Morgan Stanley surveyed a group of CIOS to understand the sources of AI budget. If hyperscalers deploy $100b in capex on AI this year & legions of software vendors hawk AI solutions, the money to justify it must come from somewhere.

41% of CIOs said net new spend. 35% said existing software spend. Only 6% said it would come from professional services.

This last point raises the question about labor replacement : will AI siphon funds from labor budgets? Perhaps the survey wasn’t designed to capture this nuance.

Compared to 2023, those originally saying they wouldn’t invest in AI are predominantly repurposing existing software spend.

Perhaps the first movers created dedicated AI budgets & the second wave reallocates existing software spend.

A distinction might lie hidden in the figures : net new software spend for AI may be earmarked for AI infrastructure to develop AI products of their own. Whereas repurposing existing software spend may be for AI applications, buying off-the-shelf AI solutions.

Both of these changes in spend will require some substantiation of return-on-investment.

Net new budget boots the AI industry if the productivity gains are defensible. Repurposing of existing spend implies greater competition amongst a larger set of competitors.

These net flows in this tug-of-war determine the potential growth rates of the next wave of software startups.