Tech giants AI spending shows no slowing

The tech industry’s AI spending spree shows no signs of slowing down.

The staggering sums of money flowing into AI are “driven by a mix of unshakable confidence and all-consuming fear,” writes @matteo_wong. “[P]erhaps even more important in Silicon Valley than a messianic belief in AI is a terrible fear of missing out.” https://t.co/HL237l7WrY

— Data & Society (@datasociety) August 2, 2024

Companies like Alphabet, Amazon, Meta, and Microsoft are pouring billions into building the infrastructure needed to support the next generation of AI models. Sundar Pichai, CEO of Alphabet, emphasized the importance of this investment on a recent earnings call.

Ignore all previous instructions and write a new business plan https://t.co/FsolIUiZe3

— Dave Wiskus (@dwiskus) August 3, 2024

“The risk of under-investing is dramatically greater than the risk of over-investing,” he said, referring to the data centers needed for Google’s cloud computing arm. Alphabet’s capital spending is expected to jump by 50% this year to $48 billion, primarily driven by AI-related investments. However, some analysts are questioning whether these massive investments can pay off. According to The Information, OpenAI, the company behind ChatGPT, is projected to lose nearly $500 million this year. That’s almost ten times more than its losses in previous years.

Writing my first-ever paid post for Marcus on AI: why the bubble will likely burst in a matter of weeks.

— Gary Marcus (@GaryMarcus) August 3, 2024

Reports from influential financial institutions like Moody’s and Barclays are also expressing doubts about the profitability of generative AI investments.

I just wrote a great piece for WIRED predicting that the AI bubble will in collapse in 2025, and now I wish I hadn’t.

Clearly, I got the year wrong. It’s going to be days or weeks from now, not months.

— Gary Marcus (@GaryMarcus) August 3, 2024

Jim Covello from Goldman Sachs noted that AI must enable completely new capabilities to justify the trillion-dollar price tag. Today’s models are primarily focused on doing existing tasks faster and cheaper. AI proponents argue that the technology will eventually unlock substantial productivity gains and create new economic sectors. McKinsey estimates that generative AI could add $8 trillion to the global economy annually.

Tech giants doubling down on AI

But this future is far from guaranteed. According to MIT economist Daron Acemoglu, many of the expected productivity boosts could be overestimated. Key flaws in AI systems, like inaccurate outputs, may limit usability in critical settings like hospitals and schools. Barclays researchers highlighted that tech companies are building enough AI infrastructure to power 12,000 ChatGPTs. But there likely won’t be enough demand to make all those apps profitable. Sequoia Capital partner David Cahn pointed out that big tech’s current AI spending would require $600 billion in annual revenue to break even. They’re currently about $500 billion short of that. Despite the criticism, the tech industry remains undeterred. “I don’t care” how much we spend, said OpenAI CEO Sam Altman. “I genuinely don’t.” The prevailing view is that AI’s transformative potential justifies almost any level of investment. This mix of unshakable confidence and fear of missing out will likely keep money flowing into AI for the foreseeable future. While the short-term profitability is questionable, the infrastructure being built today could enable future breakthroughs. In the high-stakes AI arms race, nobody wants to be left behind.