According to Bloomberg, Google is set to inject up to $40 billion into Anthropic, an artificial intelligence firm, to bolster its escalating computing requirements. The Alphabet subsidiary has initially committed $10 billion, valuing Anthropic at $350 billion for this tranche. An additional $30 billion is earmarked for future investment, contingent upon Anthropic achieving specific performance milestones, as confirmed by Anthropic itself.
This significant investment commitment emerges shortly after Anthropic unveiled its newest model, Mythos, to a select cohort of partners earlier this month. Anthropic asserts that Mythos represents its most advanced model to date, boasting substantial applications in cybersecurity. Citing concerns over potential misuse, the company has deliberately limited wider access, collaborating instead with chosen organizations to thoroughly assess and mitigate associated risks — despite reports of the model already having been accessed by unauthorized parties. Furthermore, scaling Mythos is anticipated to incur considerable operational costs.
The competitive landscape in artificial intelligence is progressively being shaped by the availability of computational power essential for training and deploying these sophisticated systems. OpenAI, a prominent player, has proactively pursued extensive capacity, forging a complex network of multi-hundred-billion-dollar agreements with cloud providers, semiconductor manufacturers, and energy suppliers, including an enhanced partnership with chipmaker Cerebras finalized this month.
Anthropic has simultaneously been engaged in its own intense efforts to secure vital resources. Following widespread user dissatisfaction regarding usage limitations for its Claude model in recent weeks, the company has countered with a series of significant infrastructure agreements. Earlier this month, Anthropic formalized an agreement with cloud computing provider CoreWeave to acquire data center capacity. Concurrently, it also secured an additional $5 billion investment from Amazon this week, forming part of a wider strategic pact under which Anthropic is projected to allocate up to $100 billion to obtain approximately 5 gigawatts of compute capacity over an extended period.
Despite Google operating as a direct competitor to Anthropic in the domain of AI model development, it concurrently functions as a crucial infrastructure provider for the company. Anthropic's operations are substantially dependent on Google Cloud for both its semiconductor components and foundational infrastructure, particularly its access to Google's tensor processing units (TPUs). These specialized chips are meticulously engineered for AI workloads and are widely regarded as prime alternatives to Nvidia's highly sought-after processors.
Anthropic's established relationship with Google extends beyond the announcements made this week. Earlier in the month, Anthropic had already disclosed a partnership with Google and Broadcom, the chipmaker responsible for designing custom AI processors for Google. This collaboration aimed to secure several gigawatts of TPU-based computing capacity, slated to become available starting in 2027, with a subsequent Broadcom securities filing clarifying this figure to be 3.5 gigawatts.
This latest investment from Google significantly amplifies the existing arrangement, as Google Cloud will now furnish an additional 5 gigawatts of computing capacity over the next five years, with provisions for further expansion as needed.
Anthropic's valuation was recorded at $350 billion as recently as February; however, Bloomberg reports indicate that investors have since expressed strong interest in supporting the company at valuations of $800 billion or higher. The AI firm is also reportedly exploring the possibility of an initial public offering (IPO) as early as October.
The Editorial Staff at AIChief is a team of professional content writers with extensive experience in AI and marketing. Founded in 2025, AIChief has quickly grown into the largest free AI resource hub in the industry.