Meta announced a $35 billion investment in artificial intelligence (AI) for this year, signaling an aggressive push in the growing tech arms race.
This massive investment raises important questions about the future of AI development and its financial viability. Industry experts are now debating the scope and impact of this funding, investigating when these investments will yield a return on investment (ROI) and how they will impact Big Tech revenue models. Can give new shape. With potential strategies ranging from subscriptions to advertising, Meta's move could set new precedents for how technology companies can take advantage of AI advances.
“As yet, there is no end in sight to the arms race,” Madhu Sidhakar, CEO of generative AI company Aisera, told PYMNTS. “AI is clearly a top strategic focus. Think of it like the transition from on-prem to cloud or desktop to mobile. These are largely secular trends that have been going on for years. So, Microsoft, Google, Meta, Or for a megatech company like Amazon, missing out on AI is why they're pouring billions of dollars into capex.
Growing AI investment
Meta's recent earnings report revealed that instead of increasing its return on investment from AI, the company increased its spending by $5 billion to develop new AI products for consumers, developers, enterprises and hardware manufacturers. has been
The company's investment in AI and its Metaverse development arm, Reality Labs, is expected to reach between $35 billion and $40 billion by the end of the year.
CEO Mark Zuckerberg also discussed last week's release of the latest version of Meta's AI assistant, MetaAI, enhanced by recent updates to its larger language model, MetaLlama 3.
Despite the massive investment in AI for Meta and other companies, observers say the return on investment may be far from over.
“We are in the investment phase of the AI cycle,” Sudhakar said. “So it's unrealistic to expect massive ROI. Another historical example is from the early days of the Internet. Had to experiment with use cases, education, adoption, big investment in infrastructure, etc. Same with AI. That's what's happening. There's one area where there's a clear ROI, though: the infrastructure providers, especially Nvidia. But over time, the ROI will widen, as will applications.”
Costs can inhibit profitability.
While enthusiasm and investment in artificial intelligence is growing, the technology — especially generative AI — is very expensive, Sudhakar noted.
“There are GPUs, data science talent, data center construction, energy consumption, data management costs,” he said. “But the good news is that costs will start to come down, which will help make money. For example, this week, Snowflake announced its LLM. It cost just $2 million to train. And 1,000 GPUs are required.
Lars Nyman, chief marketing officer of CUDO Compute, told PYMNTS that the value of AI comes not only from the products and services sold to end users, but also from the company's internal efficiencies and savings.
“Of course, some applications, such as fraud detection, chatbots, or better ad serving algorithms quickly provide clear, direct financial returns,” he said. “But for basic research projects, ROI can be ambiguous, measured in long-term innovation benefit — measured in years, not quarters.”
According to Sudhakar, big tech companies are likely to adopt blended business models to take advantage of their AI models. Initially, those with sufficient cloud infrastructure can immediately benefit by monetizing their investment through a consumption-based model of cloud services. This often includes access to large language models (LLMs) and small language models (SLMs) through an API. Additionally, there are opportunities to cross-sell development tools and various other services within this ecosystem.
Additionally, as Sudhakar notes, AI applications can be marketed as subscriptions. For example, OpenAI has seen considerable success with its ChatGPT system, while Microsoft has had similar success with GitHub Copilot and is exploring the possibilities with its Office 365 Copilot offering. In addition to subscriptions, advertising offers a viable revenue stream, especially for companies like Google and Meta, which can leverage their extensive advertising infrastructure and AI enhancements to improve ad revenue.
“However, Google is in a tough spot,” he said. Its core search business relies heavily on getting paid for links. But if a creative AI app can provide most of the information in one response, that revenue stream will be strained.
Yigit Ihlamur, general partner at Vela Partners, told PYMNTS that once advertising is introduced to AI, revenues will skyrocket.
“For example, if you use ChatGPT to book a flight, organic and sponsored results will appear,” he said. “I imagine tech companies will further monetize their models with paid subscriptions, developer-focused paid APIs, and marketplace transactions.”