New: the shifts that come back across the latest research. See what to focus on

Horizon: now

AI now does work instead of answering questions

AI is moving from answering questions to doing work, making costs, oversight, and data control critical challenges for companies.

In the report's words: “AI moves from answering questions to executing work, making cost, oversight, and reliable data production bottlenecks.”p6–7, p17

The number that matters

99%p28

organizations in a survey that lost money from AI risks

Where to focus

  1. 1Lower costs by avoiding expensive cloud models for routine tasksp6
  2. 2Define output reviews and ownership rules to scale AI projectsp17
  3. 3Track the reasoning behind business decisions to automate them safelyp12

The details

Who it affects

CTOs, AI delivery leads, and operations managers at large enterprises.p6, p9, p17

What holds them back

  • Enterprise CTOs struggle to run AI affordably because routine tasks use expensive cloud models with network-dependent response times.p6
  • Enterprise AI delivery leads struggle to scale pilots because output review, failure ownership, and repeatable delivery remain undefined.p17
  • Enterprise operations managers struggle to automate sound decisions because systems record outcomes without the reasoning behind them.p12

The numbers

  • 99% of organizations in a cited survey incurred financial losses from AI risks.p28
  • US$4.4 million average loss per affected organization in that survey.p28

How buyers search for it

  • reduce AI API costs
  • run AI on company hardware
  • move AI pilot to production
  • AI output checking tools
  • capture employee knowledge for AI
  • track why business decisions were made