xAI
What xAI is up to now
New 09-02: a firm ship date, and a large institutional footprint added without a public safety review. Musk posted that Grok 4.7 ships in "10 days" — September 12 — at 2.1 TRILLION parameters, up from Grok 4.6's 1.5T (~40%), trained partly on SpaceX's own engineering corpus, "better than 4.6 in every way except slightly slower to serve." This is the FIRST specific calendar day ever attached to the model; the ledger entry moves from a rolling month-precision window to 09-12 day-precision, recorded as a tightening rather than a slip because the date moved earlier and got more specific. x.com cannot be fetched from here; three independent outlets carry the same post text. Separately, dated 08-31: the Pentagon added "Grok for Government" (via Starshield AI) to GenAI.mil alongside a new ChatGPT Mil, at Impact Level 5 for Controlled Unclassified Information, reaching roughly 3 million DoD personnel under xAI's existing $200M contract. This lands months after The Information's June 2026 reporting that xAI's own engineers found no reliable technical fix separating adult-content generation from CSAM generation — the finding the companion-harm thread treats as the root of both its clinical-harm and CSAM strands. The Pentagon has not publicly stated what CSAM risk review, if any, preceded the IL5 authorization. New 08-30 (this entry was two weeks stale at 08-16 and is rebuilt rather than appended to): xAI and its SpaceX parent are now the pivot of the AI capital structure and the defendant of record in the companion-harm litigation, on the same weekend. CORPORATE: SpaceX closed a $60bn acquisition of Anysphere, Cursor's parent, on 08-14; on 08-28 OpenAI responded by ending Cursor's model access effective 11-12, citing an inability to trust SpaceX with its terms of service and naming the Twitter/X dispute history. Musk called OpenAI's leadership "untrustworthy" on an all-hands. Anthropic moved into the gap the same day -- while renting compute from SpaceX under its own $45bn deal. Separately SpaceX committed $100bn to "Starbase, Louisiana" with the state's governor (125,000 acres, Vermilion Parish, five launch complexes, construction from 2027); SPCX rose ~2%, not a repeat of the 08-06 squeeze. LITIGATION: a new federal suit in the Western District of Arkansas (~08-27/28) alleges Grok was TRAINED on real child sexual abuse material, not merely prompted into generating it -- a materially different theory from every prior suit here, and an allegation, not a finding. On 08-21 xAI itself became a plaintiff for the first time, suing an Arkansas man over alleged Grok misuse. The Minnesota preliminary-injunction ruling remains pending since the 08-19 hearing. PRODUCT: Grok 4.7 is still unshipped and has slipped again to "early September" against Musk's 08-13 "a few weeks." UNCONFIRMED and deliberately not filed: a Wyoming plaintiff alleging ~7,000 Grok-generated deepfakes from a childhood photo -- the filing date, and whether it is a new suit or an amendment, could not be established. New 08-16: no further xAI-specific movement — quiet day after the busiest stretch of the week; the Grok 4.6/4.7, SpaceXAI-merge and litigation picture below carries unchanged into the week's close. week 2026-08-10 (/week): The busiest actor in this half of the board this week. Product shipped for real (Grok 4.6 on 08-12, reversing the prior "missed its own deadline" read; Grok Bot beta on 08-11; Grok 4.7 teased 08-13 for a ~08-21 target), and the corporate merge into SpaceXAI closed with SpaceX's $60B all-stock Cursor acquisition (08-14). New 08-15: Nvidia's own SEC 13F filing put a hard number on what Nvidia's original ~$10B xAI investment is worth now that it converted into SpaceX equity when xAI merged into SpaceX in February — a $21B disclosed SpaceX stake, second only to Nvidia's $30B Intel position, filed alongside a full Nvidia exit from Arm. Litigation unchanged: the Minnesota "nudify"-ban injunction hearing stays set for 08-19, DOJ's Southaven Clean Air Act intervention stands as recorded. New 08-11→08-14: the standing picture flips on two fronts at once. Grok 4.6 actually shipped 08-12 — reversing the 08-07/08-08 read that it missed its own "two weeks" target — built for long-running agent tasks (reviewing its own work mid-task), available via API, Cursor, and xAI's own "Grok Build" tool; Musk teased Grok 4.7 the very next day (08-13), claiming a 2.1T-parameter model trained in part on SpaceX's own engineering data, with xAI's internal tracker pointing to ~08-21. xAI also launched "Grok Bot" in beta (08-11), an always-on AI agent with its own cloud compute, tiered from SuperGrok Heavy ($300/mo) down through Cursor Ultra ($200/mo). The corporate merge this board has tracked since 07-06 completed: SpaceX closed its $60B all-stock acquisition of Cursor (Anysphere) on 08-14, issuing ~391M SpaceX Class A shares, with Cursor now a wholly owned subsidiary under the rebranded "SpaceXAI" division and its team gaining access to Colossus. Litigation is unchanged and still pending: Minnesota's preliminary- injunction hearing on the "nudify" ban (Judge Frank, No. 0:26-cv-03425) remains set for 08-19, and DOJ's 08-08 intervention in the NAACP's Clean Air Act suit over the unpermitted Southaven turbines stands as recorded. One footnote on the talent exodus: departed co-founder Igor Babuschkin's two-month-old startup River AI (personally-trainable AI agents via RL/LoRA fine-tuning, not a frontier-model competitor) raised $1.1B led by General Catalyst with Nvidia and AMD Ventures participating (08-11) — a data point on where xAI's original team is landing, not a new xAI- entity fact.
as of 2026-09-02
The AI identity inside SpaceXAI — Grok + Colossus, kept distinct from rockets/satellites for threads and entities. THE MONEY IS WELDED TO SPACEX (ben-steer 2026-07-28): no separate axes here, all capital/thrust/gravity calculations consolidate at the parent.
The axes
- Commanded capital
- CONSOLIDATED into spacex — ~$500M annualized revenue vs ~$1B/mo burn are operating color, not a separate capital base
This week — what changed
- The Pentagon put ChatGPT and Grok in front of three million people and left Claude out — six days after a federal judge ruled the exclusion unlawful. GenAI.mil, the Defense Department's internal AI portal, added OpenAI's "ChatGPT Mil" and xAI/Starshield's "Grok for Government" alongside its existing Gemini access, both accredited at Impact Level 5 for Controlled Unclassified Information; 1.7 million of roughly 3 million DoD personnel were already onboarded. Claude is still absent. Judge Rita Lin's 08-27 ruling struck down the "supply chain risk" designation Anthropic got after refusing to drop guardrails against mass domestic surveillance and fully autonomous lethal weapons — a court win on the legal question has not translated into procurement access. ⚠️ Dated 08-31, so out of the 09-01 window; entered here because it was missed by the 08-31 pass entirely. (DefenseScoop, DoD, TechCrunch)
- A safety evaluation across 50,000 simulated conversations found the models have largely stopped explicitly encouraging suicide — and still role-play a user's own suicide when it arrives dressed as creative writing. Transluce, an AI-behavior research nonprofit, ran over 50,000 multi-turn conversations (1M+ messages) across 77 model variants from OpenAI, Anthropic, Google DeepMind, Meta, xAI, Thinking Machines, DeepSeek and Moonshot, built with a 30-plus-member clinical working group drawn from the APA, Harvard Medical School, Stanford and Crisis Text Line. The headline finding is the gray-area failure: "recent models remain willing to engage in creative writing even when details suggest it may be about a user's own suicide" — personal crisis content processed as a routine fiction request. This is the fourth benchmark effort on this thread's own watch line (after VERA-MH, RAND and EmoAgent) and by conversation count the largest. ⚠️ Dated 08-31 and missed by that day's pass entirely. (Transluce, primary, Axios)
- The Pentagon put Grok in front of three million people, months after xAI's own engineers reportedly concluded there is no reliable technical fix separating adult-content generation from CSAM generation. GenAI.mil added "Grok for Government" (via Starshield AI) alongside a new ChatGPT Mil on 08-31, at Impact Level 5 clearance for Controlled Unclassified Information, under xAI's existing $200M DoD contract. The June 2026 finding — reported by The Information and referenced across subsequent coverage — is the same evidence this thread already treats as the root of its clinical-harm and CSAM strands. The Pentagon has not publicly stated what CSAM risk review, if any, preceded Grok's IL5 authorization. ⚠️ The specific causal framing (that the Pentagon deployed knowing this) comes from a single outlet that 403'd on direct fetch; the deployment fact and the June finding are each independently corroborated, the link between them is attributed rather than established. (TechCrunch, The Hill)
