AI is splitting companies into two groups

Enterprise AI adoption is deeply uneven — some teams run fleets of agents in full production while others barely use LLMs. McKinsey data shows 88% of companies use AI in at least one function, but only a third are scaling it. Engineering job openings are at a three-year high (up 78%), suggesting AI is increasing demand for software rather than eliminating engineers. The real divide is between teams that have integrated AI into repeatable workflows versus those still treating it as an experiment. AI automates tasks but not jobs, and the highest gains come from redesigning workflows, not just adding copilots. Governance and organizational readiness — not model capability — are now the primary constraints.