AI is now part of almost every ERP conversation. That does not mean every organisation needs “AI specialists” in the way job adverts often suggest. In ERP environments, the limiting factors are rarely model complexity. They are data quality, governance, integration, process design, and adoption.
Most AI-related activity inside ERP landscapes in 2026 falls into three categories:
- Automation: reducing manual work in approvals, exception handling, document processing, reconciliations
- Decision support: forecasting, anomaly detection, prioritisation and insight generation
- User experience: search, copilots, guided workflows, knowledge support
The practical question is not “are you using AI?” It is “what work is being removed or improved, and what constraints must be respected?”
In ERP contexts, those constraints include auditability, access control, segregation of duties, data lineage, and risk management. A solution that cannot be explained, controlled or governed is difficult to operationalise—regardless of how impressive the demo looks.
This is why the most effective ERP+AI hires we see are not always those with “AI” in their job title. Strong profiles tend to combine process understanding with data fluency and governance awareness. They can work with the realities of ERP data, recognise where data quality limits outcomes, and design solutions that fit compliance requirements.
There is also a growing gap between CV language and delivery reality in this area. Not necessarily because candidates are dishonest—often because the market has rewarded broad labels. “Implemented AI” can mean anything from a small proof-of-concept chatbot to a genuinely embedded workflow that changed daily operations. Hiring teams get better outcomes when they assess candidates through practical delivery questions rather than tool lists.
Questions that usually separate practical capability from branding include:
- What business process was improved, and how was impact measured?
- What data constraints limited what could be achieved?
- How were access controls and audit requirements handled?
- What happened when adoption was low or users resisted the change?
- Which parts were deliberately not automated, and why?
Strong candidates respond with trade-offs and constraints, not only with features.
From a hiring strategy standpoint, most organisations benefit from building a small, credible capability set rather than hiring a large “AI team” within ERP:
- a business-facing lead who can identify use cases and drive adoption
- a technical/data lead who can manage architecture, integration and governance
- specialist support when needed for engineering, modelling or platform-specific work
This approach typically produces better outcomes than hiring for titles alone. It also aligns with how ERP environments operate: controlled change, measurable impact, and solutions that can be supported after initial delivery.
ERP + AI in 2026 will reward disciplined execution rather than hype. The organisations that hire well in this area will be those that define use cases clearly, measure outcomes honestly, and prioritise governance and adoption as delivery requirements—not optional extras.
