Three different things are being bundled into one headline
Cost-cutting unrelated to AI, genuine task automation, and a hiring freeze dressed up as transformation get reported identically as "AI-driven layoffs". Separating them requires asking what specific tasks the remaining team can no longer do, not what the press release claims.
Boards approving these decisions rarely get a breakdown of the three either — which is its own governance problem explored further in our strategy coverage of this exact pattern.
The roles actually at risk are narrower than the headlines imply
Work that is high-volume, well-specified, and has a short feedback loop is genuinely automatable today — first-draft content, basic data entry, tier-one support triage. Work requiring judgement under ambiguity, stakeholder negotiation, or accountability for an irreversible decision is not, regardless of how the tooling is marketed.
The honest workforce plan names the first category specifically and is silent — not silently optimistic — about the second.
What a defensible transition actually looks like
Redeploying staff from automated tasks into review, exception-handling, and the judgement layer the automation can't cover is a genuine transition. Simply removing the headcount and hoping the model holds the line unsupervised is a cost-cutting decision wearing an AI narrative.
If the workforce plan can't name the specific task being automated, it isn't an AI strategy — it's a cost reduction with better marketing.MNB AI Practice
Key actions
- Separate cost-cutting, automation, and hiring freezes before reporting them as one trend
- Name the specific tasks being automated, not the headcount reduction target
- Be explicit about what the remaining team is now accountable for reviewing
- Plan redeployment into exception-handling before announcing the reduction, not after