Goldman Sachs has released a report that warns tech giants are spending a lot of money on AI infrastructure. The investment bank thinks about $1 trillion will be spent in the coming years on things like data centers, chips, and upgrades to the power grid. However, it’s unclear whether this huge investment will pay off.

Jim Covello, who leads equity research at Goldman Sachs, said it’s important to ask what big problem AI will solve that is worth $1 trillion. He noted that using very expensive technology to replace low-wage jobs is the opposite of how technology has been used in the past to boost productivity and growth. The report points out that even though a lot is being spent on AI, few major AI applications are making money yet.

Investors may start to lose patience if this continues for too long without clear results. However, some analysts at Goldman Sachs see reasons for optimism.

Debating returns on AI infrastructure spending

They note that big spending on computing infrastructure has led to major technological advances in the past. The AI revolution may just be in its early stages, similar to past cycles that started with building infrastructure before moving on to platforms and applications. The growth of AI will also have a big impact on the power grid as electricity demand from data centers soars.

Experts estimate the US will need 47 gigawatts of extra power capacity by 2030 just for data centers. Some worry the grid may struggle to keep up with the rapid pace of AI development. Despite the concerns, the Goldman Sachs report ends cautiously, hopefully.

It suggests the AI investment trend could still have room to grow, either by proving its potential or by keeping investors excited about its possibilities. But for now, the huge costs and uncertain payoffs of the AI race are making some investors nervous.