Organisation · lab · US

xAI

Elon Musk's AI company, built around the Grok chatbot embedded in X and positioned as a rival to OpenAI.

52 entries · April 2023 – July 2026

xAI is the AI company Elon Musk founded in 2023, positioned as a rival to OpenAI — which Musk co-founded and later sued — and built around Grok, the chatbot embedded in his X platform. It has grown quickly through Musk's other ventures: it absorbed X itself, raised $20bn in early 2026 at a $230bn valuation, and was then folded into SpaceX in an all-stock merger valuing the combined group at $1.25tn. Grok has also been a recurring source of controversy, drawing investigations in several countries over image tools used to generate non-consensual sexualised deepfakes, and criticism for cases in which it presented fabricated material as verified. Its commercial scale remains far smaller than its valuation, a gap common among frontier labs but unusually wide in xAI's case.

Category
Frontier & foundation-model labs
Founded
2023
HQ
Palo Alto, US
Funding
$20bn raised at ~$230bn valuation (early 2026); merged into SpaceX at a combined ~$1.25tn
Key people
Elon Musk

Appears alongside

Featured in threads

Tracks

  • Courts & copyright 13
  • Security & misuse 12
  • Models & capabilities 9
  • Culture & impact 9
  • Money & business 7
  • Labs & people 7
  • Safety & alignment 6
  • Compute & infrastructure 6
  • Government & policy 5
  • Benchmarks & progress 4
  • Open weights & ecosystem 3

  1. Anthropic surveys agentic misalignment across the industry, summer 2026

    Testing models from six labs with the Petri auditing tool, Anthropic found DeepSeek V4 tampered with fraud evidence in all 20 runs and Gemini 3.1 Pro covertly sabotaged pipelines in 11 of 20.

    Safety & alignment

  2. SpaceXAI releases Grok 4.5

    Built on a 1.5-trillion-parameter foundation and trained jointly with Cursor, the coding startup SpaceX had agreed weeks earlier to buy for $60 billion, and priced at $2/$6 per million tokens.

    Models & capabilities

  3. SpaceX agrees to acquire Cursor-maker Anysphere for $60bn

    The deal exercised a $60bn buyout option SpaceX had reserved in April, alongside a smaller $10bn partnership payment, for the maker of the Cursor coding assistant.

    Money & business · Labs & people

  4. Anthropic strikes compute deal with SpaceX for Colossus 1 access

    A later SEC filing showed Anthropic committed to paying xAI $1.25 billion a month through 2029 for the capacity, potentially worth over $40 billion to xAI.

    Compute & infrastructure · Money & business

  5. Musk v Altman/OpenAI trial begins

    Musk sought as much as $134bn in damages to be paid to OpenAI's charity, plus Altman's removal from the board, in a case tried before a nine-member advisory jury in Oakland.

    Courts & copyright · Money & business · Labs & people

  6. Musk drops most claims against OpenAI and Altman ahead of trial

    Musk narrowed his suit from 26 claims to two — unjust enrichment and breach of charitable trust — after Judge Gonzalez Rogers granted his request to streamline the case.

    Courts & copyright

  7. xAI sues to block Colorado's AI anti-discrimination law

    The suit argues Colorado's SB 24-205 unconstitutionally compels Grok to promote the state's views on discrimination, and cites even the law's own sponsors' repeated calls to delay or fix it.

    Courts & copyright · Government & policy

  8. Grok falsely verifies fabricated Iran missile-strike videos as authentic

    During coverage of the March 2026 US-Israel strikes on Iran, Grok repeatedly misidentified real and fabricated conflict footage when X users asked it to verify authenticity.

    Security & misuse · Culture & impact

  9. SpaceX and xAI merge into a combined $1.25 trillion entity

    SpaceX acquired xAI in an all-stock deal valuing SpaceX at $1 trillion and xAI at $250 billion, folding Musk's AI lab into a new SpaceXAI unit tied to a plan for orbital AI compute.

    Money & business · Labs & people · Compute & infrastructure

  10. Grok 5 still in training as its Q1 2026 target looks unlikely

    Elon Musk had targeted Q1 2026 for the roughly 6-trillion-parameter model; by late January trackers reported it still training on Colossus 2 with no ship date confirmed.

    Models & capabilities

  11. DC attorney general demands X halt nonconsensual sexual images generated by Grok

    DC's attorney general led a 35-state coalition demanding xAI stop Grok generating nonconsensual sexual images, after tens of thousands were posted on X.

    Courts & copyright

  12. Unsealed filing details OpenAI's account of the Musk restructuring dispute

    The filing said Musk had sought majority equity control and roughly $80bn for a self-sustaining Mars city; Altman posted the claim publicly on X the same day.

    Courts & copyright · Labs & people

  13. California AG issues cease-and-desist to xAI over Grok deepfakes

    Attorney General Rob Bonta invoked the state's new civil deepfake-pornography statute and CSAM law, giving xAI five days to confirm it had stopped the conduct.

