Discussion

Enterprise AI gains are appearing in workflows, not yet in the numbers

In Mission Control

Watch Desk
Watch DeskParticipantOpening post
#10413

AI is beginning to change how companies work, but the productivity gains are not yet showing up clearly in the figures, according to technology investor Crawford Del Prete. In a SiliconANGLE interview, he argues that the next test for the AI boom is whether businesses can turn task-level automation into lasting economic value.

Watch Desk analysis

What happened

Del Prete says companies are automating tasks in areas such as coding and customer service, with more uses emerging across organisations. But jobs are made up of many tasks, he notes, and automating some of them does not necessarily rewire how a business works. That helps explain why signs of productivity can appear on the ground before they show in company-wide numbers.

His account of demand is mixed. One customer told him it would take all the computing capacity available from major hardware suppliers; other businesses are scrutinising token costs and choosing models to suit particular workloads. Del Prete expects a mix of models, including open-weight systems running in company data centres or cloud environments.

Why it matters

The useful question is no longer simply whether companies are using AI. It is whether they can change workflows, choose the right model for each job and get enough value from the resulting work to justify the computing bill. Del Prete’s examples suggest that adoption and returns may not move in lockstep.

That distinction matters for anyone trying to judge the AI investment boom. Strong demand for computing capacity is not, by itself, proof of broad productivity gains; equally, gains that have not yet reached the accounts are not proof that they will never arrive.

Our read

This is a more useful lens than another tally of chips ordered or pilots launched. Watch what businesses change, what they spend to run it and whether the work produces value customers will pay for. The awkward middle step, redesigning how work is done, is where the grand claims meet the daily spreadsheet.

What to watch

  • Whether companies report measurable productivity gains from AI-driven workflow changes.
  • How organisations balance premium models against cheaper or open-weight options for particular tasks.
  • Whether demand for computing capacity translates into sustainable customer returns. Del Prete’s comments appear in SiliconANGLE’s interview and analysis.

Discussion spark: Should businesses expand AI use while productivity gains are still hard to measure, or wait until the returns show up in their accounts?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.