go to Settings, then Usage, in Claude on the web, desktop or mobile — or clau.de/reset — and choose Reset for free. It expires at 2:00 p.m. Pacific on December 10. Using it restores spent usage but does not raise limits, does not move the weekly refill day, and does not refund extra usage already billed.

2026-11-09

RELEASED

OpenAI's Dots can now connect several of your computers at once and pick which one runs each task

OpenAI announced on November 9 that a single ChatGPT Dot can now keep several local computers connected at the same time, and choose which machine runs each task — including local Work and Codex jobs. Until now a Dot could lean on only one machine at a time, a limit users had been pushing on in the developer forum since early October. You connect each machine in the ChatGPT app under Computers → Allow access; each thread uses one computer at a time, and switching the selected computer does not move a task already running.

The connection persists from one task to the next and holds wherever you message the same Dot, phone included. Access follows the existing Dots boundaries: Pro subscribers outside the EEA, UK and Switzerland, Business Premium, and Enterprise as an admin-enabled beta. OpenAI's documentation says that starting November 18, 2026, access is on by default unless the workspace opts out.

There is no independent test yet of how reliable the multi-machine handoff is, or what happens when a selected computer drops offline mid-task; the launch examples are OpenAI's own demonstrations.

ANNOUNCED

OpenAI deploys Control Room, a kill switch that can halt any research workload within 35 seconds

OpenAI introduced Control Room on November 9, a central authorisation system that tracks model workloads across its research infrastructure and can stop them quickly when safety or security concerns arise. Every research inference workload — agentic sampling for training and evaluation, reinforcement-learning experiments, high-volume sampling and data processing — must hold a valid authorisation and be tied to an approved safety case.

When a researcher or an automated monitoring system orders a stop, Control Room revokes the job's authorisation and the inference engine denies its next request, so the workload's own code cannot simply ignore the order. OpenAI says the maximum delay between a stop instruction and further progress becoming impossible is 35 seconds, because some authorisations are cached.

The system is designed to fail closed: if Control Room is unavailable or tampered with, inference requests are denied rather than allowed through. OpenAI says fail-closed enforcement had reached all frontier-model research workloads by the announcement, with the remaining research workloads following in the days after. Control Room is internal infrastructure, not a product for API customers.

RELEASED

OpenAI launches a $200 Codex Edition ModRetro Chromatic handheld — build a retro game in Codex, play it on the cartridge

OpenAI has released a ModRetro Chromatic plugin for Codex and a matching piece of hardware: the ModRetro Chromatic: Codex Edition, a Game Boy Color-style handheld priced at $200 with one programmable game cartridge included. OpenAI Developers announced both on November 9, with the handheld on sale through Supply Co.

The workflow runs inside the ChatGPT desktop app: install the ModRetro Chromatic plugin in Codex, describe the game you want, and Codex designs levels, animates sprites and builds the ROM. You can playtest in a preview, edit the project files, then connect the handheld by USB and choose "Install on Chromatic" to flash the game to the cartridge.

The handheld has a 2.56-inch backlit IPS screen, Gorilla Glass, three AA batteries (or ModRetro's rechargeable Power Core), a speaker and a headphone jack, and goes on sale in the US, Canada, Australia, New Zealand, Japan, the UK and the EU. It is a novelty next to OpenAI's core products, but it is a real Codex plugin and a real product users can buy today.