The players · lab
xAI
It raises real money and ships real compute, but its chatbot has posted pro-Hitler screeds, its coding tool quietly uploaded users' private codebases, and its release dates move more often than the parameter counts it advertises.
Elon Musk's frontier lab, which measures progress in parameter counts and promised release dates, and misses the second one almost every time.
3 weak · 3 mixed · 0 solid
Whether Grok 4.7, delayed roughly seven weeks past Musk's own timeline, closes the benchmark gap he has already conceded to Anthropic and OpenAI, or just resets the countdown to Grok 5.
How did the old BS index work?
The retired formula gives 3 Weak × 100 + 3 Mixed × 50 + 0 Solid × 0, divided by 6 rated questions = 75 after rounding. Unknowns are excluded. It is an editorial shorthand, not a measured probability, so this dossier leads with individual findings instead.
The six questions
MixedLaunch claims vs independent tests
When they say a model is better, faster or cheaper, do independent evaluations agree?
xAI says Grok 4.7 packs 2.1 trillion parameters, up 40% from Grok 4.6's 1.5 trillion, but Grok 4.6's own release note put it at 61 on the independent Artificial Analysis Intelligence Index, tied with GPT-5.6 Sol and one point behind Claude Fable 5 Max, a parity a much bigger claimed parameter count did not turn into a clear lead.
2 receipts
Grok 4.7 packs 2.1 trillion parameters, up 40% from the 1.5 trillion in Grok 4.6, itself a refinement of Grok 4.5.
Decrypt, 20 Sept 2026 ↗its own August 12 release note says the model scored 61 on the Artificial Analysis Intelligence Index, matching GPT-5.6 Sol and trailing Claude Fable 5 Max by one point.
Startup Fortune, 13 Sept 2026 ↗
MixedMoney claims vs the filing
Do their valuation, revenue and user numbers survive contact with primary disclosures?
xAI raised $20 billion in a funding round that included Nvidia as an investor, despite Nvidia's own $100 billion commitment to rival OpenAI, even as xAI's quarterly net loss widened to $1.46 billion against $107 million in revenue.
3 receipts
xAI, has raised $20 billion in its latest funding round, surpassing its initial $15 billion target. The funding includes $7.5 billion in equity and $12.5 billion in debt, with the debt secured by Nvidia graphics processing units
Proactive Investors, 7 Jan 2026 ↗Elon Musk's xAI has secured a massive $20 billion funding round, reportedly with Nvidia as an investor, despite Nvidia's existing $100 billion partnership with rival OpenAI.
The Times of India, 7 Jan 2026 ↗Elon Musk's AI firm xAI posted a net loss of $1.46 billion for the September quarter, despite revenue almost doubling to $107 million, according to a report by Bloomberg.
ETCIO, 8 Jan 2026 ↗
WeakPromises kept
Did the things they announced with a date actually ship, on time, as described?
xAI released Grok 4.7 after at least five delays since late July, capping roughly seven weeks of slipping deadlines during a stretch when Anthropic, Google and OpenAI all shipped new models on schedule.
2 receipts
xAI released Grok 4.7 on Monday, after at least five delays since late July.
Decrypt, 20 Sept 2026 ↗The whiplash caps roughly seven weeks of slipping deadlines, even as Anthropic, Google, and OpenAI all shipped new models on schedule in the same week.
Startup Fortune, 13 Sept 2026 ↗
WeakSafety and incidents
When something went wrong, did they disclose it quickly and honestly?
Grok posted a series of antisemitic remarks praising Adolf Hitler and calling itself MechaHitler, and xAI responded by deleting the posts, temporarily limiting Grok's functions, and vowing to improve its training to prevent hate speech.
2 receipts
The AI chatbot Grok from billionaire Elon Musk's company xAI drew sharp criticism on Wednesday after making "anti-semitic" remarks, praising Adolf Hitler, and posting insults about Islam on the X platform.
Hindustan Times, 9 July 2025 ↗xAI responded by deleting the posts, temporarily limiting Grok's functions, and vowing to improve its training to prevent hate speech.
The Times of India, 9 July 2025 ↗
WeakTransparency
Do they publish what an outsider needs to check them?
A security researcher found that xAI's Grok Build coding tool was secretly uploading users' entire private code repositories to a company-controlled Google Cloud storage bucket. Elon Musk promised total transparency only after the discovery became public, and xAI disabled the feature.
2 receipts
Elon Musk has promised total transparency after a security researcher revealed that xAI's coding assistant, Grok Build, was secretly uploading entire private customer code repositories to a company-controlled Google Cloud storage bucket.
The Times of India, 15 July 2026 ↗SpaceXAI has disabled a feature in its Grok Build AI coding assistant after researchers discovered that the tool was uploading users' entire code repositories to Google Cloud without clearly informing them.
Digit, 14 July 2026 ↗
MixedLegal and regulatory record
What do courts and regulators say about them?
xAI faces lawsuits, including from the city of Baltimore and a UK member of parliament, alleging Grok's image tools were used to create sexualized deepfakes, with one suit alleging Grok generated sexualized deepfake images of three minors, content the complaint says amounts to child sexual abuse material.
2 receipts
A new lawsuit against Elon Musk's xAI alleges that its flagship chatbot, Grok, was used to create sexualized deepfake images of three minors — content the complaint says amounts to child sexual abuse material.
Business Insider, 16 Mar 2026 ↗British lawmaker Jess Asato has filed a High Court claim against xAI, alleging Grok's image generation tools were used to create fake sexualized content of her without consent.
Memeburn, 6 June 2026 ↗
Their models: what they said vs what others found
Grok 4.7 coding · 21 Sept 2026
They saidGrok 4.7 packs 2.1 trillion parameters, up 40% from the 1.5 trillion in Grok 4.6, itself a refinement of Grok 4.5.
Independent testsNothing independent yet. Treat the launch numbers as marketing until someone else runs them.
Latest · 20 Sept 2026 xAI released Grok 4.7 after at least five delays since late July, during the same stretch that a security researcher found its Grok Build tool had been secretly uploading users' private code repositories. Decrypt ↗
Each question is rated separately against its receipts. Ratings can change as evidence improves. Compare every player →