Musk and Anthropic shake hands and make peace! Under his flag, SpaceX opens the door to computing power for Anthropic.

On Wednesday, Anthropic announced a computational partnership with SpaceX, a commercial space giant owned by Musk. Anthropic can now utilize all the computational power capacity of SpaceX's Memphis Colossus 1 data center. According to the agreement, Anthropic will gain access to over 300 megawatts of computational power. Anthropic also expressed interest in collaborating with SpaceX to develop terawatt-level computational power in space. This partnership is expected to directly increase the computational power capacity for their paid subscribers Claude Pro and Claude Max. It is worth mentioning that before this agreement was reached, Musk, after merging SpaceX with his AI startup company xAI earlier this year, had criticized Anthropic for its conflicts with the US government on multiple occasions. Musk previously stated that Anthropic was destined to become the opposite of its name, "misanthropic," and questioned if there could be a more hypocritical company than Anthropic. However, on Wednesday, Musk changed his stance. He posted on the X platform that he had spent a lot of time over the past week engaging with the senior leadership of Anthropic and was "impressed." Musk stated, "Everyone I interacted with was very competent and very focused on doing the right thing. No one triggered my 'evil detector.' As long as they continue to maintain critical self-examination, Claude could do well." Musk also stated in another post on X on Wednesday that xAI would no longer exist as an independent company and would be renamed SpaceXAI. In order to compete in the thriving AI market against OpenAI, Alphabet Inc. Class C, and Anthropic, xAI has been expanding its data infrastructure, focusing primarily on the areas around Memphis. Colossus 1 is its largest facility to date. During the construction and operation of Colossus, xAI and its subsidiary MZX Tech LLC installed and operated dozens of natural gas-burning turbine generators to power the facility, claiming that federal permits were unnecessary as the equipment was only for temporary use. However, these turbine generators exacerbated air pollution issues in the Memphis area and sparked ongoing protests against xAI's local projects. Since the launch of Claude Code last year, there has been explosive growth in demand for the AI models under Anthropic. Last month, Anthropic stated that the demand for Claude had led to the "inevitable strain on infrastructure," affecting the "stability and performance" of user experiences, especially during peak times. Amidst the explosive demand and tight computational power, Anthropic adjusted the pricing model for its enterprise-level product, Claude Enterprise, last month. Under the new model, enterprise customers no longer pay a fixed subscription fee but instead pay an additional fee based on the actual computational power consumed on top of a base fee of $20 per user per month. Under the old model, enterprise customers had to pay up to $200 per authorized user per month for a certain amount of discounted Token usage. Anthropic's official explanation for this pricing adjustment was that under the old model, some heavy users frequently reached the usage limit, causing work interruptions, while others paid for an unused quota. A company spokesperson stated that the new pricing "better reflects how customers actually use Claude, as workloads move from fixed-seat productivity tools to intelligent applications." On Wednesday, Anthropic CEO Dario Amodei stated that the company had initially planned for a 10x growth but had seen an 80x year-on-year growth in revenue and usage in the first quarter, explaining why the company struggled to meet the demand for computational power. Amodei stated on Wednesday at a developer conference held by Anthropic in San Francisco, "This is the reason we are facing difficulties in computational power." He added that the company was "working to provide more" computational power capacity as quickly as possible and would "provide this computational power to users as quickly as possible." The partnership agreement between Anthropic and SpaceX is the latest effort by the company to seek computational resources to meet demand. According to recent reports, Anthropic has pledged to pay approximately $200 billion to Alphabet Inc. Class C over the next five years for cloud computational expenses related to training and deploying Claude. This investment scale reflects multiple agreements previously reached between Anthropic and Alphabet Inc. Class C. In October last year, Anthropic announced an expanded usage of Alphabet Inc. Class C's cloud TPU chips, with Alphabet Inc. Class C providing over 1 gigawatt of computational power to Anthropic, which will be put into operation this year. Last month, Anthropic reached a new cooperation agreement with Alphabet Inc. Class C and Broadcom Inc. (AVGO.US) to acquire approximately 3.5 gigawatts of TPU (tensor processing unit) computational power resources, which are expected to be rolled out starting in 2027. Also last month, Amazon.com, Inc. (AMZN.US) announced a further deepening of strategic cooperation with Anthropic. Anthropic will lock in up to 5 gigawatts of computational power from Amazon.com, Inc.'s current and next-generation Trainium chips for training and supporting its advanced AI models. Explosive growth in annualized revenue! Anthropic's valuation targets $900 billion In addition to the computational resources not keeping up with demand, the high demand for Claude directly reflects on Anthropic's revenue. Anthropic announced last month that the company's annualized revenue (ARR) had exceeded $30 billion, a significant increase from the $9 billion at the end of 2025. Meanwhile, a report by the third-party research firm Semi-Analysis in early May indicated that Anthropic's annualized revenue had risen to approximately $44 billion. This growth rate far surpasses that of OpenAI during the same period. From a revenue structure perspective, enterprise customers contribute over 80% of the revenue. Eight of the Fortune 10 companies are stable customers of Anthropic, and the number of customers paying over a million dollars a year has doubled. Among them, the annualized revenue of the standalone product Claude Code is approaching $2.5 billion, growing over five times from the mid-2025 figure of $400 million; enterprises utilizing Claude Code contribute to over half of the revenue. This indicates that Claude Code is no longer just a model function for writing code but has become an enterprise software layer that demands serious consideration in budgeting, permissions, audits, and organizational management. For stakeholders, the most significant change in this type of business is that the demand will become more sustained, relying more on stable delivery rather than sudden spikes in demand when the model is released. Data on enterprise spending from early 2026 shows that Anthropic has claimed a 73% share in businesses purchasing AI tools for the first time, surpassing OpenAI in market share. Anthropic's advantage in the enterprise market is directly related to its product positioning that prioritizes security. Anthropic is one of the leading large model manufacturers globally focusing on AI safety and alignment research, with its models demonstrating significantly better compliance and accuracy compared to similar products, rapidly penetrating industries with strict regulations such as finance, law, and healthcare. Meanwhile, sources reveal that Anthropic has initiated a new round of financing, requesting investors to submit allocation plans quickly. It is reported that this round of financing is expected to be around $50 billion and is set to be completed in mid-May. Anthropic's target valuation is approximately $900 billion, but considering the surge in demand for the company's shares, the final valuation is likely to exceed this figure. The growth trajectory of Anthropic's valuation is equally remarkable. It was valued at $183 billion in the F round in September 2025, $380 billion in the G round in February 2026, marking over a 107% increase. Now, the target valuation is set at $900 billion. If the final terms proceed as planned, the valuation will more than double from February, matching or even surpassing its main competitor, OpenAI, which completed a record-breaking $122 billion financing in March 2026 with a post-investment valuation of $852 billion.