The missing layers: Why trust and value decide who wins at enterprise AI

Published in Business Reporter

The most expensive line item in enterprise AI is no longer compute – it is the gap between what AI promised and what it is actually delivering. After two years of
pilots, boards have stopped asking what AI can do and started asking what it
returns.

By 2027, Gartner forecasts that 80 per cent of data and analytics leaders
will be recalibrating their AI value expectations: effectively a market-wide reset.
Here is what will separate the companies that quietly outperform from the rest:

• The setbacks are not technical: Models drift, pilots stall and users
  disengage – not because of the model, but because data is not maintained,
  decisions cannot be traced and people cannot see where outputs came from.
  These are governance problems wearing technical clothes.
• Trust has stopped being a checkbox: Regulation such as the EU AI Act
  places liability on the deployer, not the developer. Trust cannot be bolted on
  at the end; it must be built into the foundation, treating data lineage, quality
  and ownership as production-grade infrastructure.
• Value is engineered, not assumed: Most AI portfolios are loose collections
  of initiatives with no shared view of cost or return. Portfolio discipline – a clear
  business outcome, a metric, a reusable product and a review cadence – is
  what turns AI from a budget line item into a balance-sheet asset.
• The data and value layers must connect: Ownership has historically been
  split across the Chief Data Officer, Chief AI Officer and the business. The
  organisations pulling ahead treat governance as part of their operating model,
  mapping every initiative to a measurable outcome and retiring what does not
  work.
  
  A reset, not a retreat
  
  Four in five leaders re-examining their AI expectations is not a failure of AI – it is
  the conversation finally catching up with the reality of deployment. The winners
  will not have the largest model libraries; they will be the ones whose data and
  value layers connect, holding up not just to an auditor but to the board. Scale,
  ultimately, is the reward of trust expressed as infrastructure and value expressed
  as discipline.
  
  Read the full article here: Trust and value are the missing layers in enterprise AI

About the author: Nicolas Averseng

Chief Product Officer

With 25 years of experience in enterprise software, data, and analytics, Nicolas has built and brought to market several innovative software and data platforms across multiple industries. His focus has always been the same: ensuring that technology serves a clear business purpose and delivers real value.