Enterprises Waste AI Spend Without Process Redesign, Need 2-Tier Developer Mix
Updated
Updated · InfoWorld · Jun 15
Enterprises Waste AI Spend Without Process Redesign, Need 2-Tier Developer Mix
3 articles · Updated · InfoWorld · Jun 15
Summary
Copilot rollouts and AI agents are delivering uneven returns because many enterprises bolt them onto existing software workflows instead of redesigning how work is specified, tested, reviewed and shipped.
The article argues AI's real payoff comes from changing the 'factory' itself: deleting outdated steps, rewriting processes around AI-native workflows, and judging success by better delivery rather than more code or tokens.
Senior engineers remain critical for security, compliance, architecture and technical judgment, especially as easier code generation can also accelerate technical debt and risky changes.
Junior developers, used well, can challenge inherited assumptions and explore new workflows—such as executable specs, AI-readable API contracts and agent-led routine migrations—under guardrails set by experienced colleagues.
The proposed model is small mixed teams of newer and senior developers, with leaders defining approved patterns and boundaries so experimentation can happen safely at enterprise scale.
Can youthful impatience fix corporate AI, or will it just create new problems for experienced engineers to solve?
With 80% of AI projects failing, are companies just repeating mistakes from the industrial revolution?
Unlocking AI ROI: Process Redesign, Talent Strategy, and Data Governance for Enterprise Success
Overview
By mid-2026, enterprises face growing pressure to prove that their significant AI investments deliver real business value. The landscape has shifted to demand measurable outcomes, not just adoption. However, many organizations still fail to measure the financial return on AI, leaving them without feedback to know what works. Most never even assess their AI ROI, leading to missed opportunities for improvement. Competitive pressure pushes companies to deploy advanced AI, but often without clear profitability. As a result, AI investments frequently fall short, highlighting the need for better measurement and a more disciplined, outcome-focused approach.