Sravan Vadigepalli outlines six guidelines for governing enterprise AI agents
Lowe's enterprise AI strategy lead Sravan Vadigepalli proposed six guidelines for organizations deploying AI agents: delegate routine data work, prioritize principles over rigid rules, encode organizational culture, route decisions through confidence thresholds, strengthen context and management, and manage exceptions. An accompanying IEEE Spectrum commentary argues that AI expands engineers' role from performing technical steps to framing problems, coordinating specialists, testing answers, and retaining responsibility for final decisions. TechNews also relayed a Project NANDA report estimating $30 billion–$40 billion in enterprise generative-AI investment, with only about 5% of integrated pilots creating value.
The framework gives enterprises a practical governance structure for assigning work to AI agents while retaining human judgment, accountability, and auditable review.


美國家居零售商 Lowe′s 企業 AI 策略主管、IEEE 資深會員 Sravan Vadigepalli 5 日於《IEEE Spectrum》撰文指出,AI 代理(AI agent)接手越多日常工作,工程師、產品經理、分析師與營運人員的角色,正從親自執行任務轉為設定意圖、原則與邊界,他稱之為「管理者轉變」(governor shift)。
他引用麻省理工學院媒體實驗室(MIT Media Lab)Project NANDA 的 2025 年報告指出,企業對生成式 AI 的投資估算達 300 億至 400 億美元,但多數組織尚未展現可衡量的損益影響,僅約 5% 整合試驗專案有創造價值。他認為,落後企業使用的模型與成功者並無二致,真正欠缺的是能指揮系統並為結果負責的人才。
#Sravan Vadigepalli#Lowe's#IEEE Spectrum#IEEE#MIT Media Lab#Project NANDA