Leslie Lee, Head of Product at Squid AI, brings 20 years in enterprise AI, data analytics, scalable automation and cloud product management.
Over the past year, I’ve had dozens of conversations with business leaders trying to make sense of AI agents. The interest is real, and so is the confusion.
Some assume agents are just chatbots with a fresh coat of paint. Others see the potential but think the tech isn’t quite enterprise-ready. Meanwhile, companies further along in their digital transformation are piloting agents yet are unsure how to scale them responsibly.
If you’re trying to cut through the noise, you should know a few things before diving in. The AI landscape is evolving fast; according to McKinsey & Company, 72% of surveyed businesses had adopted AI in at least one business function as of early 2024, and many are looking to move beyond testing. My goal is to help leaders see where to begin when there’s so much hype and open questions about AI.
What An AI Agent Is
Think of an AI agent as a digital teammate that doesn’t just understand what you need but takes action quickly.
Unlike traditional LLMs that simply generate responses, AI agents are designed to drive outcomes. For example, instead of summarizing a help desk ticket, an agent can route it to the right team and close the loop automatically, in real time.
It’s about turning passive data into meaningful results. Most organizations are sitting on huge volumes of underused data. Agents can help unlock that value by embedding intelligence directly into private workflows.
It’s important to note that AI agents are designed to work alongside people. While they can automate parts of a process, employees remain in control. This model reflects a shift toward human-AI collaboration.
Not Just A Tech Project
One of the biggest misconceptions I see is that AI agents are “just an IT thing.” Since agents sit at the intersection of systems, data and people, they affect the entire business.
Leaders across the C-suite, operations, sales and even billing should be involved early. The best use cases are tied to how your business works. If you can pinpoint the manual workflows that slow your teams down, you’ve found prime candidates for agentic AI.
Three Things To Know Before You Start
Here’s what I tell leaders who are exploring agentic AI for the first time:
1. You don’t need perfect data to get started.
Many companies delay adoption because their data isn’t “ready.” However, today’s platforms can often handle fragmented systems and unstructured inputs like PDFs. The goal is progress. You’ll learn faster by doing than by waiting for the stars (or your data) to align.
2. Start with low-risk, high-friction tasks.
Look for repetitive, well-defined problems like triaging tickets or syncing CRM fields. These tasks help you prove value, build confidence and secure internal buy-in.
3. Build with oversight in mind.
Governance isn’t optional. You need to know what agents are doing, who has access and how every action is tracked. Look for platforms that offer audit trails, role- and attribute-based access and compliance with standards like SOC 2. Accountability should be built in from day one, and security should be a core part of how you evaluate any solution.
How Enterprises Are Using Agents
Here’s what we’ve seen firsthand while supporting midmarket and enterprise teams.
Executive Dashboards
A CISO at a national health services organization needed better visibility into real-time security and operational issues across multiple systems. We deployed an agent-powered dashboard that pulls live data from their tools and allows executives to ask plain-language questions to surface real-time insights instantly.
Instead of waiting weeks for analyst-prepared reports, leadership can now identify risks in real time and take action immediately. The CISO also uses the system to pinpoint urgent issues and proactively assign owners, closing the loop faster.
Sales Operations
A global cloud communications company deployed an agent to streamline Salesforce CRM automation. The agent runs in the background with no new tools or workflow changes and automatically enriches opportunity records based on call notes and internal communications.
The company saved over $3 million annually, saw more than 20x ROI and achieved payback in under two weeks. With administrative tasks off their plate, reps now spend significantly more time selling.
Billing Compliance
A major law firm piloted an AI agent to improve billing compliance. The agent interprets firm-specific billing guidelines and scans invoice line items to flag inconsistencies before bills are sent.
The COO expects accelerated collections, reduced manual review hours and increased client trust. These are especially valuable in firms where even minor errors can trigger compliance risks or audit concerns.
The common thread among these three use cases? These companies didn’t wait to get moving. They found a real bottleneck, launched a pilot agent and scaled from there.
The Shifting Competitive Edge
AI agents are quickly becoming essential to staying competitive. A July 2024 Salesforce report found that sales teams using AI were 1.3 times more likely to see revenue growth than those who weren’t. From my perspective, sales is just the beginning. Early adopters across industries are seeing meaningful gains in efficiency, decision making speed and cost savings.
Yes, there’s plenty of hype in the market, but that shouldn’t cloud your strategy. Focus on solving real problems. Involve your business teams early. Most importantly, don’t wait for perfect conditions. Start where you are, and scale what works.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
