Pinterest said it will pay Amazon Web Services (AWS) $4 billion for cloud services through 2031, marking the social media company’s largest-ever cloud infrastructure agreement and deepening a partnership that has spanned more than a decade.

The announcement reflects Pinterest’s growing focus on artificial intelligence as it seeks to enhance user engagement, improve advertising performance, and compete more effectively with digital platforms such as TikTok, Instagram, and Facebook.

Shares of Pinterest rose nearly 5% following the news, while Amazon gained about 1.7 percent, underscoring investor optimism about the expanded collaboration and its potential impact on both companies.

Under the agreement, AWS will provide Pinterest with access to its custom-designed chip processors, including Graviton and Trainium. These technologies are intended to help Pinterest scale its AI workloads more efficiently while managing infrastructure costs.

“This expanded commitment with AWS gives us the compute flexibility, hardware optionality, and infrastructure efficiency to accelerate our AI vision,” Pinterest Chief Technology Officer Matt Madrigal says.

Also Read: Decoding India's E-Commerce Shift: AI, Speed & Real-Time Retail

Pinterest has increasingly invested in AI-powered capabilities across its platform. The company has rolled out upgrades to its Performance+ advertising suite, which uses automation and machine learning to improve ad targeting and campaign effectiveness. Such investments are aimed at driving growth in an increasingly competitive social media and digital advertising market.

The company said it has worked with AWS since 2010 to improve the reliability, scalability, and performance of its core services. The latest agreement builds on that long-standing relationship and aligns with Pinterest’s broader strategy of expanding its AI infrastructure.

Also Read: India's AI-Energy Race: Can Power Drive the Digital Leap?

Last month, Pinterest forecast second-quarter revenue above Wall Street estimates, signaling continued momentum in its advertising business. As part of the new deal, the company plans to diversify its accelerated computing resources using Amazon’s custom silicon, seeking better price-performance efficiency for AI applications.

Also Read: WebEngage and India Shelter to Bolster Digital Conversion Funnel

This includes leveraging AWS Trainium chips to support large language models and vision-language models that power features such as personalized visual search, recommendation systems, content understanding, and AI-assisted discovery experiences across the Pinterest platform for users globally.