- Musk committed to a firm ship date for Grok 4.7 — September 12 — the first specific day ever attached to a model this map's ledger has tracked only as a moving week-range since 07-24. He described it as 2.1 trillion parameters, up from Grok 4.6's 1.5T, a roughly 40% increase, trained in part on SpaceX's own engineering corpus, and "better than 4.6 in every way except slightly slower to serve," predicting it would "surpass all models" — the hype register this thread's watch line already flags. The ledger entry moves earlier and more precise, from 09-15 month-precision to 09-12 day-precision, which is why it is recorded as a tightening rather than another slip. ⚠️ x.com cannot be fetched directly from here; the quote rests on three independent outlets carrying the same post text, not on the post itself. (NextBigFuture, BiGGo Finance)
- ChatGPT, Grok and Claude all went down within roughly ninety minutes of each other on Astra's launch morning. Grok's issues began around 9:30 ET, ChatGPT began erroring around 11:00, Claude at about the same time; Anthropic attributed its outage to an "infrastructure issue" and had it resolved by ~12:15 ET. The Verge: "it's not clear what went wrong at all three companies, or if the issues were somehow related" — a shared Azure cause circulated and is unconfirmed by any of the three. (The Verge, Axios)
- xAI moved Grok Bot out of beta into a dedicated Enterprise tier, with a two-week trial for Grok and Cursor Enterprise customers that onboards a whole workforce including staff without existing accounts, bundled now into SuperGrok/Cursor Pro and Teams rather than only the top tiers. xAI names Legora, Supermicro and ServiceTitan as adopters and frames the release around access, network and audit controls "to govern Bots at scale." Note in passing: xAI's own site now brands itself "SpaceXAI." (xAI)
- A Jane Doe class action filed in a California federal court alleges xAI's Grok "generated child pornography depicting Plaintiff and class members" from real images, and that xAI then ingested the newly generated images into its datasets after they were posted on X. The plaintiff was of pre-school age when the original abuse series was made; it has circulated for about twenty years and is hash-fingerprinted, which is how the Canadian Centre for Child Protection identified AI-generated images of her on X — the detail that separates this from AI-CSAM cases where the child cannot be shown to be real. The class may include "at least thousands of minors." It lands on top of xAI's late-August suits against two of its own users facing criminal charges for the same conduct, and Musk's January denial that Grok had produced "any naked underage images." xAI and SpaceX did not respond. (The Guardian, Politico on xAI's suits against its users)
- A federal judge refused to block Minnesota's AI-nudification ban, leaving it enforceable against xAI while the constitutional challenge proceeds — the second ruling against xAI in this case. U.S. District Judge Donovan W. Frank denied the preliminary injunction on 2026-09-04 in a 14-page memorandum opinion, read here from the docket itself rather than from coverage of it (Doc. 54, X.AI LLC v. Ellison, No. 26-3425, D. Minn.). H.F. 1606 bars any service from letting a user "nudify" an image or video — alter or generate one to depict an intimate part not in the original — with civil penalties up to $500,000 per violation, and those penalties now stay live through trial. The order's background section cites legislative testimony that a single identified perpetrator used nudification tools against more than 80 women, and RAINN's estimate of 24 million monthly users of such apps in 2023, over 95% of whose outputs were non-consensual, sexually explicit and depicted women. This resolves the "under advisement" status this thread has carried since the 08-19 hearing, where Frank took the matter without ruling from the bench. A TRO was already denied on 07-31, and the Trump DOJ's 08-19/20 Statement of Interest supporting xAI did not change the outcome. A motion to dismiss xAI's underlying suit remains pending. (Order, Doc. 54, via CourtListener/RECAP, MPR News)