On mvanemmerik.com, Nov 16: NFL Week 10 chalk talk: Detroit dazzles, Cincinnati takes the North
vanemmerik.aiClaude, ChatGPT and Grok news, twice a day

Friday, November 13, 2026

OpenAI launches ChatGPT Wallet, letting ChatGPT Dots make purchases for US Pro subscribers

OpenAI's new Wallet lets dots check out with a saved card, approving each purchase with Face ID or a fingerprint. OpenAI also reportedly folded its robotics team into research and Applied. SpaceX, which now runs xAI, plans over $74 billion of AI campuses and reportedly cut about 60 Grok Imagine jobs. For Anthropic, Claude had a login outage, and its IPO is reportedly oversubscribed while banks turn away China and Hong Kong orders.

Friday, November 13, 2026

OpenAIFeature

OpenAI launches ChatGPT Wallet so ChatGPT Dots can buy things for you, starting with some US Pro subscribers

OpenAI introduced ChatGPT Wallet on Nov. 13, letting a dot choose items and check out with a saved US-issued Visa, Mastercard or American Express card. It starts with a limited number of US ChatGPT Pro subscribers aged 18 and up; each purchase is shown for review and approved with a check such as Face ID or a fingerprint, and the card is charged through a payment token so neither the dot nor the merchant sees the card number. OpenAI says spending limits and rules for pre-approved purchases are planned but not yet available.

Why it matters: It is the first time one of the always-on agents can pay for things itself, which raises the stakes on both usefulness and mistakes.

Try it: If you are a US Pro subscriber with access, add an eligible card to Wallet, then ask your dot to help with a purchase; you approve each checkout.

ChatGPT DotsChatGPTHappened Nov 13, 2026 · Sources: OpenAI Developer Community: Introducing ChatGPT Wallet · TechCrunch: ChatGPT's new Wallet lets AI agents shop and check out for you

OpenAIBusiness

OpenAI breaks up its stand-alone robotics team, moving researchers into its main research group and hardware staff into AppliedReported

Fortune reported on Nov. 13, and an OpenAI spokesperson confirmed, that the dedicated robotics unit has been dissolved: robotics researchers now report to chief research officer Mark Chen, while applied robotics and hardware staff move to the Applied organization under president Greg Brockman. OpenAI said robotics stays part of its long-term plans and that its general models, including GPT-6 Astra, can now be applied directly to robots. OpenAI has not published its own announcement.

Why it matters: It signals OpenAI is betting on its general-purpose models, not separate robot-specific ones, to drive its robotics work.

OpenAIHappened Nov 13, 2026 · Sources: Fortune: OpenAI disbands its stand-alone robotics team

GrokBusiness

SpaceX's AI group reportedly cuts about 60 jobs from the team building Grok ImagineReported

The Information reported on Nov. 13 that SpaceX's AI division, formerly xAI, laid off roughly 60 people working on Grok Imagine, its image and video generator, leaving a small team to maintain it. The report says management is prioritizing AI coding tools and the Grok Bot agent. SpaceX has not publicly commented, and Grok Imagine remains available.

Why it matters: It suggests Grok Imagine will get fewer major updates as the company shifts resources to coding and agents.

Grok ImaginexAIHappened Nov 13, 2026 · Sources: Yahoo Finance (Investing.com): Musk's AI group lays off 60 staffers from Grok video unit

ClaudeBusiness

Banks reportedly turn away mainland China and Hong Kong orders for Anthropic's IPO, citing US export rulesReported

Bloomberg reported on Nov. 13 that several banks on Anthropic's IPO have rejected orders from mainland China and Hong Kong investors or see allocations to them as unlikely, with some people involved pointing to US ITAR export rules. Anthropic declined to comment. Pricing is still expected on Nov. 18, with trading under the ticker ANT the next day.

Why it matters: It narrows the investor pool for one of the largest IPOs on record and shows how export controls now reach AI company shares.

