xAI

Elon Musk posted that xAI's Grok Bot will use the 'best back end model for any given task,' naming Anthropic's Claude Opus 5.5 (as relayed), Midjourney and Suno — 'Whatever is most likely to give you the best outcome,' per City AM. The post describes a direction, not a documented launch or partnership; neither xAI nor Anthropic has published commercial terms, and the post did not specify routing, pricing, user controls or data handling.

SpaceX is in early talks to borrow about $40 billion for Nvidia AI chips, the Financial Times and Bloomberg report. The FT, via Reuters, puts the structure at roughly $10B bank loans plus $30B investment-grade debt, with Apollo expected to lead and Pimco among lenders, targeting a 2027 close. Bloomberg independently confirms talks and Pimco interest, and stresses the discussions are early-stage and may not happen. The borrower is SpaceX — not xAI. No party has confirmed a deal.

On Oct. 5, OpenAI, Anthropic, Google and Meta testified under oath to New York’s City Council on a package of proposed AI bills. None put a number on catastrophic risk or promised that failing an internal or third-party safety test would block a release. xAI did not appear despite a Council subpoena; Speaker Julie Menin said the Council is pursuing it in court. The bills remain proposals — nothing has passed.

After a White House AI luncheon, Trump posted a Joint Commitment on Frontier Responsibilities with Anthropic, OpenAI, Google, Meta, Nvidia, and xAI. The text pledges four layers of self-policing — including an independent external auditor and a board committee — with no enforcement, company-chosen auditors, and no duty to publish findings. It is Washington’s political product after the Astra shelf and rogue-agent week, not a statute and not the earlier UK-tester hold.

xAI shipped Grok 4.7 at the same $2/$6 token prices as Grok 4.6, with a fast variant at 2× speed and 2× price. Lab tables lead EEBench and Harvey’s legal-agent bench; Andon’s Vending-Bench put it in the middle — closer to Opus 5.5 than to Sol.

xAI shipped Grok Voice Transcribe 2.0 at unchanged $0.10 batch / $0.20 streaming per hour. Lab short-phrase WER falls from 20.6% to 6.8%; xAI cites first place on Artificial Analysis’s streaming accuracy board among 32 models.

Anthropic CEO Dario Amodei published “We Must Pace the Frontier,” committing Anthropic unilaterally to METR-class evaluators with employee-like access. Sam Altman said OpenAI will do the same; Elon Musk said Dario is right. Pacing is not a halt.

xAI made Grok Bot available for enterprises with access, network, and audit controls — two weeks of free usage for Grok and Cursor Enterprise customers, including people without an existing seat. Bot is no longer a prosumer teammate.

xAI shipped the first Grok Bot integration with X: connect an X account, auto-provision a developer account if needed, and let paid Bot users spend complimentary X API credits to search posts, read timelines, check mentions, and pull trends. Lab calls it the first version of this integration.

OpenAI will stop serving Cursor after SpaceX’s acquisition of Anysphere closes, with a proposed November 12 cutoff. Cursor CEO Michael Truell said OpenAI is about 5% of Cursor traffic; Anthropic’s Tom Brown said Claude compute will fill the gap. A 5% cut is survivable; the precedent is not.

xAI put Grok 4.6 on Microsoft Foundry in public preview at the same $2 / $0.50 cached / $6 per million tokens as Bedrock and the first-party API — two weeks after the model GA and a week after Amazon and Google cloud ports. Watch: Foundry Learn docs list 200k context against xAI’s 500k card.

Nvidia said xAI will deploy Vera CPUs — the 88-core Olympus CPU Nvidia bills as built for agents — to orchestrate Grok’s next agentic workloads, scale Grok infrastructure on Vera Rubin toward gigawatts, and base its first-generation Starmind AI satellite on an optimized Vera Rubin NVL72 rack. No matching x.ai/news post — Nvidia primary is the source. Starmind is planned, not a launch date.

xAI expanded Grok Bot from the August 11 Heavy/Ultra/Teams Premium beta to SuperGrok Plus, Cursor Pro+, Cursor Ultra, and all Cursor Teams plans. Enterprise remains on a waitlist. Same product line as the beta, new seat map — how xAI tries to make persistent teammates the default coding-plan extra, not a Heavy novelty.

xAI made Grok 4.6 available on Google Enterprise Agent Platform via Model Garden at $2 / $0.50 cached / $6 per million tokens — the same list as Bedrock two days earlier — with a 500k context window and low through xhigh reasoning effort. Google-cloud distribution, not a dual-date of the August 12 model GA.

xAI moved Grok Build out of SuperGrok Heavy early beta: describe an app, game, site, or dashboard in Grok and it builds live in chat on web, iOS, and Android, on every plan. Publish to grok.me, remix, custom domains, GitHub export, and in-app xAI APIs ship the same day.

xAI put Grok 4.6 on Amazon Bedrock at the same $2 / $0.50 cached / $6 per million tokens as the first-party API, with a 500k context window and low/medium/high/xhigh reasoning effort. Same pattern as OpenAI Daybreak on Bedrock (August 11) — not a second model GA.

Guidelight AI Standards graded Anthropic, OpenAI, Google, xAI, and Meta on six rogue-model control practices. No lab scored above 3/5 on any practice. Anthropic and OpenAI tied at C+ (2.50); Google D+; xAI D−; Meta F. Anthropic got a zero on containment despite the safety brand.

xAI said Grok 4.6 is live in GitHub Copilot across VS Code, cloud agents, and the Copilot CLI, with some businesses needing to enable the model in Copilot settings. API pricing remains $2/$6 per million tokens. Distribution ship — not a second model GA.

xAI released Grok 4.6 — a post-training upgrade of Grok 4.5 aimed at long-running agents, coding, and interactive visual work — that ties GPT-5.6 Sol Max at 61 on the Artificial Analysis Intelligence Index, ships at $2/$6 per million tokens, and is live in Cursor, Grok Build, and the xAI API. DeepSWE and Terminal-Bench still trail Sol/Fable.

xAI opened Grok Bot early beta: always-on agents on their own cloud computer that sign into the same apps a coworker would, gated to SuperGrok Heavy, Cursor Ultra, and Cursor Teams Premium on desktop and iOS. Distinct from Grok 4.6 the next day — hosted teammate surface, not a model ship.

xAI upgraded Imagine Video 1.5 with text, image, and voice references (up to seven image refs), native 1080p, and prompt-only text-to-video—rolling first to US SuperGrok Heavy and Plus. The trade is production-ready character consistency, not another one-shot clip model.

xAI’s Grok Voice Think Fast 2.0 cites Artificial Analysis speech-to-speech quality at 82.9% (ahead of GPT-Realtime-2.1 High at 79.1%), 0.70s time-to-first-audio, and $0.08/min. grok-voice-latest migrates to 2.0 on August 5 unless pinned—force a pin strategy before the default flips.

xAI launched Build Mode in early beta for SuperGrok Heavy—describe an idea, get a live preview in chat, publish to grok.me or a custom domain on web, iOS, and Android. End-to-end creation gated to the top paid tier; competes with ChatGPT Sites-style hosted builders.

xAI put Grok 4.5 in GitHub Copilot across VS Code, Copilot CLI, and cloud agents—selectable from the model picker (enterprise enablement may be required). API at $2/$6 per MTok. Copilot availability changes where developers invoke the model, not Arena or AA ranks by itself.