An Executive’s Crash Course In AI Agents

In this segment of “Office Optional with Larry English,” Larry gives an overarching look and what executives need to know about AI agents.

OpenAI launching agents in 2025. Salesforce’s CEO announcing AI agents as the third wave of AI. Microsoft adding agent capabilities to Copilot.

The message here is clear: AI agents are going to be big, and leaders need to begin strategizing now about how to incorporate this powerful technology into their organizations.

If you’re not sure what AI agents are, you’re already behind the AI curve. You’re also not alone — AI is developing at a dizzying pace, and most leaders are struggling to keep up. “Innovation is happening faster than you can imagine or adapt to, and large organizations are racing against time to move from data to value to insights to action,” notes Abhas Ricky, chief strategy officer at Cloudera, a hybrid data platform.

Read on for an AI agent crash course, including a definition of this new technology and answers to questions about security, team impact and the investment required for leaders to get their organization caught up.

What Are AI Agents?

AI agents are advanced AI systems that can complete complex tasks and make decisions on their own. They can analyze data, make predictions, offer insights, converse, solve problems, create strategies and more. They learn over time and adjust to real-time data, offering a high level of accuracy, efficiency and agility.

How are AI agents different from ChatGPT and other LLMs? AI agents work independently, following instructions to use a variety of tools to complete tasks. ChatGPT doesn’t do anything on its own — humans must enter a question or prompt to get a response.

Like any tool, AI agents aren’t going to magically solve every business problem. But they are extremely powerful — especially when you combine agents together to create agentic workflows, which allows them to accomplish complex tasks.

Answers To 7 Burning Questions About AI Agents

Any new technology brings with it a wave of concerns, fears and excitement. AI agents are, of course, no different. Here are some answers to top questions from leaders about the technology:

Are AI agents just fancy chatbots?

Nope. This is a common misperception. During a recent webinar on AI agents my company Centric Consulting hosted we asked attendees what they thought AI agents were. Nearly 20 percent responded with “chatbots.” Chatbots are reliant on user input, whereas agents use AI and natural language processing. AI agents can have a conversational interface — just like a chatbot — but it’s not a requirement.

How do I minimize security risks when handing over entire processes to AI?

Agents should have clear rules around what they’re allowed to do. Agents that try to be all things to all people tend to fail spectacularly. Once the agent is live, actively monitor inputs and outputs during the initial use phase. This helps provide transparency and explainability, creating an audit trail so you can have confidence in the technology. As you scale, you can transition out to passive monitoring to flag anomalies.

“In the first phase of deploying agents, you need to put humans in the loop all the time,” says UiPath CEO Daniel Dines.

Is there any data yet on how AI agents will impact organizations?

One study predicts that agentic AI will achieve 60 percent productivity gains for organizations. AI agents are most powerful when combined to create agentic workflows. Compared to single, one-off AI agents, agentic workflows can tackle more complex tasks, solve more complex problems and achieve greater boosts in efficiency and productivity.

What do agentic workflows look like? Ricky shares an example of agents working together to achieve tax invoice reconciliation and loan underwriting in an autonomous fashion. “Imagine agent one is reading tax documents, agent two is extracting additional sources, agent three is benchmarking the tax data, agent four is writing the memo for you, agent five is fact-checking the memo and agent six is formatting the memo,” Ricky says. “There’s potential for 99 percent productivity improvements, as well as vast improvements in consistency.”

OK, but how are companies using AI agents in real life?

Some forward-thinking organizations have already deployed AI agents successfully. The technology is making inroads across many industries, including insurance, marketing, manufacturing, customer service, financial services, supply chain and healthcare.

For example, my company recently helped a healthcare technology organization build an AI agent to analyze and extract demographic data from disparate sources (patient charts, pharmacy orders, hospital discharge notes, etc.) to help manage prescriptions. The tool reduced manual labor by 82 percent and increased accuracy to nearly 100 percent.

To share just a few more quick examples:

• Hippocratic AI, a generative AI healthcare company, designed an AI agent for “low-risk, non-diagnostic, patient-facing healthcare tasks,” reducing the burden on overworked nurses, social workers and nutritionists.
• An energy company that is a client of Centric is using AI agents to extract valve specifications of each valve piece from engineering specification documents and sales brochures. Along with inspection forms, these specifications are used in a machine learning model to determine if an inspection has potential anomalies.

How will AI agents impact jobs?

AI will cause some jobs to become obsolete. But it will also create new opportunities — although these new jobs will take some time to emerge. Leaders must figure out how to create workers of the future who are adept at using AI to solve problems and innovate.

There’s also a cultural component to how AI impacts jobs. Yes, some leaders will choose to cut headcount. But leaders can instead choose to position the technology as a tool for accelerating market growth or super augmenting your most valuable asset — your people.

“It’s more about job transformation than job elimination,” Dines says. “Jobs will evolve as robots and agents take over some tasks. But it’s actually very hard with the existing technology to replace a job. Agents can do very specific tasks, while most jobs are broader.”

How can my team keep up with AI?

It is possible for organizations to keep pace with AI’s quick progression, but it requires investment and nimble strategy. Ricky suggests that leaders need a change in processes around innovation. Agile doesn’t cut it any longer, he says. “The challenge with applying the agile mentality to the AI world is there’s a new model every Sunday, a new agent every Monday and a new framework every Tuesday. By the time your agile team adapts, you’re already behind. You have to embed the core skill set and processes of AI into your development cycle.”

How can I get started with AI agents?

Unfortunately, there’s no shortcut here. This is a heavy strategy lift. But to paint a broad picture of what you need to do to get started with AI agents:

First, figure out your top priority use cases with AI vision workshops. Repeat this exercise every six months at a minimum. If you revisit your use cases and priorities just once a year, you’ll be woefully behind — the landscape is evolving rapidly, and what’s possible today will have changed by next month.

“This is probably the fastest growing technology there is,” Ricky says. “Agentic workflows and regenerative agents are being developed by large organizations and multiple vendors for a large variety of use cases at a breakneck speed. Designing agentic systems to enable agents that use foundation models to execute complex multi-step workflows across a digital world will help move from thought to action.”

Then, create an AI roadmap and define agent goals and KPIs. Set up data, security and compliance governance. Like any other AI tool, data is the foundation for your success. “You need to be able to trust the data that you’re going to use to train your model,” Ricky says. “Only then will you get the insights and actions you can trust.”

Finally, build a smart team that understands the goal and role of AI in your organization. This team should take charge of continuous learning and adaptation as AI advances.

Ricky cautions leaders that AI and agentic systems — done correctly — is a capital-intensive game. “We’re looking for a relatively larger than usual capital investment into AI technologies today with the expectation that it will yield results many times bigger than what you’re investing in,” he says.

“You can’t wait for ROI to come through on a test use case and then deploy a bigger budget. This is one of those technologies that requires a leap of faith. By the end of the year, companies investing in AI applications and agentic workflows will outpace those that aren’t. They’ll leverage agentic systems that manage multiplicity, respond to natural language and work seamlessly with existing software tools and platforms — accelerating their benefits and beating the competition in a shorter time.”

While integrating AI agents into your organization can be challenging — there’s a lot of strategy to consider, important governance to put in place and team members to involve — the potential benefits are enormous. Leaders need to act now to begin strategizing how to use this powerful technology to transform their organizations and capture AI ROI.

This article was originally published on Forbes.com.

Are you ready to explore how artificial intelligence can fit into your business but aren’t sure where to start? Our AI experts can guide you through the entire process, from planning to implementation. Talk to an expert