Research from Forrester and Orgvue indicates that around 55% of organisations that cut roles as part of AI-driven decisions now say they regret those choices, with analysts forecasting that roughly half of AI-attributed job losses may be reversed or reshaped by 2026.
These reversals are rarely announced. They tend to happen quietly through rehiring into different roles, greater use of contractors, or shifts in delivery models, rather than through explicit public admissions.
Alongside this, Gartner’s research suggests that a significant share of AI initiatives fail to translate into sustained business value, not because the technology lacks potential, but because organisations adopt AI ahead of clear processes, accountability, governance, and data maturity.
Taken together, this points to a consistent pattern. AI can deliver value, but only where there is a clear understanding of what problem it is intended to solve, how success will be measured, and what the organisation is realistically equipped to absorb.
Large enterprises have the scale, budget and internal capability to experiment, misjudge, and course-correct. Smaller and mid-market organisations generally do not. Yet many are making similar decisions under pressure, investing in tools before their needs are well defined, automating before workflows are stabilised, or assuming productivity gains without accounting for cultural and structural readiness.
In practice, AI tends to amplify existing strengths and weaknesses rather than compensate for them. When ambiguity exists in strategy, process or accountability, technology adoption often makes that more visible rather than resolving it.
Over the next 12 to 24 months, the organisations most likely to see sustainable value from AI will be those that move with intent rather than urgency. They will be clear about where AI genuinely supports performance, where human judgement remains essential, and what foundations need to be in place before investment begins.
Hype moves quickly, but unwinding poorly considered decisions is expensive. Clear, organisation-level understanding at the outset remains one of the most effective ways to reduce risk and create lasting value.
