Slop, productivity, and why the AI-fueled world is going nowhere mighty fast

2025-06-07 · Source: Marcus on AI · Field: Business & Management — Artificial Intelligence & Machine Learning, Economic Analysis & Policy, Corporate Strategy & Leadership · Depth: Fundamental Awareness, short

Summary

Analysis of recent data from the FT, MIT, McKinsey, and Bain indicates that while AI significantly increases content output across domains like mobile apps, books, music, and scientific papers, this nominal productivity has not translated into substantial Return on Investment or material GDP growth. This surge often results in "slop"—low-quality, unreliable content, as exemplified by concerns from mathematicians in the "Leiden Declaration" regarding AI-generated proofs. Furthermore, the economic reality of generative AI, particularly in agentic coding, reveals massive financial losses for providers like OpenAI and Anthropic, with some analyses suggesting they spend "\$1000 for every \$100" earned. This raises the prospect that AI solutions could become more expensive than the human labor they aim to replace, highlighting a disconnect between output volume and real value creation.

Key takeaway

For executives evaluating generative AI investments, recognize that high output does not guarantee Return on Investment or quality. Your focus should shift from mere content volume to verifiable value creation and cost-effectiveness, as current models often produce "slop" and AI solutions may become more expensive than the human labor they aim to replace. Prioritize pilot projects with clear, measurable quality and economic benefits to avoid significant financial losses.

Key insights

AI boosts nominal output, but real value and economic impact remain elusive.

Principles

• High AI output does not equate to high quality or economic value.
• AI costs may exceed human labor costs for equivalent tasks.

Topics

• Generative AI
• Economic Productivity
• Content Quality
• AI Costs
• Agentic Coding
• Leiden Declaration

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Executive, Investor, Consultant

Related on AIssential

• The Convenience Trap — AI is increasing workload and content volume, not reducing it, due to a "verification burden" and the Jevons paradox.
• Coders are refusing to work without AI — and that could come back to bite them — Widespread AI coding tool adoption masks dubious productivity gains and potential increases in maintenance costs.
• Are we already getting past the point of no return with Slop software? — The discussion explores the concern that AI-assisted software development is generating an excessive amount of "slop" software, primarily categorized as "Agentic Wrappers around LLMs" or "Slop apps th…
• A Need for Nuance: The Economist’s Andrew Palmer — AI excels at rapid prototyping but faces significant challenges in production deployment and organizational integration.
• AI Is Too Expensive — AI's current economic model is unsustainable due to massive hyperscaler investments, high operational costs, and unproven profitability of AI labs.
• 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 Marcus on AI.