Musk and Anthropic are going to space to find electricity

Original by Odaily Planet Daily (@OdailyChina)

Author | Azuma (@azuma_eth)

Musk and Anthropic have actually joined forces!

In the early morning of May 7, Beijing time, a joint announcement by Anthropic and SpaceX instantly shocked the entire AI industry. According to the announcement, the two parties have signed a cooperation agreement, Anthropic will use the entire computing capacity of the SpaceX Colossus1 data center, which will provide Anthropic with over 300 megawatts of new computing power within a month (equivalent to over 220,000 NVIDIA GPUs), directly improving the user experience for Claude Pro and Claude Max subscribers.

This is the first time Musk's business empire has reached such a direct, formal, and large-scale cooperation with Anthropic. Over the past many years, Anthropic's core partners have been Amazon and Google. Whether in cloud infrastructure, chip supply, or model training, Anthropic has been deeply tied to the AWS and Google TPU ecosystem for a long time. Musk has not only never publicly invested in Anthropic, but has even publicly criticized its AI safety route and political leanings multiple times. Previously, there were almost no public records of infrastructure cooperation, model collaboration, or business alliances between the two parties.

When the most capable large model company and the most newsworthy world's richest man finally connect. Such a partnership naturally comes with its own buzz.

Musk's Hidden Agenda

For Musk, the timing of this cooperation is even more delicate.

Just ten days ago (April 27), Musk's lawsuit against OpenAI officially entered the trial stage in the U.S. District Court for the Northern District of California. The two sides have been trading heated arguments in court in recent days, with a strong smell of gunpowder. This lawsuit, regarded by the outside world as the "biggest case of the AI era," has almost become a full public airing of the long-standing grievances between Musk and OpenAI.

And Anthropic happens to be one of OpenAI's most core and direct competitors. Thus, this cooperation naturally gains an additional layer of intriguing significance -- the enemy of my enemy is my friend. As long as it makes OpenAI uncomfortable, Musk will do anything.

Looking deeper from the perspective of the AI competitive landscape, this cooperation also foreshadows the evolution of Musk's strategic thinking in the AI era.

On the surface, this is just a standard computing power transaction -- SpaceX provides GPU clusters, Anthropic gains more inference resources, and both sides get what they need. But this matter is clearly not that simple.

Because what Musk is doing around AI now is far beyond just "building models." In the past two years, since Musk personally started xAI (now renamed SpaceXAI), the outside world has mostly viewed large model companies like OpenAI and Anthropic as Musk's potential competitors. But with the completion and commissioning of the supercomputing cluster Colossus, its data center capacity has begun to overflow, and Musk's role has quietly changed.

Musk is increasingly becoming the "arms dealer" of the AI world -- anyone who lacks computing power can come to him, even if they were once a potential competitor.

Anthropic, Caught in the 'Degradation' Storm, Finally Meets Its Savior

For Anthropic, the importance of this cooperation might be even higher than the outside world imagines.

Over the past few months, Claude's reputation has been undergoing subtle changes. On one hand, Claude Opus 4.7 and the mysterious Mythos are still regarded as the best models on the market; but on the other hand, voices about Claude's "performance degradation" have become increasingly frequent in the community.

This sentiment is particularly strong among heavy developer groups. Some have found that Claude's reasoning ability shows a "cliff-like decline" when handling long code and complex engineering tasks; some research reports indicate that the "thinking budget" or response length of certain Claude models has been significantly cut; more retail users report that Claude's hallucinations are getting worse, making it easier for the model to "speak nonsense with a straight face" when processing complex information.

After the "performance degradation" storm escalated, Anthropic officially released a technical review report, acknowledging that between March and April, due to product-level adjustments and bugs, Claude's models experienced "performance regression" on complex tasks.

However, this reason did not convince the market. Public opinion still widely believes that high inference costs and computing power shortages are the main reasons for the performance fluctuations of large model products like Claude in actual applications.

The stronger the model's capability, the higher the inference cost; the larger the user base, the more terrifying the GPU consumption. All AI companies face an unavoidable business contradiction -- users want the model to always run at full capacity, but companies must control costs. Thus, mechanisms like dynamic throttling, inference budget adjustments, response length control, and priority scheduling almost inevitably appear, and the user's final perception is that "the model has degraded."

This is precisely why this computing power cooperation with SpaceX is so critical for Anthropic.

Anthropic stated in the announcement that with the cooperation, the experience for Claude's core users will be directly improved:

• First, Anthropic will double the five-hour usage limit for Claude Code available to Pro, Max, Team, and seat-based Enterprise plan users.
• Second, Anthropic will lift the throttling measures for Claude Code on Pro and Max accounts during peak hours.
• Third, Anthropic will significantly increase the API rate limits for the Claude Opus model.

AI model evaluation expert Alex Finn commented on this cooperation, stating that Anthropic had been somewhat losing steam over the past few months, with reduced quotas and dumber models... and then Musk came to the rescue, giving Anthropic access to the world's largest supercomputing cluster. Anthropic's computing power crisis has always been the company's Achilles' heel, causing a continuous decline in user sentiment and market perception, and Musk solved this problem with just one deal.

Alex Finn used a metaphor easier for American fans to understand: "Musk's help is like letting Anthropic draft Victor Wembanyama!"

Final Fantasy -- Seeking Power in Space

In the cooperation announcement, there was a small detail overlooked by many -- "The parties also have an interest in collaboratively developing multi-gigawatt-scale orbital AI computing power." In plain English, Musk and Anthropic want to move AI data centers into space. This sounds science fiction, but the issue it reflects is very real.

Aakash Gupta, one of the most influential analysts in the AI circle, explained: "Earth's power, land, and cooling capacity can no longer meet demand fast enough."

Anthropic has already secured approximately 15 gigawatts of computing power, equivalent to the electricity consumption of 11 million households, but it's still not enough... Nvidia can produce chips, and Anthropic has plenty of money, but what truly cannot be built on time is power, land, and cooling capacity -- and the growth rate of model demand has far exceeded the construction speed of these infrastructures.

The frontier of the computing power race is now beginning to leave Earth. And only one company in the world truly has the capability to launch gigawatt-scale solar arrays into orbit on a large scale -- SpaceX.

If this interstellar travel-like story eventually becomes reality, the only person who can do it might be Musk.