![\"\"](\"https:\/\/blogs.infosys.com\/quality-engineering\/wp-content\/uploads\/2026\/08\/QE-Shifts-from-Implementation-to-Business-Intent-with-AI.jpg\")
\nFig 1: How AI-driven software development shifts defects upstream<\/p>\n
For years, organizations have used shift-left and shift-right practices for the enhancement of software quality. Shift left embedded quality assurance (QA) into software design, enabling early issue detection. Shift right implemented quality practices in production to validate application performance. Together, these approaches minimized defects, accelerated feedback, and strengthened delivery.<\/p>\n
However, as enterprises adopt artificial intelligence (AI) and agentic AI development, these traditional quality engineering (QE) approaches are falling short.<\/p>\n
A new QE model is required: one that goes beyond validating requirements and code reviews. QE should start by validating AI-generated output against business objectives, constraints, assumptions, policies, and regulatory expectations.<\/p>\n
Besides code, AI also generates user stories, business requirements, test cases, and operational as well as decision-making workflows. As a result, defects can originate earlier in the development lifecycle, with business intent becoming machine-readable input.<\/p>\n
To summarize, the shift-left approach moved quality into software requirements. AI-driven development moves quality further upstream to business intent. This blog explores how AI is driving a change in QE approaches, from validating software requirements to focusing on business intent.<\/p>\n
Traditionally, QE focused on identifying issues during implementation. This included misinterpreted requirements, incorrect business logic, and missed edge cases. With AI-driven delivery, the risk shifts from implementation to interpretation.<\/p>\n
AI does not produce new types of defects. Instead, it amplifies misinterpretations at scale. For example, a missing business rule, an overlooked exception, or an incomplete prompt can span multiple delivery stages, as illustrated in Figure 1. Although requirements may appear complete, code may function correctly, and automated tests may pass, the intended business outcome may still not be achieved.<\/p>\n
When business intent is unclear or incomplete, AI can generate technically accurate output that is misaligned with organizational goals. The result is a solution that works as designed but not as intended. This is especially significant in regulated industries and customer-facing processes. Without business context, even technically accurate output can compromise enterprise operations, compliance, or customer experience.<\/p>\n
\nFig 1: How AI-driven software development shifts defects upstream<\/p>\n
Consider a scenario where a regulatory requirement includes an exception for a small customer group. If the exception is missing in the AI prompt, the resulting requirements, code, and test cases may appear correct. However, the output in fact is incorrect. Here, the risk lies in the missing business context, not in development or testing.<\/p>\n
Additionally, during product delivery, intent can drift from the original business goal, creating a predictable pattern. Requirement misinterpretations can cascade across code, testing, and operational performance. Therefore, QE must address risks at the source before they intensify across delivery stages.<\/p>\n
In an AI-driven environment, QE teams should focus on:<\/p>\n
Organizations should embed traceability across business intent, AI-generated output, test results, and production outcomes. A single implementation can provide insights for scaling across the enterprise.<\/p>\n
Shift left improved quality by moving testing earlier in the development lifecycle, while shift right extended quality into production. However, AI presents a new frontier for QE.<\/p>\n
As AI tools can code, test, and support decision making, risks are not confined to development alone. They can arise when business intent is defined, interpreted, and translated into action.<\/p>\n
Organizations that validate and preserve business intent throughout the delivery lifecycle are better positioned to scale AI responsibly and deliver reliable business outcomes.<\/p>\n
In sum, the future of QE primarily depends on code interpretation accuracy in addition to code quality with a clear focus on business intent.<\/p>\n","protected":false},"excerpt":{"rendered":"
For years, organizations have used shift-left and shift-right practices for the enhancement of software […]<\/p>\n","protected":false},"author":1131,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"inline_featured_image":false,"footnotes":""},"categories":[50,89],"tags":[170,63,237],"coauthors":[236,238],"class_list":["post-1029","post","type-post","status-publish","format-standard","hentry","category-ai","category-quality-engineering","tag-agentic-ai","tag-quality-assurance","tag-shiftlefttesting"],"acf":[],"_links":{"self":[{"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/posts\/1029","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/users\/1131"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/comments?post=1029"}],"version-history":[{"count":4,"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/posts\/1029\/revisions"}],"predecessor-version":[{"id":1037,"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/posts\/1029\/revisions\/1037"}],"wp:attachment":[{"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/media?parent=1029"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/categories?post=1029"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/tags?post=1029"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/blogs.infosys.com\/quality-engineering\/wp-json\/wp\/v2\/coauthors?post=1029"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}