New Techreviewer.co Research: 89% of Software Companies Now Use AI to Write Code, but 90% Report at Least One Downside

Techreviewer.co's survey of 127 software-development companies found that 89% now use AI to write or assist with code. The median respondent said AI manages 26% to 50% of its codebase, while roughly one-quarter reported AI generates more than half of their code. The findings indicate broad and growing AI adoption in software development, though the survey is unlikely to have significant near-term market impact.

Analysis

This is directionally supportive for AI infrastructure and developer-platform demand, but the survey is too small and self-selected to alter near-term estimates. The investable question is whether coding-assistant adoption converts from seat growth into sustained inference consumption and incremental cloud workloads. MSFT (GitHub Copilot/Azure), GOOGL (Gemini/Google Cloud), and AMZN (AWS) benefit if AI-assisted development increases software release velocity, because faster product cycles can expand application deployment, testing, security scanning, and cloud usage rather than merely replace developer seats.

The more material second-order effect is margin pressure on labor-arbitrage IT services. If clients begin underwriting projects on output rather than billed headcount, EPAM, GLOB, WIT, and potentially ACN face a structural mix shift: productivity gains initially support delivery margins, but contract renewals should eventually pass much of that gain to customers. Over the next 6-18 months, watch bookings, revenue-per-employee, utilization, and fixed-price-contract pricing; rising productivity without corresponding revenue growth is bearish for services multiples. The contrarian view is that AI-generated code raises integration, security, and governance workloads, delaying net headcount displacement and favoring premium engineering vendors over lower-end offshore providers.

No immediate directional trade is warranted from this release alone. A credible catalyst requires independently reported evidence that enterprise software-development budgets are being redirected from external labor toward AI tooling and cloud consumption; the first confirmation should appear in GitLab, Microsoft, Alphabet, Amazon, EPAM, and Globant guidance over the next two earnings cycles.

Key Decisions for Investors

• Maintain a 3-6 month relative-value watch: long MSFT or GOOGL versus short EPAM or GLOB only after services companies disclose declining revenue-per-head or pricing pressure while hyperscalers cite accelerating developer-AI usage. Target 10-15% relative return; exit if services bookings reaccelerate and utilization improves simultaneously.
• Prefer MSFT over standalone developer-tool names for AI-coding exposure: Copilot monetization is additive to a diversified earnings base, whereas GTLB requires sustained seat expansion and premium valuation support. Reassess after the next two quarterly reports for disclosed AI revenue, net retention, and cloud-consumption commentary.
• Monitor ACN, WIT, EPAM, and GLOB for fixed-price backlog mix and gross-margin guidance over the next 1-3 months. A margin expansion accompanied by flat hiring is not automatically bullish; it may precede customer repricing at renewal, making any rally in labor-intensive services a potential short-entry setup.
• Do not buy broad AI software ETFs solely on this signal. Trigger a more constructive sector stance only if enterprise CIO surveys or public-company disclosures show incremental budgets for code-generation, testing, security, and cloud deployment rather than substitution within existing IT spend.