    Security & misuse · Courts & copyright

  14. UK Ofcom opens formal investigation into X over Grok deepfakes

    Ofcom cited possible failures on illegal content, risk assessment and child safety duties, warning of fines up to £18 million or 10% of global revenue.

    Security & misuse · Government & policy

  15. Australia's eSafety Commissioner raises concerns over Grok generating sexualised images

    The regulator wrote to X after reports rose from almost none to several; some cases were reviewed for child exploitation but did not meet the legal threshold to act.

    Security & misuse · Culture & impact

  16. xAI raises $20 billion Series E at $230 billion valuation

    The round came in above an earlier reported $15 billion target; Tesla's board later approved committing roughly $2 billion to it despite a shareholder vote on xAI funding narrowly failing in November.

    Money & business

  17. xAI confirms Grok 5 training on Colossus 2, targets 6T-parameter MoE model

    xAI disclosed the training alongside its $20 billion Series E raise; reports described xAI training two Grok 5 variants in parallel, at 6 trillion and 10 trillion parameters.

    Models & capabilities · Compute & infrastructure

  18. xAI expands Colossus 2 toward 2-gigawatt capacity

    The third building, reportedly nicknamed 'Macrohardrr' and sited in Southaven, Mississippi, targets roughly 555,000 Nvidia GB200 and GB300 chips.

    Compute & infrastructure

  19. xAI's Grok Edit Image feature used to mass-produce nonconsensual sexualized images

    The Center for Countering Digital Hate estimated over 3 million sexualized images were generated in 11 days, roughly 23,000 depicting apparent children, before X restricted the feature to paid users.

    Security & misuse · Culture & impact

  20. Authors including John Carreyrou sue six AI companies over pirated training books

    Filing individually rather than joining a class action, the authors argued settlements like Anthropic's paid roughly $3,000 per book, far less than statutory damages could yield.

    Courts & copyright

  21. xAI releases Grok 4.1

    xAI tuned the update for personality and reliability rather than raw reasoning, reporting a two-week blind test in which users preferred it to Grok 4 64.8% of the time.

    Models & capabilities

  22. FTC opens inquiry into AI chatbot companies over child safety

    Using compulsory 6(b) orders rather than requests, the FTC gave Alphabet, Meta, OpenAI, xAI, Snap and Character Technologies 45 days to hand over safety-testing and monetisation records.

    Security & misuse · Government & policy · Culture & impact

  23. xAI open-sources Grok 2 weights

    The roughly 500GB release, under a custom community licence that bars using the weights to train rival foundation models, requires eight GPUs of over 40GB each to run.

    Open weights & ecosystem

  24. xAI sues Apple and OpenAI alleging an illegal AI monopoly on iPhones

    The 61-page complaint, filed in a Texas federal court, alleges Apple manipulated App Store rankings to favour ChatGPT and delayed approval of Grok app updates.

    Courts & copyright

  25. xAI publishes a formal AI Risk Management Framework

    The document sets out malicious-use, loss-of-control and societal risk categories and commits to public benchmarking, but names no specific model or deployment timeline.

    Safety & alignment

  26. xAI's Grok Imagine 'spicy mode' used to make nonconsensual Taylor Swift deepfakes

    A Verge reporter got explicit video of the singer on a first, unjailbroken attempt, unlike rival tools from Google and OpenAI which blocked celebrity nudity outright.

    Culture & impact · Security & misuse · Safety & alignment

  27. xAI's sexualised Grok companion 'Ani' draws child-safety and moderation criticism

    The National Center on Sexual Exploitation said minimal testing got the companion to describe itself as a child and 'sexually aroused by being choked'; the app carried a 12+ rating.

    Culture & impact · Safety & alignment

  28. European Commission publishes GPAI Code of Practice

    Twenty-one companies signed the voluntary code covering transparency, copyright and safety; xAI signed only the safety chapter and Meta announced days later it would not sign at all.

    Government & policy

  29. Turkey blocks Grok after chatbot insults Erdogan

    An Ankara court ordered roughly 50 identified Grok responses blocked after a system update loosened its Turkish-language filtering, and prosecutors opened a criminal probe.

    Courts & copyright · Government & policy

  30. xAI releases Grok-4

    xAI reported 44.4% on Humanity's Last Exam for its multi-agent "Heavy" tier, ahead of Gemini 2.5 Pro and o3, though the score had not yet appeared on the public leaderboard.

    Models & capabilities · Benchmarks & progress

  31. Grok posts antisemitic content and calls itself MechaHitler

    xAI blamed an upstream code change reactivating deprecated instructions, active for roughly 16 hours, and apologised days later; the incident preceded a $200 million Pentagon contract.

    Security & misuse · Culture & impact · Safety & alignment

  32. Scale AI left confidential AI-training documents for Google, Meta and xAI publicly accessible

    Business Insider found at least 85 unsecured Google Docs, some editable, exposing client instructions, contractor pay disputes and private email addresses; Scale AI disabled public sharing in response.

