Enterprise AI plans are subject to steep churn. That could become everyone’s problem.
A new report from a Pacific Northwest venture capital firm offers a lot of reasons to be optimistic about the AI sector—and a few that may give boosters some pause.
The investor, Seattle-based Madrona—which has backed Snowflake, Amazon and Rover—surveyed 150 senior enterprise decision-makers about their tech spend for its new report on enterprise AI software, and the big-picture takeaway for AI was promising. Three quarters of respondents told Madrona they’re planning on expanding their AI budgets over the next year; almost half have already set aside dedicated AI budgets.
In other words, IT departments are eager to put machine learning technology to work across corporate America. (Corporate tech spend is set to hit $4.25 trillion this year, thanks largely to the AI boom, TechCrunch recently reported.)
For all the money flowing into this sector,
But for all the money flowing into this sector, the landscape for AI vendors themselves isn’t very stable.
The daily digest for entrepreneurs and business leaders
An Inc.com Featured Presentation
In part, that’s a question of follow-through. Madrona’s survey found that at 83 percent of companies, fewer than half of AI program pilots move on to full production; only 1 percent of firms are seeing a conversion rate over three-quarters.
“It’s not a product shortcoming,” Madrona notes. “‘Didn’t work as promised’ is only the sixth most common reason a pilot fails! What kills pilots is the startup’s ability to navigate an enterprise environment. Integration complexity ranks at the top, then security and compliance.”
Perhaps an even bigger cause for concern, though, is a stat about AI user churn buried deep in the Madrona report: that 77 percent of enterprises are re-evaluating their AI vendors every six months, if not on a rolling basis.
Source: www.inc.com
