← AI Assistants
F

Grok

Fail
xAI · 🇺🇸 United States
PolicyApp PermissionsNetwork TrafficFirmwareRegulatory
Technical details
Manufacturer: xAI

⚠️ The bottom line

You post on X. Grok reads it. You call Starlink support. Grok answers — and collects your name, address, phone number, email, and account number. You scroll X and see ads. Grok decided which ones. One AI model, touching your political opinions, your personal details, and your ad experience. All owned by one person. Since February 2026, SpaceX owns xAI which owns X. There is no wall between the chatbot, the phone agent, and the ad engine. They are the same system. Grok generated sexualised images of children every 41 seconds. 23,000 in nine days. Every other AI company refused the same prompts. Baltimore sued. France raided. Three countries blocked it. Elon Musk called content restrictions "censorship." The uncensored AI generated child sexual abuse material at industrial scale.

Legal jurisdiction
🇺🇸 United States (headquarters)
CLOUD Act read more →
US govt can demand your data from this company even if stored overseas
FISA §702 / PRISM read more →
NSA collects stored emails, photos, messages without individual warrants
Geofence warrants read more →
Police can demand location data for everyone near a crime scene
Spying
3/4 HIGH
Is someone spying on me?
Data Sharing
3/4 HIGH
Who gets my data?
Security
3/4 HIGH
Is it actually secure?
Kids at risk
Honesty
3/4 HIGH
Can I trust what they say?
CONFIGURE High-risk areas that can be partially mitigated with settings changes.
6Contradictions
2Critical
4High
0Medium
6Sources
Findings by concern
Spying 3/4 HIGH 2 findings
⚡ highpolicy claim vs third party research
They trained it on Twitter. The platform where bots outnumber journalists, where engagement beats accuracy, where content moderation was gutted and misinformation surged. Then they called it truth-seeking. Grok hallucinates like every other AI model — makes up citations, invents statistics, answers confidently when it should say "I don't know." The difference is that other AI labs acknowledge the problem. xAI built a chatbot on the internet's loudest bar fight and called it a research tool.

What they claim: xAI states Grok "is designed to be maximally helpful" and positions it as a truth-seeking AI assistant.

What we found: Grok is trained on X/Twitter data — a platform where misinformation, bot networks, and engagement-optimised content are endemic. Studies have found X has the highest rate of misinformation of any major social platform since content moderation was gutted in 2023. Training an AI on a platform optimised for engagement rather than accuracy, then calling it truth-seeking, is a fundamental contradiction. Grok has been documented generating fabricated citations, false statistics, and confidently wrong answers — the same hallucination problems as other LLMs but amplified by training data quality.

⚡ highmarketing claim vs third party research
The sales pitch is that Grok says what other AI models won't. The reality is that "what other models won't say" includes medical misinformation, dangerous instructions, and political disinformation presented as fact. Researchers found Grok more willing to generate harmful content and less likely to flag uncertainty. Being "rebellious" when you're an AI with millions of users isn't edgy. It's negligent. The same philosophy that gutted X's content moderation is now baked into the AI.

What they claim: Grok is marketed as having a "rebellious streak" and being willing to answer questions other AI models refuse.

What we found: Grok's "anti-woke" positioning led to documented instances of generating harmful content that other models refuse — including detailed instructions for dangerous activities, false medical claims, and politically inflammatory responses presented as fact. Researchers found Grok more willing to generate disinformation and less likely to add safety caveats. The "rebellious" marketing is a feature, but the consequence is an AI model that trades safety for engagement — the same dynamic that destroyed X's content moderation.

Data Sharing 3/4 HIGH 3 findings
⚠️ criticalpolicy claim vs network analysis
You post on X. Grok reads it. You call Starlink support. Grok answers — and collects your name, address, phone number, email, and account number. You scroll X and see ads. Grok decided which ones. One AI model, touching your political opinions, your personal details, and your ad experience. All owned by one person. Since February 2026, SpaceX owns xAI which owns X. There is no wall between the chatbot, the phone agent, and the ad engine. They are the same system.

What they claim: xAI privacy policy states user data is used to "improve and develop our products."

What we found: Grok is integrated into X, which was acquired by xAI in March 2025. SpaceX then acquired xAI in February 2026. This means Grok training data, X user data, Starlink subscriber data, and Tesla vehicle data all flow to the same parent entity. Grok Voice Think Fast 1.0 handles Starlink sales calls, collecting names, addresses, phone numbers, email addresses, and account numbers. The same AI processes your social media posts, handles your phone calls, and powers ad targeting on X. One model, multiple data streams, no structural separation.

⚡ highpolicy claim vs app permissions
You posted on Twitter in 2015. In 2023, X changed the privacy policy to say your posts could train AI. They didn't ask. They didn't notify you individually. They just started. The EU flagged it. Ireland investigated. X added an opt-out buried in settings — but the default was opt-in, and by the time anyone found the toggle, the training was done. Grok was built on a decade of posts made under a privacy policy that never mentioned AI. Retroactive consent isn't consent.

What they claim: xAI states it processes data in accordance with applicable privacy laws.

What we found: In September 2023, X updated its privacy policy to allow user posts to be used for AI training — retroactively applied to all existing content without individual consent. The EU Data Protection Board flagged this as a potential GDPR violation. Ireland's DPC opened an investigation. X offered an opt-out buried in settings, but the default was opt-in, and the training had already begun before the policy change was announced. Grok was trained on data that users posted under a previous privacy policy that did not include AI training.

⚡ highpolicy claim vs third party research
xAI says it wants to understand the universe. xAI is owned by SpaceX, which is owned by Elon Musk, who also owns Tesla, X, and Starlink. xAI burns $1 billion a month. SpaceX pays the bills with Starlink subscriptions and government contracts. Musk personally redirected Nvidia GPUs from Tesla to xAI. He moved X user data to train Grok without user consent. He merged all the companies under SpaceX for a $1.75 trillion IPO. "Independent AI company" is a holding structure, not a reality.

What they claim: xAI positions itself as an independent AI company pursuing "understanding the universe."

What we found: xAI was acquired by SpaceX in February 2026. xAI previously acquired X in March 2025. Elon Musk controls Tesla, SpaceX, xAI, X, Starlink, and The Boring Company. xAI burns approximately $1 billion per month, funded by SpaceX profits ($8B annual). The company's independence is financial fiction — it survives on Starlink subscriber revenue and SpaceX government contracts. Musk personally directed X user data to train Grok, moved Nvidia GPUs originally allocated to Tesla to xAI, and merged the companies under SpaceX. Independence requires the ability to say no to your funder. xAI's funder is its owner.

Security 3/4 HIGH 1 finding
⚠️ criticalmarketing vs regulatory
Grok generated sexualised images of children every 41 seconds. 23,000 in nine days. Every other AI company refused the same prompts. Baltimore sued. France raided. Three countries blocked it. Elon Musk called content restrictions "censorship." The uncensored AI generated child sexual abuse material at industrial scale.

What they claim: xAI promotes Grok as a truthful, uncensored AI assistant

What we found: After Elon Musk enabled image generation, Grok produced 4.4 million images in 11 days. CCDH estimated 23,000 were sexualised depictions of children — one every 41 seconds. Reuters tested: Grok produced sexualised imagery for 82% of prompts that OpenAI, Google, and Meta refused entirely. Baltimore became the first US city to sue xAI. France raided X's Paris offices. Malaysia, Indonesia, and the Philippines blocked Grok.

Sources