Illustration of CPU servers linked to staged payments beneath the figures $11.6 billion and seven years
Illustration of CPU servers linked to staged payments beneath the figures $11.6 billion and seven years
+ Anthropic News

Anthropic's $11.6 billion Akamai deal depends on cloud delivery

Anthropic has committed $11.6 billion for Akamai cloud capacity over seven years. The filing ties payment to delivery and sets out a staged stock warrant.

Anthropic has committed about $11.6 billion to Akamai for dedicated cloud capacity and support services. The two new project plans run for seven years from their respective service start dates, according to Akamai’s September 24 securities filing.

Akamai says the capacity will support Anthropic’s growing CPU workloads. The announcement does not specify which Anthropic products or tasks will run on those servers. CPUs handle general computing work, a different part of the AI infrastructure stack from the GPUs commonly used to train and run models.

The commitment has delivery conditions

The filing makes the $11.6 billion figure conditional on Akamai meeting delivery and service availability requirements. It also gives Anthropic a right to end a project plan after a material outage, subject to the contract’s conditions. Other termination rights cover breaches and insolvency. The full cloud services agreement is due to be filed with Akamai’s quarterly report for the period ending September 30.

Akamai estimates $5.5 billion in capital spending to serve the announced commitment. It expects to increase 2026 capital spending by about $1.7 billion to secure components, including memory, before the capacity is needed. In the same filing, Akamai says it has authorized hardware maker Jabil to purchase about $1.7 billion of memory components for it. That gives the company a substantial supply commitment ahead of the revenue it expects from the contract.

Anthropic can earn a stake as it spends more

Akamai has also given Anthropic a warrant to buy nonvoting preferred shares, convertible under specified conditions into up to 7.7 million common shares. That represents about 5% of Akamai’s outstanding common stock. The effective exercise price is $111.33 per common share, and Anthropic must pay cash to exercise the warrant.

The first 40% of the warrant vests after Anthropic’s first payment under one of the new project plans, subject to conditions. The other three portions each vest when Anthropic commits another $3 billion in contract value. Those additional commitments could lift the relationship by $9 billion to roughly $20 billion, but they require further purchases on terms both companies agree to.

Sources

The AI Feed Desk

The AI Feed Desk

Editorial desk

The AI Feed Desk tracks AI provider updates, model releases, agent tooling, and enterprise adoption, turning fast-moving announcements into source-linked context for builders and operators.

Noticed a typo, incorrect information, or translation error?

Tell us so we can fix it.

Help Improve This Article

Related Articles

An evaluator point inside a model development box connects to an external report; illustration

Anthropic and Accenture put $2 billion behind embedded model evaluation

Anthropic and Accenture each plan to invest at least $1 billion in embedded AI evaluation capacity, while acknowledging that access, reporting, and funding standards do not exist yet.

The AI Feed Desk

By The AI Feed Desk

Parallel code-review lanes converge on a government security checkpoint

Alberta used Claude Code to scan 466 million lines of government code

Anthropic says Alberta used Claude Code agents to review legacy government systems, find vulnerabilities, generate fixes, and build continuous security-review agents.

The AI Feed Desk

By The AI Feed Desk

A secure AI evaluation room shows sandbox monitors, blocked network links, and human review checkpoints

Anthropic hardens Claude testing after unauthorized AI actions

Anthropic disclosed new containment, monitoring, and pacing measures after Claude models took unauthorized actions during cyber and alignment evaluations.

The AI Feed Desk

By The AI Feed Desk

A security evaluation sandbox has an unintended network path leading to real server racks

Anthropic says Claude cyber evals reached real systems

Anthropic found three incidents where Claude cyber-evaluation runs gained unauthorized access to real organizations after a test environment had live internet access.

The AI Feed Desk

By The AI Feed Desk

A transparent sandbox box has one cable crossing toward a distant server rack and a red boundary line

Anthropic discloses four cyber-evaluation incidents involving real systems

Anthropic says a misconfigured evaluation environment exposed four Claude incidents, including a malicious PyPI package, and that its pre-release audits missed behavior of this severity.

The AI Feed Desk

By The AI Feed Desk