Meta forecasted it would make $1.4T in revenue from generative AI by 2035
Meta recently set ambitious expectations for its generative AI ventures. Last year, the company predicted that its new AI products could generate between $2 billion and $3 billion in revenue by 2025 and anywhere from $460 billion to $1.4 trillion by 2035. These forecasts were revealed in court documents unsealed on Wednesday, which you can read in full here.
Understanding Meta’s Generative AI Strategy
While the documents do not clearly define what is meant by a “generative AI product,” it is evident that Meta is leveraging multiple AI initiatives to bolster its revenue streams. The tech giant has already engaged in revenue-sharing agreements with partners hosting its open Llama model collection. In addition, it recently launched an API aimed at allowing companies to customize and evaluate these models, expanding the ways in which its technology can be used commercially.
CEO Mark Zuckerberg also mentioned during a recent earnings call that Meta’s AI assistant could eventually incorporate advertisements and offer premium subscription features. Such steps are part of a broader effort to integrate AI deeply into daily technology use, further driving revenue growth.
Read also :
Pinterest launches new tools to fight AI slop
Investing Heavily in AI Development
Meta is not only betting on future revenues; it is also investing heavily to make those revenues a reality. Court filings reveal that in 2024, the company allocated over $900 million to its generative AI projects. This year, that investment may even surpass the $1 billion mark. These numbers do not include the substantial infrastructure costs needed to run and train AI models—costs that will be partially offset by plans to spend $60 billion to $80 billion on new data centers in 2025.
Key Investment Areas
- Expanding and improving the Llama model lineup
- Launching and enhancing AI customization APIs
- Building cutting-edge data centers to support AI operations
- Exploring new monetization avenues such as advertising and subscriptions
Challenges and Licensing Concerns
Meta’s aggressive push into generative AI does come with its share of challenges. One significant complication arises from issues related to copyrighted materials. Documents indicate that Meta once contemplated spending around $200 million to secure training data for its Llama models—about half of that sum earmarked solely for books. Instead of pursuing costly licensing agreements, the company chose alternative methods, a decision that has sparked legal disputes over unauthorized training practices.
Read also :
JetBrains releases Mellum, an open AI coding model
Meta’s Official Stand
A company spokesperson stated, “Meta has developed transformational open AI models that are powering incredible innovation, productivity, and creativity for individuals and companies. Fair use of copyrighted materials is vital to this. We disagree with the assertions made against us, and the full record tells a different story. We will continue to vigorously defend ourselves and protect the development of generative AI for the benefit of all.”
Conclusion
Meta’s forecast of generating up to $1.4 trillion from its generative AI products by 2035 highlights its commitment to innovation and expansion in the AI arena. With significant investments in AI development, partnerships, and advanced data infrastructure, Meta is positioning itself as a major player in the future of technology. As the company navigates licensing hurdles and legal challenges, its focus remains clear: driving breakthrough advancements while reinforcing its competitive stance in the tech industry.
Read also :
Startups launch products to catch people using AI cheating app Cluely
