Get a Good Return on Your AI Investments

2026-05-27 · Source: AI & ML – Radar · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Cloud Computing & IT Infrastructure · Depth: Intermediate, medium

Summary

A recent O'Reilly Infrastructure & Ops superstream, "Platform Engineering in the Age of AI," featured a fireside chat with Nathen Harvey from Google Cloud's DORA team, discussing findings from DORA's "ROI of AI-Assisted Software Development" report. DORA's 2025 data indicates a 10% improvement in code shipped to production with AI assistance, but also higher instability, suggesting individual productivity gains don't always translate to production efficiency. AI acts as an amplifier, exacerbating existing process friction in low-performing teams. Key challenges include a "verification tax" due to low trust (30% trust AI output little or not at all, 46% "somewhat"), potential AI-induced burnout from context switching, and growing "cognitive debt" as developers lose shared understanding in agentic workflows. DORA emphasizes that effective AI adoption depends on "how" AI is used, advocating for investment in quality internal platforms to provide guardrails and shift production complexity.

Key takeaway

For Directors of AI/ML or VPs of Engineering evaluating AI investments, recognize that AI tools alone do not guarantee production efficiency. Your teams may experience increased throughput but also higher instability and "cognitive debt" if underlying processes are weak. Prioritize investing in robust internal platforms to provide essential guardrails and scale verification efforts. Use DORA's ROI framework to model the full costs, including verification and training, ensuring your organization creates conditions for sustainable AI productivity rather than just individual gains.

Key insights

AI amplifies existing software development processes, demanding robust platforms and verification to realize true productivity gains.

Principles

• AI amplifies existing processes, good or bad.
• Verification is a critical "tax" on AI productivity.
• Investment in internal platforms is crucial.

Method

DORA's ROI framework and calculator model real costs of AI-assisted development, including learning, verification, and pipeline changes.

In practice

• Model AI's true costs with DORA's ROI calculator.
• Prioritize platform investment over feature development.
• Foster a culture of continuous learning and experimentation.

Topics

• AI-Assisted Development
• Software Delivery Performance
• DORA Metrics
• Platform Engineering
• Cognitive Debt
• ROI Calculation

Best for: CTO, Director of AI/ML, VP of Engineering/Data, AI Architect

Related on AIssential

• The ROI of AI in manufacturing: Where adoption becomes advantage — Industrial AI offers substantial ROI and operational improvements across manufacturing, sustainability, and workforce empowerment.
• Amazon's Big Spending Plans and Bitcoin's Rebound — Massive AI infrastructure investments are reshaping market valuations and driving new business models across tech and media.
• The Sequence Opinion #860: Every Company’s Last eXam: Some Reflection About Practical AI Evals — Company-specific, dynamic evaluations are now essential for production AI, forming a fourth pillar alongside compute, data, and models.
• New DORA Report Claims Strong Engineering Foundations Drive AI Return on Investment — AI amplifies existing engineering foundations, requiring strong organizational systems for positive ROI.
• The Convenience Trap — AI is increasing workload and content volume, not reducing it, due to a "verification burden" and the Jevons paradox.
• the companies actually making money with AI aren't using it the way this sub thinks they are — Real AI value stems from incremental automation of mundane tasks, not revolutionary AGI or human replacement.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI & ML – Radar.