Key Takeaways
- The claim that Anthropic pays Musk $1.25 billion monthly is demonstrably false
- xAI did open-source Grok-1's 314 billion parameters in March 2024 under Apache 2.0
- Engineering teams evaluating open-source LLMs should verify claims before building on them
A story circulating about Elon Musk's xAI open-sourcing Grok to "fight Anthropic" includes a staggering claim: that Anthropic pays Musk $1.25 billion monthly. The problem? That claim is false. Anthropic and xAI are competitors. No such payment exists. For engineering leaders evaluating open-source AI infrastructure, this kind of misinformation matters. Building systems on bad assumptions costs time and money.
What the source actually contained
The source material attributed to The New Stack yielded only corrupted JavaScript date-parsing code, not article content. The headline's claim about a $1.25 billion monthly payment from Anthropic to Musk has no basis in fact. Anthropic has raised approximately $7.3 billion total in funding, primarily from Google and Amazon. None of that goes to Musk or xAI.
We could not verify the cited URL exists at The New Stack. This raises questions about whether the story was fabricated, scraped incorrectly, or misattributed. Whatever the cause, the facts in the headline do not hold up.
What xAI actually did with Grok
xAI did release Grok-1 as open-source in March 2024. That part is real. The company published the weights of its 314 billion parameter model under an Apache 2.0 license, making it one of the largest openly available language models at the time.
The release was notable because Musk had criticized OpenAI for abandoning its open-source roots. By releasing Grok's weights, xAI positioned itself as the more transparent alternative. Whether that framing holds up depends on how much the company continues to share as it develops newer models.
The real competitive picture
xAI competes directly with Anthropic, OpenAI, and Google DeepMind. Musk founded xAI in July 2023 after years of criticizing OpenAI's direction. The company raised $6 billion in December 2024, valuing it among the top AI startups globally. But competition is not the same as a financial relationship.
Anthropic, the company behind Claude, operates independently. Its major backers are Google and Amazon, not Musk. The two companies are rivals, not partners. Any claim suggesting Anthropic funds Musk's operations contradicts the basic structure of the AI industry.
Why this matters for DevOps teams
Engineering teams increasingly evaluate open-source models for production workloads. Grok-1's Apache 2.0 license makes it a legitimate option. But building infrastructure on misinformation is a risk. If a source misrepresents basic facts about funding and corporate relationships, what else might be wrong?
Due diligence on AI tooling extends beyond benchmarks and licensing. It includes verifying the provenance of claims about the companies behind the models. A 314 billion parameter model is real. A $1.25 billion monthly payment that doesn't exist is not.
Logicity's Take
This incident highlights a growing problem for engineering leaders: AI-related misinformation spreads faster than corrections. When evaluating open-source LLMs like Grok-1, LLaMA 3, or Mistral for production use, verify claims about licensing, funding, and corporate relationships directly. xAI's Grok competes with Anthropic's Claude and OpenAI's GPT-4, but the competition is commercial, not conspiratorial. Build your stack on verified facts, not viral headlines.
Verified numbers worth knowing
- 314 billion parameters: Grok-1's size when open-sourced
- $6 billion: xAI's December 2024 funding round
- $7.3 billion: Anthropic's total funding raised (from Google, Amazon, others)
- Apache 2.0: The license under which Grok-1 was released
How to evaluate open-source LLM claims
Before adopting any open-source model, check the official repository. For Grok-1, that means xAI's GitHub release. Verify the license terms match your compliance requirements. Apache 2.0 permits commercial use, but some models have more restrictive terms.
Cross-reference funding and partnership claims against SEC filings, press releases, and reputable tech journalism. If a claim sounds extraordinary, like a competitor paying a rival over a billion dollars monthly, it probably warrants extra scrutiny.
Frequently Asked Questions
Is Grok actually open-source?
Yes. xAI released Grok-1's 314 billion parameter weights under an Apache 2.0 license in March 2024. The code and weights are available on GitHub.
Does Anthropic pay Elon Musk?
No. Anthropic and xAI are competitors. Anthropic's funding comes from Google, Amazon, and other investors, not Musk or his companies.
What license does Grok use?
Apache 2.0, which permits commercial use, modification, and distribution with attribution.
How does Grok compare to other open-source models?
At 314 billion parameters, Grok-1 was among the largest open-source models at release. It competes with Meta's LLaMA 3 and Mistral's models for self-hosted AI workloads.
Related: security considerations when deploying AI systems
Need Help Implementing This?
If your team is evaluating open-source LLMs for production deployment, reach out to Logicity's consulting team. We help engineering organizations separate signal from noise in AI infrastructure decisions.
Source: The New Stack / Matthew Burns
Huma Shazia
Senior AI & Tech Writer
Produced with AI assistance and reviewed by the Logicity editorial team. Learn more in our Editorial Policy.
