Apollo's chief economist has argued that artificial intelligence has so far produced little or no financial payoff for most companies outside the technology sector, according to a report cited by TechCrunch. While the claim reflects one economist's assessment rather than a formal regulatory or market-wide ruling, it adds to a growing business debate about whether AI investment is translating into measurable returns.

For European businesses, the point is less about whether AI matters in the long term and more about how companies evaluate its near-term commercial value. Many organisations are experimenting with generative AI, copilots, chatbots and workflow automation, but not all pilots produce clear savings, higher conversion rates or operational improvements.

What happened

TechCrunch reported on comments from an Apollo economist who said AI has delivered "zero financial payoff" to most companies in corporate America outside the tech industry. Based on the source provided, this is an assessment of current business outcomes rather than a new study, legal development or technology launch.

The argument challenges a common assumption in the market: that AI spending automatically creates business value. In practice, many organisations are still in early adoption stages, where costs are easier to identify than benefits. These costs can include software subscriptions, model access, integration work, data preparation, employee training and governance measures.

Why it matters for European businesses

European SMEs and mid-market firms are under pressure to adopt AI for productivity, marketing and customer service. At the same time, budgets remain tight and management teams increasingly want evidence that AI projects improve margins, reduce manual work or support revenue growth.

This matters because AI can create a gap between activity and outcome. A company may deploy AI tools broadly across teams, yet still struggle to show measurable gains if workflows are poorly designed, data quality is weak, staff are insufficiently trained or the chosen use cases are too generic.

For many businesses, the most realistic value may come from targeted automation rather than broad AI transformation. Examples can include drafting repetitive content, summarising customer interactions, supporting internal search, accelerating product data workflows or improving service triage. Even then, the business case depends on whether time savings actually convert into lower costs, faster delivery or better sales performance.

The story is also relevant because it supports a more disciplined approach to AI investment. European businesses adopting AI alongside compliance, privacy and cybersecurity requirements cannot always afford long experimentation cycles without clear business objectives.

Who may be affected

• SMEs and founders evaluating whether AI tools justify their subscription, integration and training costs.
• Marketing teams using generative AI for content, campaign support, SEO and customer engagement without clear measurement frameworks.
• E-commerce businesses testing AI for product descriptions, support automation, merchandising or personalisation.
• IT and digital teams responsible for implementing AI tools while proving value to management.
• Companies outside the tech sector where processes, data maturity and operating models may make returns slower to realise.

What companies should consider

• Define the business problem first. Start with a specific operational or commercial bottleneck, not with the tool itself.
• Measure ROI at use-case level. Track outcomes such as hours saved, support resolution time, conversion uplift, content production cost or error reduction.
• Separate pilot success from business impact. Strong user feedback does not automatically mean financial value.
• Prioritise workflow integration. AI is more likely to deliver returns when embedded into existing processes, systems and team responsibilities.
• Review total cost. Include licensing, integration, governance, security, review time and staff training when assessing value.
• Focus on repeatable use cases. The clearest returns often come from high-volume, repetitive tasks rather than one-off experimentation.
• Maintain governance. Especially in Europe, businesses should align AI projects with privacy, security and emerging compliance obligations instead of treating adoption as purely a productivity exercise.

The broader takeaway is not that AI lacks value, but that value is uneven and may take longer to prove outside the technology sector. For European businesses, the more practical question is which AI deployments can produce measurable gains under real operating conditions, rather than whether AI should be adopted in principle.