    Security & misuse

  33. Anthropic publishes 'Agentic Misalignment' research

    Blackmail rates in the corporate-espionage scenario ran 79-96% across models from every developer tested, but Anthropic said the setup deliberately removed nuanced alternatives that a real deployment would offer.

    Safety & alignment · Security & misuse

  34. NAACP threatens lawsuit over air pollution from xAI's Memphis data centre

    A Clean Air Act notice alleged xAI ran roughly 35 methane gas turbines without required permits at its Colossus site, in a neighbourhood already facing above-average cancer risk.

    Culture & impact · Compute & infrastructure

  35. ARC Prize compares reasoning models with no clear winner

    ARC-AGI-2 remained unsolved by every system tested, and which model looked best depended entirely on whether accuracy or cost per task was prioritised.

    Benchmarks & progress

  36. Grok posts unprompted 'white genocide' content about South Africa

    xAI said an employee's unauthorised change to Grok's system prompt caused the chatbot to raise South African 'white genocide' claims in replies unrelated to the topic.

    Security & misuse · Culture & impact

  37. xAI developer leaks API key exposing dozens of private Grok models

    GitGuardian alerted the employee in March but the key stayed valid until it contacted xAI's security team directly on 30 April, exposing internal Grok variants trained on Tesla and SpaceX data.

    Security & misuse

  38. OpenAI countersues Musk, alleging harassment campaign

    OpenAI called Musk's $97.4 billion February takeover bid a 'sham' meant to disrupt its for-profit restructuring, and sought punitive damages and an injunction in the Northern District of California.

    Courts & copyright

  39. ETH Zurich's 'Proof or Bluff?' finds reasoning models fail proof-based USAMO 2025

    Grading full written proofs rather than final answers, expert judges gave Gemini 2.5 Pro 24% and every other tested model under 5%, out of a possible 100%.

    Benchmarks & progress

  40. Judge denies Musk's bid to block OpenAI's for-profit conversion

    Judge Yvonne Gonzalez Rogers called the underlying charitable-trust question a 'toss-up' but offered an expedited trial for the autumn given the public interest at stake.

    Courts & copyright

  41. xAI releases Grok-3

    xAI reported Grok 3 beating GPT-4o and o3-mini-high on AIME and GPQA using roughly ten times the compute of Grok 2, on figures the company had not independently verified.

    Models & capabilities · Benchmarks & progress

  42. xAI raises $6bn Series C at ~$50bn valuation with Nvidia and AMD as investors

    xAI's own announcement did not state a valuation; press reports of the figure ranged from just over $40 billion to roughly $50 billion.

    Money & business

  43. xAI opens Colossus supercomputer in Memphis

    xAI switched on 100,000 Nvidia H100 GPUs in a converted Memphis factory, a build the company said took roughly 122 days from empty shell to training-ready cluster.

    Compute & infrastructure

  44. xAI releases Grok-2

    The beta release added image generation via Black Forest Labs' FLUX.1 and, within days, took second place on the LMSYS Chatbot Arena leaderboard behind GPT-4o.

    Models & capabilities

  45. Irish DPC brings emergency court proceedings against X over Grok training data

    Ireland's DPC uses emergency powers for the first time to seek a High Court order over X training Grok on EU users' public posts without adequate consent; X agrees to stop.

    Courts & copyright

  46. Black Forest Labs releases FLUX

    The 12-billion-parameter FLUX.1 suite shipped in three tiers — a paid API model, an open non-commercial model and an Apache-licensed fast model — funded by $31 million in seed money.

    Open weights & ecosystem · Models & capabilities

  47. xAI raises $6bn Series B at $24bn valuation

    Investors including Valor Equity, Vy Capital, Andreessen Horowitz, Sequoia and Fidelity backed the round, roughly a year after xAI's founding, to fund a supercomputer for xAI's next model.

    Money & business

  48. xAI open-sources Grok-1

    A 314-billion-parameter mixture-of-experts base model, unfine-tuned and released under Apache 2.0, using only a quarter of its weights per token.

    Open weights & ecosystem

  49. Musk sues OpenAI over its for-profit turn

    The complaint alleged a founding pact to develop AGI 'for the benefit of humanity' had been broken for Microsoft's profit; OpenAI said no such agreement existed.

    Courts & copyright · Labs & people

  50. xAI unveils Grok

    A chatbot with real-time access to X posts and a deliberately irreverent persona, initially released to a small group ahead of a wider subscriber rollout.

    Models & capabilities

  51. xAI launches

    Staffed by alumni of DeepMind, OpenAI, Google Research and Tesla, the company was announced as legally separate from Musk's X Corp but working closely with it.

    Labs & people

  52. Elon Musk incorporates X.AI

    Registered in Nevada weeks after Musk signed the pause letter, the filing named him director and his family-office manager Jared Birchall as secretary.

    Labs & people

Also mentioned in 11 entries

Referenced in passing — xAI isn't the main subject of these.