You see, I’ve been following the adventures of Elon Musk’s AI company xAI, and I’ve come to a conclusion. The only real idea is, “What if it’s Elon Musk and AI this time?”
Based on publicly available information about xAI, it appears that Musk showed up late to the generative AI party without even drinking a beer. Here he is in 2024. The party is crowded now. xAI doesn’t seem to have anything that stands out beyond Musk.
That didn’t stop Musk from pitching his idea to investors. Last December, xAI announced in a filing with the Securities and Exchange Commission that it was seeking to raise $1 billion. (This is not the same company as Hex, formerly known as Twitter.) financial times Musk has said he is seeking up to $6 billion in funding.
xAI (not Twitter) has one product so far, a supposedly cheeky LLM called Grok.
Just in case, Mr. Musk tweeted “xAI has not raised any funding and we have not discussed this with anyone.” Musk has said a lot of things in public, only some of which are true, so I I’m going to be swayed by this application that I saw with my own eyes.
xAI (not Twitter) is kind of weird. There doesn’t seem to be any clear purpose other than to interact with X (Twitter). The xAI pitch deck is obtained in the following way: bloomberg It depends on two things:
- OpenAI achieved great success in 2023. Musk, of course, was the founder of OpenAI and left the company in 2018 in a huff. He recently said he was offered his stock by OpenAI, but he turned it down.
- When you invest in xAI, you get access to “Muskonomy”. Notes: Those words are not my fault. It’s true that Mr. Musk is involved in a number of companies and is the putative CEO of several of them. Besides xAI (not Twitter) and hex (Twitter), there’s Neuralink, SpaceX, The Boring Company, and Tesla. We’ll be at Tesla soon.
xAI (not Twitter) has one product so far, a cheeky LLM called Grok. Users pay $16 per month to X (a Twitter company) and gain access through the X (Twitter) interface. xAI (not Twitter) does not have a standalone interface for Grok. My colleague Emilia David characterizes this as having “no reason to exist” because it’s not significantly better than free chatbots offered by competitors. Its most obvious feature is that it uses X (Twitter) data as real-time input, acting as a kind of operatic glass for platform drama. The Discover/Trends section of the Twitter app is being reworked internally to feature news summaries from Grok, according to people familiar with the development.
Grok was developed very fast. One possible explanation is that Musk hired the most in-demand team in the field. The other is that it’s a fine-tuned version of an open source LLM like Meta’s Llama. Maybe there’s a secret third factor in his life that explains his speedy development.
According to , besides X (Twitter), another data source for xAI (not Twitter) is Tesla. bloombergThis is a report by That’s interesting! Musk said in January that he would “like to develop products outside of Tesla” unless he was given “up to 25 percent voting rights.” Musk also said he doesn’t feel he has enough ownership in Tesla to feel comfortable “growing it into a leader in AI,” and that if Tesla stock doesn’t increase, he won’t “develop products outside of Tesla.” I like it,” he said.
Tesla has been working on developing AI related to self-driving cars for quite some time and has experienced some of the same obstacles as other self-driving car companies. I think there’s also an Optimus robot. These seem like specific use cases that are much less common than building another he LLM. Tesla’s data is valuable and goes back many years. If xAI is siphoning money, how will Tesla shareholders feel about that?
Who would want to fund yet another very popular AI company in a crowded field?
Certainly, AI has practical applications. Data bricks exist! It’s not intended for consumers, but it does seem to have a specific purpose: data storage and analysis. There are also smaller, specialized companies that handle industry-specific types of data. Take Fabric AI as an example. Its purpose is to streamline patient intake data for telemedicine. (He’s also created a chatbot that threatens to replace WebMD as the scariest place to ask about symptoms.) Or Abnormal Security, an AI approach to blocking malware, ransomware, and other threats. We don’t know if these companies will achieve their goals, but they at least have a compelling reason to exist.
So I’m wondering who would want to fund yet another very common AI company in a crowded space. And I wonder if Mr. Musk’s denial that he’s raising money at all means he doesn’t really care about xAI and is trying to minimize embarrassment. Why would one of the richest people in the world need outside funding for this anyway?
Silicon Valley’s reputation for Musk is surprisingly solid, perhaps because he has made so many people make so much money in the past. But the failure at X (Twitter) was disastrous for his investors. And Mr. Musk was distracted by it at a critical time for Tesla, which is facing increased competition. Tesla’s newest product, the Cybertruck, ships without a clear coat. Some owners say it’s rusty. (Tesla engineers claim the orange spots are surface contamination.) And Tesla warned in its recent earnings call that growth is slowing. Meanwhile, Rivian’s CEO has publicly stated that he is directly trying to undermine Tesla.
In a perhaps underappreciated development over the past 20 years or so, we’ve seen Elon Musk grow from an investment bellwether to a top signal. Take, for example, the “GameStonk” movement, where Musk’s tweets served as the perfect sell signal for individual investors as well as sophisticated hedge funds. Or the Dogecoin crash that happened when he called himself the Dogefather on SNL. Or even Twitter, which is certainly not worth the amount Mr. Musk ultimately paid and has rapidly declined in value since then, and whose transactional debt was classified as an investment by a firm specializing in distressed debt. It’s called “impossible.”
I don’t see a convincing case being made for xAI. It has no special purpose. Grok is also running as his LLM and is intended to enhance existing products. X.xAI isn’t selling AI-native applications, it’s mainly just saying, “Look at OpenAI.”
Well, I looked into it and ChatGPT is no longer growing.
As the hype around generative AI begins to die down, Musk is trying to pitch new AI startups without a clear focus. This isn’t just his ChatGPT. Microsoft’s Copilot’s usage drops sharply after his first month. There are currently open questions about whether the productivity gains from AI will be sufficient to justify its costs. What I’m wondering here is how many investors believe that “just adding Elon” will solve the problem?
With reporting by Alex Heath.
