Artificial Intelligence (AI) has been hailed as a game-changer for businesses and governments, with the potential to address labor shortages, increase wage growth, and boost tax revenues. However, the widespread adoption of AI technologies has been relatively slow and limited […]
Artificial Intelligence (AI) has been hailed as a game-changer for businesses and governments, with the potential to address labor shortages, increase wage growth, and boost tax revenues. However, the widespread adoption of AI technologies has been relatively slow and limited to specific sectors of the market.
In this insightful analysis, we delve into the latest trends in AI adoption, the use of generative AI, different implementation models, and AI rankings based on consumption and market presence. Our aim is to provide a fresh perspective on the impact of AI on costs and technology adoption.
AI Initiatives Tap into Other Budgetary Funds
Let us first examine the overall enterprise information technology market and the influence of AI costs on other sectors.
The graph above represents nearly 30 sectors tracked in the Technology Spending Intentions Survey (TSIS). It shows the Net Score or consumption momentum on the vertical axis, and sector penetration or market presence on the horizontal axis. The dotted red line at 40% indicates a high rate of cost momentum in the sector.
Notably, the line for ML/AI appears distorted. When we transitioned from an isolation economy, AI momentum slowed alongside the overall IT spending. However, observe what happened a month after the announcement of ChatGPT. ML/AI took over as the number one sector in terms of consumption momentum.
The issue lies in our research findings from “Lower for Longer,” which revealed that expectations for general enterprise technology spending have been declining from a high growth forecast of 7.5% to the current level of 2.9% over the past two years. Chief Financial Officers and CEOs are generally not allocating funds for discretionary spending on AI research. Instead, AI initiatives are diverting money from other sectors, many of which are sacrificing their budgets to fund AI experiments.
Erik Bradley from ETR provides additional context to these findings: “We have a few data points that I want to highlight. First, yes, ETR is growing the number of sectors we track. We just added FinTech, and we will continue to expand our coverage as needed. But in terms of the spending, yes, 3% might not sound too exciting, but we’re coming from 0.8% last year at one point, and 3% is huge growth from that. The second thing I want to emphasize is that the low sub-1% numbers were coming from the largest organizations in the world, Fortune 100, Fortune 500, Global 2000, all of which had the worst spending. Now, we see them moving back toward the mean. So, when we see the largest organizations in the world actually increasing their spending, I think we’re coming out of it. I think that’s a good thing. That’s point one.”
“Second point, when we talk about net scores for ML/AI sector-wise, it’s at historical levels. We’ve been tracking spending for over 12 years. And currently, ML/AI has a sector net score of 52%. That’s up from 38% just 12 months ago. In comparison, the overall average score in our survey is only 20%. ML/AI is three times higher than the survey average. Only container management is approaching the 40% tipping point. AI is truly at a rare level. It’s unique in its category.”
“And finally, your comment about AI cannibalizing budget. We conducted in-depth research recently, and it showed that 67% of those Global 2000 respondents stated that they are not using net new dollars to fund their generative AI efforts. They are actually taking it from other areas, while only 33% are funding gen AI by introducing fresh money.”
Examining AI Adoption Trends…
Now, let’s look at the latest data on AI adoption spending.
The chart above displays the findings from ETR’s October spending survey. IT decision-makers were asked whether their organizations evaluate generative AI and large language models for business use cases, and if so, which ones. You’ll notice a sharp decline from the April survey for organizations not evaluating gen AI – the gray to yellow bars shifting from 52% to today’s 26%. Additionally, there is a significant increase in business use cases from April to October.
It is important to note that the percentage of clients in the Global 2000 evaluating gen AI is much higher than the industry average, with only 14% of Global 2000 organizations stating that they do not evaluate gen AI.
…but Most Initiatives Remain Experimental
Looking at the data below regarding what is actually happening in production, we can draw two significant conclusions: 1) Most activities are still in the evaluation phase, and 2) The business use cases are primarily focused on productivity and relatively straightforward work, such as ChatGPT.
This does not mean that these business use cases are insignificant, but generally, they do not indicate a radical restructuring of processes within organizations.
Daren Brabham adds the following points to these findings: “I mean, especially in this macro-environment, organizations are going to be cautious as much as possible. They are excited and overwhelmed by the buzz around generative AI at a rate we’ve never seen before with this technology. But limited budgets and an uncertain business environment mean that organizations are still tiptoeing into areas they are most familiar with.”
So, on the previous chart, it’s incredible how quickly organizations are evaluating generative AI for business use cases. It went from six months ago, where half of organizations were not evaluating, to now being only 14%.