From OpenAI to xAI

In 2015, Elon co-founded OpenAI alongside Sam Altman and others. The original mission was to develop artificial general intelligence safely and ensure its benefits were distributed broadly — hence the "open" in the name. Elon was a major early donor and board member.

He departed from OpenAI's board in 2018, citing conflicts of interest with Tesla's own AI work. Over the following years, his relationship with OpenAI soured dramatically. He became a vocal critic of the organization's shift toward a for-profit model and what he saw as a departure from its founding mission. In 2024 he filed a lawsuit against OpenAI and Sam Altman, alleging breach of their founding agreement.

In 2023, Elon founded xAI — a new AI company with the stated goal of building AI that is maximally curious, truthful, and beneficial to humanity. xAI created Grok, a conversational AI integrated directly into the X platform. Elon has released open-weight versions of some Grok models, making them freely available for researchers and developers to use and modify.

Grok and the Colossus Supercomputer

Grok is xAI's flagship AI model. It is available to X Premium subscribers and competes directly with ChatGPT, Claude, and Google's Gemini. Elon claims that Grok 3, released in early 2026, outperforms competing models on key benchmarks.

To train Grok, xAI built Colossus — a massive computing cluster in Memphis, Tennessee. Elon has described Colossus as the most powerful AI training facility in the world. It was assembled at extraordinary speed, going from empty building to operational in a matter of months. Elon posts frequently about Colossus expansions and the compute milestones xAI achieves.

One distinctive feature of Grok is its access to real-time data from X. Because X owns the firehose of public posts, Grok can answer questions about current events in ways that models trained only on static datasets cannot.

What Elon Says About AI Risk

Elon has been warning about the dangers of advanced AI for over a decade — longer than almost any other tech CEO. He has described AI as potentially the most transformative and most dangerous technology humanity has ever developed. He believes there is a non-trivial chance that a sufficiently advanced AI, if misaligned with human values, could pose an existential threat.

At the same time, he doesn't advocate for slowing AI down or banning its development. His position is that if powerful AI is coming regardless, it's better to have safety-conscious developers at the frontier than to cede that ground to those who are less careful. He founded xAI in part because he felt no existing lab had the right combination of ambition and caution.

He is also a strong advocate for open-source AI — the idea that AI models should be publicly available, not locked behind corporate walls. His argument is that open models allow more people to identify problems, improve safety, and prevent any single company from monopolizing a technology that will reshape civilization.

Key Terms Explained

Artificial Intelligence (AI)

Computer systems that can do tasks that normally require human thinking — like understanding language, recognizing images, writing text, or solving problems. Modern AI learns from enormous amounts of data rather than being programmed with specific rules.

Large Language Model (LLM)

The type of AI behind Grok, ChatGPT, and Claude. Trained on vast amounts of text, these models learn to predict and generate language — which turns out to enable a surprisingly wide range of tasks.

xAI

Elon's AI company, founded in 2023. Its mission is to build AI that helps humanity understand the universe. It created the Grok AI model and built the Colossus supercomputer to train it.

Grok

xAI's AI assistant, available through X. The name comes from a sci-fi term meaning to understand something deeply and intuitively. It has access to real-time X posts, unlike most other AI models.

Open-Source / Open-Weight AI

AI models where the underlying code or trained model weights are made publicly available. Anyone can download, study, modify, and use them. Elon supports this approach as a check against any single company controlling powerful AI.

AGI (Artificial General Intelligence)

A hypothetical AI that can do any intellectual task a human can — and potentially much more. We don't have AGI yet. Current AI is narrow (good at specific tasks). AGI is the long-term goal of OpenAI, xAI, and others — and the thing Elon is most cautious about.

Elon's Posts on AI

Live post links will be added once X API integration is complete. Check back soon.