By Annapurna Roy
AI agents have been the talk of the town, pegged as the 鈥渘ext big thing鈥 after Generative AI due to its ability to autonomously carry out tasks, marking a shift from language generation and analytics to execution. Nearly 62% of global enterprises are currently experimenting with AI agents, ranging from proof-of-concepts (PoC) to scaled pilots, as per a Nasscom report.ETtech
However, the nature of these experiments is still largely internal, focused on task-level automation with human oversight.ETtech
Despite growing adoption and experimentation with agentic AI, reports also indicate limited maturity of the technology or the use cases, leading to high chances of failure of proofs-of-concept. Gartner expects four in 10 agentic AI projects to be canned by end of 2027. ETtech