Big Tech’s AI Power Plays

AI Adoption

Artificial Intelligence (AI) is increasingly becoming an integral part of modern business. Its potential to automate processes, optimize operations for increased profitability and support decision-making is invaluable. But what enables AI systems to have such wide applications is their ability to learn and expand, swiftly adapting to new situations and improving performance over time.

Advances in AI are transforming the economy. Behind the wave of AI innovation lies a less visible but important trend: the role of large technology firms across the AI supply chain. AI applications are becoming pervasive, integrating into various aspects of daily life and industries, with hopes to revolutionize the way we live and work. Specifically, generative AI (GenAI) is being adopted at a rapid pace. GenAI is a type of artificial intelligence that creates new content, such as text, images, music and code, based on learned patterns from existing data. Like humans learning through experiences, experimentation, and absorbing new information, AI develops its skills by analyzing vast amounts of data.

Big Tech Power

The influence of big technology firms on AI extends far and wide. Training data is the foundation of AI, and big tech firms have access to some of the richest pools of user-generated data in the world. For example, Meta (META) has Instagram, Facebook and WhatsApp. Google (GOOG) has YouTube, Gmail, Maps, Play Store and Google Search. Amazon (AMZN) has Amazon Web Services, Amazon Prime and a vast access to third-party sellers.^[1] In addition to accessing their current data supply in house, big tech companies are actively acquiring or partnering with data-rich firms.

These firms are not only developing their own fundamental AI models, they are also integrating them into consumer-facing products. Simultaneously, they are producing AI hardware and investing billions of dollars in data centers to fuel their expansion. Meta, Microsoft (MSFT), Amazon and Google are expecting to spend a cumulative ~$325 billion on AI capital expenditures and investments in 2025, driven by their continued commitment to building out their respective AI infrastructures.^[2] This vertical integration of internal data to customer-facing AI models is allowing big tech firms to capture strong value along the AI supply chain, creating a self-reinforced data loop. By controlling computing resources, these firms can produce better AI models. In turn, the models generate more data, which can be fed back into their systems extensively at new and improved versions. As more users adopt the AI model or platform, its value increases.

Chart 1: Big Tech’s Capex AI Investment Accelerates^[3]

Google AI Integrates YouTube

We saw reports last week that emphasized our point about big tech’s ability to leverage rich pools of user data to fuel AI growth. We learned that Google was using its expansive YouTube library to train its AI models, including Gemini and the Veo 3 video and audio generator. Gemini is Google’s AI model, and Veo 3 is a specific AI video generation model that is part of the Gemini ecosystem.

The tech company turned to its catalog of 20 billion YouTube videos to train these new-age AI Tools. Given the platform’s scale, training on just 1% of the catalog would amount to 2.3 billion minutes of content, which is more than 40 times the training data used by competing AI models.^[4]

While it is unclear which of the 20 billion videos on YouTube were used for AI training, Google said it honors agreements with creators and media companies, stating that “they have always used YouTube content to make our products better, and this has not changed with the advent of AI.” This new evolution in AI learning can enhance its capabilities, transforming search and content discovery which can greatly improve Gemini and Veo 3’s understanding of human communication and interaction.

Project Rainier

In Seattle, Mount Rainier has a commanding presence, towering over the surrounding terrain. Its prominence explains why Amazon Web Services (AWS) borrowed its name for a project that creates what is expected to be the world’s most powerful computer for training AI models, Project Rainier. As part of Amazon’s $100 billion commitment to invest in AI infrastructure, they designed this massive project to usher in the next generation of AI.

AWS customer AI safety and research company Anthropic (who AMZN has an $8 billion stake in) will use this AI compute-cluster to build and deploy future versions of its leading AI model, Claude. Project Rainier is projected to provide five times more computer power compared to Anthropic’s current largest training cluster.^[5]

Project Rainer is designed as a massive set of “Ultra Servers”.  Traditionally, servers in data centers operate independently. If these servers want to share information, they must travel through external network switches which can cause delays when running at a large scale. Now enters the Ultra Server. An Ultra Server combines four Trainium2 servers, each with 16 Trainium2 chips. These servers communicate through high-speed connections called Neuron Links. These connections allow data to move much faster within the system and significantly accelerate complex calculations. When you connect these Ultra Servers and point them to the same problem, you get project Rainier-a mega “Ultra Cluster.”

Potentially, Project Rainier could deploy computational power that could catapult AI into possessing the ability to solve problems that have long been unsolvable by humans, transforming the technological landscape as we know it.

META Going All-In on AI

Recently, we have seen Meta CEO Mark Zuckerberg on a spending spree like we’ve never seen before, pushing his company to a position at the forefront of the AI boom. Recently, Meta increased their capital expenditure range from $64 billion to $72 billion, reiterating to investors that AI is a major theme right now, transforming everything the company does.^[6]

Meta’s unique open-source approach to their AI model is built around a Llama family of models. Its most recent update in April, the Llama 4 AI models, were not well received.

That recent failure may explain the ongoing pile of cash being spent to revamp Meta’s AI organization. Last week, Meta announced a $14.3 billion investment in Scale Ai (a 49% stake). This investment also brought the startup’s founder, Alexandr Wang, with it, which included a small group of his top staffers to join Meta’s new “Superintelligence Lab”. The acquisition addressed Meta’s most pressing challenge in the AI race, which is access to the specialized datasets required to train competitive large language models. Scale AI provides curated and labeled data sets to model developers for AI training. It can also provide evaluation and help to improve reasoning models, providing key services to ensure Meta can put out its best products.

According to OpenAI CEO Sam Altman, Meta has also tried to lure OpenAI employees by offering signing bonuses as high as $100 million, with even larger annual compensation packages. Clearly, Meta is making a concerted effort to stay competitive in the rapidly evolving AI landscape.

Chart 2: Meta’s AI Spending Spree^[7]

Stephanie Link’s TV Schedule:

Sources

^[1] Source: Centre of Economic Policy Research, As of May 16, 2025.

^[2] Source: CNBC, As of February 2025.

^[3] Source: Financial Times, As of February 2025.

^[4] Source: CNBC, As of June 19, 2025.

^[5] Source: About Amazon, As of June 24, 2025.

^[6] Source: CNBC, As of June 21, 2025.

^[7] Source: Yahoo Finance, As of June 16, 2025.

Disclosures

Investment Solutions is a group comprised of investment professionals registered with Hightower Advisors, LLC, an SEC registered investment adviser. Some investment professionals may also be registered with Hightower Securities, LLC, member FINRA and SIPC. Advisory services are offered through Hightower Advisors, LLC. Securities are offered through Hightower Securities, LLC. This is not an offer to buy or sell securities. No investment process is free of risk, and there is no guarantee that the investment process or the investment opportunities referenced herein will be profitable. Past performance is neither indicative nor a guarantee of future results. The investment opportunities referenced herein may not be suitable for all investors. All data or other information referenced herein is from sources believed to be reliable. Any opinions, news, research, analyses, prices, or other data or information contained in this presentation is provided as general market commentary and does not constitute investment advice. Investment Solutions and Hightower Advisors, LLC or any of its affiliates make no representations or warranties express or implied as to the accuracy or completeness of the information or for statements or errors or omissions, or results obtained from the use of this information. Investment Solutions and Hightower Advisors, LLC assume no liability for any action made or taken in reliance on or relating in any way to this information. The information is provided as of the date referenced in the document. Such data and other information are subject to change without notice. This document was created for informational purposes only; the opinions expressed herein are solely those of the author(s) and do not represent those of Hightower Advisors, LLC, or any of its affiliates.