AnthropicHappened Nov 13, 2026 · Sources: The Decoder: Banks bar Chinese and Hong Kong investors from Anthropic IPO

ClaudeSafety

Claude.ai, Claude Code and the Claude Console hit a login outage for about an hour on Nov. 13

Anthropic's status page flagged sign-in failures on Claude.ai and Claude Code at 09:58 UTC on Nov. 13, listing Claude.ai as a major outage, Claude Code and the Console as partial outages and Cowork as degraded, while the API kept working; a fix went out at 11:05 UTC and Anthropic moved to monitoring. No cause was given. Separately, since Nov. 12 a Microsoft 365 service incident has made Claude's mail, calendar and search tools fail for most requests and affected Claude for Outlook, which Anthropic's status page still listed as unresolved.

Why it matters: Login failures lock people out of the apps and coding tools even when the models themselves are up.

Try it: If Claude still won't sign you in, check status.claude.com before resetting anything on your side.

ClaudeClaude CodeClaude APIHappened Nov 13, 2026 · Sources: Claude Status · AI Weekly: Anthropic logs login outage on Claude.ai and Claude Code · Business Today: Claude AI is down

GrokBusiness

SpaceX plans more than $74 billion of AI supercomputing campuses at Starbase, Texas, and in Cochise County, Arizona

SpaceX said on Nov. 12 that it will spend more than $24 billion on a supercomputing campus at Starbase, Texas, which would make its own power from a gas-fired plant, and about $50 billion on one in Cochise County, Arizona, which would use dry cooling and local electric cooperatives. The Arizona site still needs county approval at a December hearing, and a petition with about 2,400 signatures opposes it; Starbase's city commission weighs a tax abatement on Dec. 9. SpaceX did not give power capacity or completion dates.

Why it matters: It is the biggest compute commitment yet from the company that now trains Grok, and both sites still face local approvals.

xAIHappened Nov 12, 2026 · Sources: The Star (Reuters): SpaceX to invest over $74 billion in Texas and Arizona supercompute facilities

ClaudeBusiness

Update: Anthropic's IPO is reportedly drawing several times more orders than the 825 million shares on offerReported

Bloomberg reported, citing people familiar with the matter, that orders for Anthropic's IPO are several times the shares available, mostly from large institutions; one report puts demand at about six times. Pricing is expected on Nov. 18 with Nasdaq trading under ANT from Nov. 19, and the filed range is still $70 to $78 a share. Anthropic has not commented on demand, and no new range has been filed.

Why it matters: Strong demand makes a price above the filed range more likely for what could be the largest IPO on record.

AnthropicHappened Nov 12, 2026 · Sources: Seoul Economic Daily: Anthropic IPO demand swamps order book · Dataconomy: Anthropic IPO draws orders six times shares available

OpenAIPolicy

OpenAI reportedly tells Congress it backs the Chip Security Act, AI Overwatch Act and MATCH Act on AI chip exportsReported

OpenAI has told congressional offices it supports three bipartisan bills on advanced chips, according to reports from Bloomberg and Politico on Nov. 12. The Chip Security Act would require location checks on export-controlled AI chips, the AI Overwatch Act would give Congress more review of chip export rule changes, and the MATCH Act would press allies to match US limits on chipmaking tools. OpenAI has not published a statement on the bills.

Why it matters: It puts OpenAI publicly in the debate over how tightly the US limits AI chip sales abroad.

OpenAIHappened Nov 12, 2026 · Sources: Plain English: OpenAI backs Chip Security Act, AI Overwatch and MATCH Act

OpenAIFeature

ChatGPT Dots can now connect to several of your computers and run each task on the one you pick

OpenAI said on Nov. 9 that one dot can now be linked to multiple local computers, so it can, for example, build and test an app on a spare Mac while you keep working on your laptop. Each computer is approved separately in the ChatGPT app, has to stay online with the app open, and can be cut off with Revoke access; each thread uses one computer at a time. Dots still run on OpenAI's cloud computers unless you connect your own.

Why it matters: It lets OpenAI's always-on agent spread work across machines instead of taking over the one you are using.

Try it: In the ChatGPT app on each computer, open your dot's profile, find the computer under Computers and choose Allow, then Allow access.

ChatGPT DotsChatGPTHappened Nov 9, 2026 · Sources: OpenAI Developer Community: Connect multiple local computers to the same dot · ChatGPT Learn: Connect computers and apps to your dot · TokenPost: OpenAI lets users choose which computer runs ChatGPT Dot tasks