Revolutionary CEO Proposes Alternative to Traditional Data Centers
WHO:
Fast Company - Tom Leighton, cofounder and CEO of Akamai, a content delivery network company valued at $18 billion.
Fast Company - Meta, a technology company building a large AI infrastructure project.
WHAT:
VentureBeat - Only about one in five (21%) enterprises run AI in production at scale, while 76% are either experimenting or running only some workloads in production.
VentureBeat - 83% of enterprises that operate GPUs report utilization of 50% or less, with nearly half (49%) running at 25% or below.
VentureBeat - 64% of enterprises plan to switch or add an infrastructure provider within twelve months, and 38% within the next quarter.
VentureBeat - The single most-cited planned evaluation area for AI infrastructure over the next 12 months is AI-specialized clouds, at 45%, which is a category almost none of these enterprises use today.
Fast Company - Akamai's cloud infrastructure revenue has grown 40% year over year.
Fast Company - Akamai's architecture can reduce inference latency by as much as 2.5 times compared with traditional hyperscaler infrastructure and lower inference costs by up to 86%.
Fast Company - Akamai's published benchmarks show RTX PRO 6000 Blackwell GPUs on its cloud delivering up to 1.63 times the inference throughput of Nvidia H100 GPUs, sustaining roughly 24,000 tokens per second per server under 100 concurrent requests.
Fast Company - Meta's Hyperion campus is expected to cost more than $50 billion and run on about 5 gigawatts of power.
Fast Company - Akamai has entered into a four-year, $200 million agreement with an unnamed major U.S. technology company to deploy one of the world’s largest clusters of Nvidia RTX PRO 6000 Blackwell GPUs.
Fast Company - Akamai announced a seven-year, $1.8 billion cloud infrastructure commitment from a leading U.S. frontier AI model developer, identified as Anthropic.
Fast Company - Akamai claims its architecture can reduce inference latency by as much as 2.5 times compared with traditional hyperscaler infrastructure and lower inference costs by up to 86%.
Fast Company - Akamai's published benchmarks show RTX PRO 6000 Blackwell GPUs on its cloud delivering up to 1.63 times the inference throughput of Nvidia H100 GPUs.
WHERE:
Fast Company - The Hyperion campus is located in rural Richland Parish, Louisiana.
Fast Company - Akamai is supporting production AI deployments for an AI-powered video intelligence platform in India and a U.S.-based consumer AI company.
WHEN:
Fast Company - The agreement with Anthropic is the largest contract in Akamai’s 28-year history.
WHY:
Fast Company - Leighton argues that operational decisions, not just raw compute, will determine the performance, reliability, and economics of AI applications at scale.
Fast Company - Communities are pushing back against data centers due to their demands on power and water and their impacts on rural and suburban areas.
Fast Company - Leighton believes that the AI industry may be trying to solve too many problems with the same enormous building.
HOW:
VentureBeat - Enterprises are choosing AI infrastructure providers based on integration with the existing stack (41%) and total cost of ownership (35%), rather than on headline price, which is the deciding factor for just 8%.
Fast Company - Akamai's strategy includes AI Grid, an orchestration layer developed with Nvidia, which determines where inference workloads should run based on latency requirements, operating costs, and performance needs.
Fast Company - Leighton argues that operational decisions, not just raw compute, will determine the performance, reliability, and economics of AI applications at scale.
(UPDATED July 23, 2026 21:20 UTC)
A recent report reveals that only about one in five enterprises are running AI in production at scale, while a significant 76% are still experimenting or only utilizing some workloads. Additionally, 83% of enterprises operating GPUs report that their utilization is at 50% or less, with nearly half running at 25% or below, indicating a potential gap in efficiency. Looking ahead, 64% of enterprises plan to switch or add an infrastructure provider within the next twelve months, with AI-specialized clouds being the most-cited area for evaluation, as 45% of them express interest in this category, which is currently underutilized.
--
(UPDATED July 23, 2026 18:23 UTC)
Akamai has reported a remarkable 40% growth in its cloud infrastructure revenue year over year, highlighting the increasing demand for its services. The company's architecture significantly enhances performance, reducing inference latency by as much as 2.5 times compared to traditional hyperscaler infrastructure and cutting inference costs by up to 86%. Additionally, Akamai's benchmarks indicate that its RTX PRO 6000 Blackwell GPUs can deliver up to 1.63 times the inference throughput of Nvidia H100 GPUs, achieving around 24,000 tokens per second per server with 100 concurrent requests, underscoring the importance of operational decisions in optimizing AI applications.
--
(UPDATED July 23, 2026 09:44 UTC)
Akamai, led by cofounder and CEO Tom Leighton, has made headlines with a significant four-year, $200 million agreement to deploy one of the world’s largest clusters of Nvidia RTX PRO 6000 Blackwell GPUs, alongside a record-breaking seven-year, $1.8 billion cloud infrastructure commitment from Anthropic. Meanwhile, Meta is investing over $50 billion in its Hyperion campus located in rural Richland Parish, Louisiana, which will require about 5 gigawatts of power to support its ambitious AI infrastructure project. Leighton emphasizes that the AI industry may be attempting to tackle too many challenges with massive facilities, suggesting that operational decisions will be crucial for the performance and economics of AI applications at scale.