Silicon Valley is embarking on a comprehensive AI strategy, with Mark Zuckerberg, CEO of Meta, at the forefront. According to Zuckerberg, by 2027, billions of people will have their own personal AI agents working around the clock to achieve various goals.
An AI agent goes beyond mere chatbots; it automates tasks and operates in the background even when users are away from their devices. During Meta’s latest earnings call, Zuckerberg highlighted that these personal agents would form the backbone of the company’s future products and revenue streams.
Creating such advanced AI agents is not without significant costs. Like large language models (LLMs), developing AI agents requires substantial investment billions of dollars in research and development. This heavy spending has impacted Meta’s financial reserves, with its free cash flow experiencing a dramatic drop.
In the second quarter of 2023, Meta’s free cash flow fell to $784 million from $8.55 billion in the same period last year its lowest since 2022. The company has also raised its capital expenditure forecast for 2026 to $130 billion, up from $125 billion. These figures reflect Meta’s commitment to AI despite the financial strain.
Google is not far behind. Recently, Alphabet reported negative free cash flow of $5.9 billion in the second quarter as it continues to spend more than it earns. This trend suggests that both Meta and Google are willing to burn through their cash reserves to stay ahead in the AI race.
Meta’s AI strategy includes doubling its overall computing power to 7 gigawatts this year, followed by an expansion to 14 gigawatts next year. The company currently operates or is constructing 32 data centers. Additionally, Meta has released a new Muse family of AI models within less than a year, including Muse Spark 1.1 and Muse Image, developed by the Meta Superintelligence Labs (MSL).
Zuckerberg emphasized that messaging apps will play a crucial role in this AI landscape. WhatsApp is currently the largest platform for Meta’s AI initiatives, with other messaging surfaces expected to become increasingly important as users interact with multiple agents.
Despite these ambitious plans, Meta faces additional costs beyond AI infrastructure. The Reality Labs division, responsible for AR and VR products, lost approximately $4.6 billion in the quarter and has incurred more than $80 billion in operating losses since 2021. Moreover, the company has laid off around 8,000 employees about 10% of its workforce resulting in severance expenses.
In summary, while Meta is investing heavily to ensure it leads the AI revolution, these substantial costs come with trade-offs that could affect the company’s financial health and organizational structure.
Thinking about building an AI product?
Get in Touch