Akamai’s AI cloud gambit: stock soars, margins squeezed

By mid-2026, Akamai demonstrated early signs of a strategic pivot towards AI-driven cloud infrastructure, with Q2 revenues reaching $1.1 billion—a 5.4% year-over-year increase—primarily fueled by robust demand for AI workloads and multi-year enterprise contracts. Notably, the company secured over $2.8 billion in multi-year cloud infrastructure commitments, including a landmark $600 million, four-year deal with a U.S. technology firm focused on robotics infrastructure, underscoring management’s confidence in long-term, predictable revenue streams.

However, this aggressive expansion came with margin pressures as Akamai’s operating margin halved from 14.5% to 7.3% year-over-year, reflecting substantial upfront capital expenditures and increased headcount to scale GPU capacity and data center infrastructure. CEO F. Thomson Leighton and CFO Edward McGowan acknowledged these investments as necessary growing pains, emphasizing that all existing GPU inventory was sold out and that the company planned up to $500 million in additional capital spending to meet surging AI cloud demand.

Looking ahead, Akamai’s management expressed cautious optimism for 2027, projecting accelerated growth in the Cloud Infrastructure Services segment, which already grew 39% year-over-year in Q2 2026. With over $2.8 billion in multi-year contract commitments providing strong revenue visibility, CFO McGowan highlighted the importance of take-or-pay structures in ensuring robust cash flows, while full-year 2026 guidance anticipates revenue between $4.45 billion and $4.53 billion and adjusted EPS of $6.40 to $7.05, contingent on disciplined execution amid ongoing investments.

Akamai's stock has experienced significant volatility amid its aggressive AI cloud infrastructure expansion, with sharp swings reflecting investor concerns over profit pressures and margin compression. Despite a 5% year-over-year revenue increase to approximately $1.1 billion in Q2 2026, net income and EPS fell by about 23%, driven largely by heavy capital expenditures that surged to 32% of revenue and are expected to reach 40% for the full year. This intense spending, coupled with a 6 to 9 month lag before large AI contracts translate into recognized revenue, has fueled skepticism about near-term earnings quality and led to a nearly 7% single-session stock drop in early August.

Investor sentiment has oscillated sharply in response to major AI partnership announcements, illustrating the market's balancing act between long-term growth potential and short-term financial strain. For instance, Akamai's shares surged 13% on August 7, 2026, following a $600 million, four-year AI robotics and cloud infrastructure deal with a leading U.S. tech firm, despite the company lowering its adjusted EPS outlook due to upfront hardware and colocation costs. Similarly, the announcement of a $1.8 billion, seven-year cloud infrastructure agreement with AI company Anthropic sparked a 42% weekly rally and multiple brokerage upgrades, underscoring how transformative AI partnerships can temporarily override concerns about margin pressure and softer quarterly forecasts.

Despite these headline-grabbing deals and a 70% year-to-date stock gain marking Akamai's strongest performance since 2006, the company's stock has underperformed broader industry growth, rallying 44.3% over the past year compared to the sector's 57.6%. This divergence highlights persistent investor caution as earnings estimates for 2026 have declined while 2027 projections have improved, reflecting uncertainty about the near-term impact of AI investments versus anticipated longer-term benefits. Akamai's stable stock price around $123 and a market cap near $17.7 billion as of mid-August 2026 further illustrate a market still weighing the trade-offs between immediate profit pressures and future AI-driven growth.

In August 2026, Akamai solidified its position in the AI cloud infrastructure arena by securing a landmark $600 million, four-year partnership with a major U.S. tech firm, catalyzing a 13% surge in its stock price. This deal not only reflects Akamai's aggressive expansion of its GPU and AI capabilities but also underscores its commitment to scaling data center capacity and hardware orders to meet the escalating demand driven by AI automation. CEO F. Thomson Leighton highlighted this momentum, positioning the company for sustained growth in cloud and security services amid the AI boom.

Just weeks later, Akamai announced an even more substantial $1.8 billion, seven-year cloud infrastructure agreement with AI pioneer Anthropic, propelling its stock to a 42% gain and marking a major leap in GPU capacity to support intensive AI workloads. Despite near-term margin compression caused by upfront hardware and colocation expenses, CFO Edward McGowan emphasized that such large contracts typically take six to nine months to translate into recognized revenue, signaling robust long-term revenue growth. This strategic deal, coupled with strong first-quarter revenue and elevated component costs, has prompted multiple brokerages to upgrade their price targets and ratings, reinforcing investor confidence in Akamai’s AI-driven trajectory.

Akamai is aggressively expanding its distributed cloud infrastructure across the Asia-Pacific region by deploying thousands of NVIDIA Blackwell GPUs and pioneering the Inference Cloud, the first global implementation of NVIDIA AI Grid. This strategic move enables intelligent orchestration of AI inference workloads across its edge network, addressing the surging demand for low-latency, GPU-accelerated processing essential for production AI applications such as fraud detection, personalized banking, and media processing. Jay Jenkins, Akamai’s CTO of Cloud Computing, emphasizes that inference has become the most compute-intensive phase of AI, requiring real-time reasoning at planetary scale, which their distributed architecture is uniquely positioned to deliver.

To navigate the complex and varied infrastructure landscape of the Asia-Pacific markets, Akamai has designed its distributed cloud approach to balance centralized AI model training with regional cloud resources and edge sites. This layered infrastructure strategy not only respects data residency and sovereignty requirements but also optimizes latency by bringing AI inference closer to users and data. By positioning its network as an infrastructure layer spanning centralized facilities, regional clouds, and edge locations, Akamai effectively meets the diverse cloud maturity and connectivity needs across APAC, enabling enterprises to transition AI workloads from experimental phases to robust, production-grade deployments.