The three pressures that will test every system built in the next two years

Regulatory shift, AI disruption, and workforce change are not future risks. They are present constraints that most current briefs do not account for.

Most system briefs are written in response to a present problem. The system that results is designed to solve that problem, in the conditions that exist today, with the technology, regulatory framework, and workforce that the organisation currently has.

The problem is that none of those conditions will remain constant. Three pressures are already reshaping the environment in which institutional systems operate. They are not future risks. They are present constraints that most current briefs do not account for.

Regulatory shift

Data sovereignty requirements are tightening across most jurisdictions. The question of where data is stored, who has access to it, under what conditions it can be processed, and what must be disclosed when a system makes a decision affecting a regulated party: these requirements are expanding in scope and specificity.

A system built today that stores data in a structure incompatible with emerging data sovereignty frameworks will require architectural redesign before the regulatory deadline. That redesign is significantly more expensive than an architecture decision made at the outset.

AI regulation is developing in parallel. Systems that use machine learning to inform decisions affecting people are subject to explainability requirements, audit trails, and bias assessment frameworks in an increasing number of jurisdictions. A system built without these considerations embedded will fail regulatory review, not immediately, but within the medium term.

Three expanding pressure zones

DATA SOVEREIGNTY

Now → 3 years

Requirements for where data is stored, processed, and governed are expanding. Systems built without considering jurisdiction will face mandatory redesign.

AI REGULATION

Emerging now

Audit trails, explainability requirements, and accountability frameworks are becoming compliance conditions in regulated industries.

WORKFORCE TRANSITION

Ongoing

Role structures are changing. Systems designed around current job descriptions will require modification as responsibilities shift.

AI disruption

Many systems built in the next two years will be built to support processes that AI is actively transforming. A document review system built today may be partially redundant in eighteen months if the underlying review process has been automated. A routing optimisation system may be superseded by a foundation model deployment before the first system has recouped its build cost.

This is not an argument against building. It is an argument for modularity: building systems where the components most likely to be disrupted can be replaced without replacing the entire architecture. It is an argument for shorter commitment cycles and more explicit technology refresh planning.

A platform built only for today's problem becomes legacy quickly.

Workforce change

The systems built in the next two years will be operated by a workforce that is changing in at least three dimensions. Skills profiles are shifting: roles that existed three years ago have changed, and roles that will exist three years from now are not yet defined. Generational transition in institutional leadership is accelerating: systems designed for one generation of users may face adoption resistance from the next.

Remote and hybrid work patterns have permanently changed how systems are used. Systems built for co-located use require different architecture than systems designed for distributed access. A system that does not account for the workforce that will operate it is a system that requires ongoing workarounds from the day it launches.

Monolithic

Built as a single integrated system.

Pressure on any layer affects every other layer.

Regulatory change, AI requirement, or role shift triggers a full system review.

Replace everything or nothing.

Modular

Built as replaceable components with clear interfaces.

Pressure on one layer is absorbed without destabilising others.

The AI layer, the compliance layer, and the data layer can each evolve independently.

Replace only what changes.

Designing for forward pressure

Designing for these pressures is not speculative. It is structural. It means building modularity into the architecture so that components can be replaced. It means embedding compliance requirements from the outset rather than retrofitting them. It means selecting technology partners with roadmaps that account for the regulatory and AI environment, not just the current feature set.

Systems designed to be resilient to these pressures will not require replacement. They will require evolution. That is a significantly smaller cost. The brief that does not account for forward pressure is a brief for a system that will begin obsolescence at launch.

←All insights

If this resonates with a challenge you are navigating, we are happy to talk through it.

Start a conversation