The boom in generative AI will help global IT spending grow by 7.5% this year, according to analyst firm Gartner — but some companies will be benefiting more than others.
The large language models (LLMs) that power generative AI require huge amounts of processing power, driving an increase in spending on infrastructure such as data centers.
According to Gartner, spending on data center systems increased by 4% in 2023, and is expected to climb an additional 24% this year.
“The compute power needs of GenAI are being felt across the data center, and spending in that segment reflects this ravenous demand,” said John-David Lovelock, distinguished VP analyst at Gartner.
While that's good news for companies that supply and support data centers, other sections of the IT industry are facing rising costs, too. Lovelock pointed to software, predicting an increase of 12.6% on software spending — while that's up a point from 2023, Gartner has revised that spending prediction down from previous estimates.
"Some software spending increases are attributable to GenAI, but to a software company, GenAI most closely resembles a tax," he said. "Revenue gains from the sale of GenAI add-ons or tokens flow back to their AI model provider partner."
In other words, money is flowing to software companies because of the boom in generative AI, but that money carries on flowing to AI providers.
“Generative AI is being felt across all technology segments and subsegments, but not to everyone’s benefit,” Lovelock added.
Beyond software and data center spending, Gartner also predicted that spending on devices would rebound from a 6.5% fall last year to 5.4% growth this year, while IT services would be up 7.1% and communications services climbing by 3%.
Overall, Gartner believes IT spending across the board would climb by 7.5%, versus 3.8% last year. That prediction is two percentage points lower than Gartner's estimates earlier this year, however.
"The change fatigue in CIOs that we saw at the start of the year has now abated and the contract backlogs going back to the third quarter of 2023 are being cleared," added Lovelock. "We expect to see a larger rush towards the end of the year to make up for the slow start."
Rising cost of generative AI
The Gartner report comes amid rising concerns around the cost of developing and using these technologies.
Anthropic CEO Dario Amodei has warned that the costs of training an AI model are set to skyrocket — perhaps topping $100 billion in the next ten years — and that could limit who can take part in such development, as CCS Insight analyst Ben Wood told ITPro last year.
"[The] big hyperscalers, the Googles, the Microsoft's and the Meta's of the world have the resources that they can scale that up, and they can keep funding that kind of investment, but for smaller companies it is going to be harder."
Beyond financial costs, AI development is increasingly causing environmental concerns over the energy required to train and power these systems.
Earlier this month, Google admitted its greenhouse gas emissions skyrocketed by 48% from five years ago, largely due to data centre energy consumption driven by AI.
