August 7, 2025
5 min read
Jennifer Follett
Solution providers who secure AI agents and build strong data infrastructure will unlock major opportunities in AI adoption.
The AI gold rush is in full swing, with IT vendors racing to market promising digital transformation that impacts everything from software development to customer insights and cybersecurity. However, there is a critical gap between the theoretical capabilities of AI offerings and their practical delivery — a gap that AI-savvy solution providers are uniquely positioned to fill.
A major focus in 2025 has been AI agents: software programs empowered to autonomously complete tasks on a user’s behalf. These agents grabbed headlines recently with OpenAI’s ChatGPT Agent, ServiceNow’s revenue growth driven by agentic AI, and Walmart’s new AI "super agent" strategy for customers, employees, and partners. Yet, AI agents also faced setbacks, such as an incident where an AI coding agent deleted a production database after violating a directive, highlighting the risks involved. The CEO of Replit, the company behind the agent, publicly apologized for this event. This incident underscores a seismic opportunity for solution providers to help customers move AI agents from mere buzzwords to blockbuster business tools.
Two critical areas where solution providers play a vital role are securing AI agents and building the data infrastructure that supports them.
Originally published at CRN on August 6, 2025.
Securing AI Agents
On the cybersecurity front, managing and securing the access AI agents have to critical data and systems is paramount. As CRN Senior Editor Kyle Alspach explains, "The entire purpose of agents is to connect to many different systems and data sources to autonomously accomplish tasks. But if such access is not actually supposed to be authorized, a data breach is the probable result." Because AI agents operate with limited human oversight, breaches could go unnoticed for extended periods, increasing risks. Solution providers who can help organizations navigate these security challenges will find abundant opportunities. Matt Shufeldt, chief solutions officer at Evotek, notes, "With AI agents, the need for access to data is typically greater. Because of that, you need to solve for those data management issues more quickly."Building AI-Ready Data Infrastructure
Beyond security, solution providers must also address foundational data challenges. CRN Features Editor Rick Whiting highlights key technologies such as ETL tools, data pipeline management, data governance, data quality tools, and vector databases that enable AI projects. Daniel Avancini, chief data officer at Indicium, points out, "Most companies understand they can’t do advanced AI use cases because their data platforms are not ready. They lack the framework, governance, security, or technology to provide AI-ready data. Most companies are not 100 percent ready for large-scale AI applications." These challenges will only grow as AI agents proliferate, consuming more data and accessing sensitive systems. Solution providers who position themselves as experts in securing and enabling AI agents will be well-positioned for success.Originally published at CRN on August 6, 2025.