Report: Half of Genetic AI Projects Could Go Over Budget by 2028
In a recent report titled “10 Best Practices for Optimizing Generative and Agentic AI Costs,” Gartner reveals that many companies may be underestimating the financial demands of generative AI as it transitions from experimental phases to full-scale production.
Understanding the Financial Implications of GenAI
Gartner researchers caution that organizations shifting from GenAI pilot programs to production environments could face significant cost escalations. “Organizations moving from GenAI pilots to production are in for a rude awakening when it comes to cost,” they note. “Creating a production-ready GenAI system can be orders of magnitude more expensive than running a pilot project.”
By 2028, experts predict that at least half of GenAI projects will exceed budget expectations due to suboptimal architectural decisions and inadequate operational expertise.
The Reality of Scaling AI Systems
This warning underscores a pressing issue in the AI industry. While much of the discourse has focused on the capabilities of AI models, Gartner emphasizes that the real challenge for companies will be efficiently operating AI systems at scale.
Inference, the process where a trained AI model is used to generate responses, analyze data, or perform tasks, is a primary cost driver. Unlike training, which incurs upfront costs, inference costs accumulate each time the model is used in production. Gartner forecasts that inference will constitute at least 70 percent of a model’s lifetime cost, shifting the focus from training to the daily demands of deploying AI workloads at scale.
The Complexity of Agent AI
The challenge is even more pronounced with agent AI. Unlike traditional chatbots that provide single responses, AI agents can initiate multiple model calls, retrieve data, access external tools, and execute complex workflows.
As companies implement more autonomous systems, the usage of AI and the associated costs are likely to rise significantly.
Strategies for Cost Management in AI
The key takeaway is that success in the AI era extends beyond model performance. Gartner asserts that to scale generative and agentic AI without incurring unsustainable costs, companies must prioritize cost control, architectural efficiency, model selection, and usage monitoring.
The report reiterates, “By 2028, at least 50% of GenAI projects will exceed their budgeted costs due to poor architectural decisions and a lack of operational expertise.”
For further details and insights, the full report can be accessed Here.
“`
