A new Gartner forecast has cast a shadow over the current enthusiasm surrounding agentic AI, predicting that over 40% of such projects will be discontinued by 2027. The advisory firm attributes this potential fallout to unsustainable costs, unclear business benefits, and underdeveloped risk management strategies.
Agentic AI, which refers to systems capable of autonomously executing tasks and pursuing defined goals, is emerging as a step beyond traditional automation. Yet, most of today’s implementations are still limited to early-stage experiments or proof-of-concept deployments. According to Gartner’s January 2025 poll of 3,412 participants, only 19% of organisations had made significant investments in agentic AI, while 42% were approaching the space cautiously and 31% remained unsure or on the sidelines.
Agent washing clouds the market
Compounding the issue is a trend Gartner refers to as “agent washing,” where vendors rebrand conventional technologies such as chatbots, robotic process automation (RPA), and virtual assistants as agentic AI, even though they lack true autonomous capabilities. The report estimates that only about 130 vendors globally possess genuine agentic functionality, creating a credibility gap in the ecosystem.
Anushree Verma, senior director analyst at Gartner, noted that most agentic AI projects today are driven more by hype than by operational value. “This can blind organisations to the real cost and complexity of deploying AI agents at scale, stalling projects from moving into production,” she said.
From buzz to business value
Despite the current hurdles, major technology companies such as Oracle and Salesforce continue to invest heavily in agentic AI to enhance efficiency and cut costs. Gartner expects that by 2028, up to 15% of routine workplace decisions will be made autonomously by AI agents, a significant jump from zero percent in 2024. Similarly, one-third of enterprise software is predicted to integrate agentic features by that time.
Also read: 90% of Public Sector to Adopt Agentic AI: Capgemini
Agentic AI holds potential to automate decision-making, orchestrate workflows, and learn continuously from its environment. But its long-term success depends on how well businesses can navigate the transition from pilot testing to large-scale implementation, while mitigating emerging risks.
A cautious road to adoption
Gartner’s outlook serves as a cautionary note for CIOs, CTOs, and innovation leaders. The firm recommends that enterprises prioritise agentic AI use cases that clearly demonstrate business value and are compatible with existing data, compliance, and infrastructure frameworks. Without this clarity, organisations risk investing in technologies that don’t mature past their experimental stages.
As the AI industry evolves, clarity and credibility will be essential. While agentic AI promises to redefine automation and enterprise intelligence, the path ahead will require discipline, purpose-driven deployment, and a firm grip on both risks and returns.
