Author: Ejaaz, Limitless; Compiler: BitpushNews

xAI has come under fire recently. Yet even though Musk has repeatedly proven over the decades that his extreme skeptics were wrong, I think people are still underestimating him.

Don’t forget, we’re talking about a two-year-old startup. But they built a breakthrough data center in 122 days (this usually takes 4 years), are pushing their products to 600 million monthly active users, and have something that other AI labs do not have - a physical carrier (yes, a humanoid robot).

I will analyze why xAI’s soon-to-be-released Grok 5 will not only equal its competitors, but is also expected to surpass them comprehensively.

Here are a few reasons why xAI may dominate the competition...

Musk’s energy giant network

In 2026, as xAI rapidly expands the scale of its data center, it will have a structural advantage in computing power. It is estimated that its currently operational computing power capacity is equivalent to approximately 500,000 GPUs, exceeding its top competitors.

In addition, through the efforts of the "Colossus One" and "Colossus Two" data centers, he aims to bring 900,000 GPUs online by the second quarter of this year, and is promoting more than 1 million GPUs in the short term.

How the hell do you compete with this scale? It’s not just about capital expenditures or even the hardware itself – it’s about how they get there.

For example, the power grids in Tennessee and Memphis couldn't fully support his data centers, so he flown in gas turbines to make up the difference. He chose to bypass the entire state’s power grid entirely—just to get there faster.

His thinking on the entire power architecture is also unique; he has deployed Tesla Megapack battery packs of up to 250 MWh in relevant facilities to ensure model training when power demand surges and the grid cannot support it.

In terms of "moving mountains and reclaiming seas" at the physical level to realize his vision, Elon's advantages are obvious; this is establishing xAI's huge computing power advantage over its competitors.

If the formula "more computing power = better model" still holds (it seems to be the case so far), then the rumored Grok 5 with 7 trillion parameters will become an absolute behemoth when it is released (a jump of more than 2 times from the 3 trillion parameter model of Grok 4).

You have to understand that the regulatory burden, talent acquisition, and operational logistics required to achieve this scale are unprecedented. xAI currently appears to be well ahead in the race to scale AI infrastructure, with a strategy of "execute first, ask questions later."

Unless other labs follow suit, xAI's model will continue to lead.

How can this compete?

Elon is literally playing Civilization in the real world, and he is number one in the AI rankings.

This guy flew power plants directly from abroad to power his GPU.

He expanded a data center to 300 megawatts in less than 4 months, which should have taken 4 years.

Well, for those who want to know the actual details:

"Macrohard is not just a bad joke - the word is actually painted on the roof of the Colossus 2 data center, clearly visible from the satellite image. It's a classic Musk spoof: the name is used to poke fun at software giants built from scratch using AI, in this case Microsoft.

X: The winner of the social media campaign

So, I think I've clarified the computing advantage of xAI, but top models require more than just GPUs - they also require massive amounts of data.

And it’s not just any data. More and more AI labs are realizing that real-time data is the key to unlocking personalized AI, which can deeply understand your desires and goals and help you achieve them.

Google's latest "personal intelligence" product may be the clearest indication that this will eventually be the direction of model development, but xAI has a unique advantage that competitors such as Google do not have...

None of its competitors has a social media platform that can deliver more than 100 million posts per day and map the pulse of global culture in real time. The X platform’s data deluge also allows the team to develop a deep understanding of virality and human behavior at scale—perhaps better than any other data set in the world.

Whereas other models only tell you what happened, Grok tells you what happened and how people felt about what happened - and faster than anyone else. It's hard to argue with X's data moat.

But it’s not just data, X has amazing distribution channels, with 250 million daily active users and 600 million monthly active users on the X and Grok apps. Every user will see an "Ask Grok" button next to every post on the platform.

It is not difficult to foresee that xAI will integrate multiple services in the same application in the future, such as real-time prediction of the market, shopping, banking and dating - all driven by Grok.

Today, most model labs base their valuations on GPUs, smart benchmarks, and reputation. xAI has all of this, and the opportunity to break into a number of different Internet monopolies - don't forget that their stated goal is to become a "universal app".

Today, Platform X’s algorithm is powered by Grok – it analyzes each post to make recommendations. Tomorrow, it will provide personal intelligence services to every user.

Obviously, Grok is much more than a standard large language model, and its valuation should reflect this.

Physics AI Advantages

At this point, it should come as no surprise that robotics will play a huge role in advancing the world over the next five years. Technology is finally smart enough.

Everything from manual labor in factories to last-mile delivery, fast-food chains, and elite surgeons will be assisted by robots, or even completely replaced.

Progress once limited to viral videos is coming out of the lab, with fleets of self-driving cars emerging and surprising humanoid robots on the horizon. However, after decades of development, when the two are combined, people still only think of one company - Tesla.

A car that drives better than humans is no longer a fantasy, the latest v14.2.2.3 update is already technically better than you at driving. Once regulations catch up - you'll see self-driving Teslas driving people around. Likewise, the vision of a personal humanoid robot is becoming a reality; Elon says Optimus will begin shipping to the public by the end of next year.

So, what does all this have to do with xAI?

Grok needs diverse data sources to understand the world around it, and that data will come from Tesla's robots.

Driving these machines requires a multimodal brain, and in Tesla's case, that will be Grok.

This symbiotic relationship between the two companies gives xAI an almost unfair advantage over its competitors. I think Google is the only company that can compete at this level, thanks to Waymo, but they are still lagging behind.

Today, Grok is integrated into Tesla vehicles - with the latest update, you can even ask Grok to command the FSD to drive you to your destination...while playing music for you...and teaching you Roman history.

The big game Musk is playing must be acknowledged. He is not just trying to build a big language model, he is trying to build the ecosystem that the entire AI relies on to survive and operate.

Even as I write this, I admit that this all sounds great, but also ambitious... which leads me to the last part of this article...

Yes, there are risks

Of course, risks are everywhere. Perhaps a person's energy ceiling is to run five companies at the same time, and the sixth one is the critical point... But Musk has silenced doubters too many times. This past credit deserves our more patience.

After all, what he did was extraordinary.

Nonetheless, in my opinion, there are three main risks:

• The King of Controversy: Look, Elon is a regular at making headlines. The man is currently involved in a $130 billion lawsuit with OpenAI, is under investigation by regulators in the European Union and India, and has an often...complicated relationship with President Trump.

• Execution Risk: xAI consumes approximately $1 billion per month – a huge bill. Elon is also just one person, running (at least) five other companies at the same time.

• The Law of Scaling: xAI places all its bets on the formula of "more computing power = better model", but if a better model training architecture is discovered, this formula may be overturned. (Andrej Karpathy has stated many times that he believes that large language models are not the final form.)

That’s all I’m thinking about. In the final analysis, I think the current doubts about xAI's breakthrough of the boundaries of intelligence may have overlooked the fact that it is reshaping the rules of competition at an alarming rate. What this company brings together is not only algorithms and computing power, but also a majestic ambition to break the constraints of the physical world.

Perhaps we should all temporarily put aside our prejudices and watch the ongoing AI revolution driven by infrastructure reconstruction. Because real disruption often starts in undervalued corners.