The world of agents is already diverse: Travel agents help us plan dream vacations. Insurance agents (sometimes) test our patience and our faith in humanity. Secret agents [REDACTED]. Literary agents are still working on publishing my first novel…But in the world of business, AI agents are the next hot new tech.
AI has already reshaped the workplace. AI-powered copilots can handle employee inquiries. Recruitment platforms can filter through thousands of résumés. AI tools can populate professional development coursework for employees looking to grow specific skills, and much more. Now as agentic AI solutions begin to emerge, workplace tools are beginning to act independently, make decisions, and execute complex workflows with minimal human input.
What makes an agent? Tech Brew reported last month that many top tech companies such as IBM and Amazon would consider tech that is “able to interact with other tools and plan and execute workflows on their own” as agents.
“These offerings, potentially some of them, they were not possible a few months ago or a year ago,” said Chano Fernandez, co-CEO of the AI-powered recruiting tool Eightfold. “You can decide which ones of them are to augment work in combination with your teams and which ones are not.”
This shift to agentic AI represents a major unlock in workplace technology. While many AI systems and solutions can already help users make better decisions by generating insights and recommendations, they still require humans to complete tasks. But agentic AI can initiate and complete tasks on its own, learning from interactions and adapting to changing conditions without human direction.
Agent vs. copilot. While AI copilots assist users in decision-making, agents can perform autonomous tasks and make decisions on behalf of a person or organization. Agents are also able to perform complicated operations within specific human-defined parameters, but don't require constant oversight. Copilots are great with summarizing large swathes of content, analyzing complex sets of data, or even creating new content.
“Agentic AI is, at the core, the autonomous nature of agents that can perform tasks, autonomously, semi-autonomously, mimicking human-like behavior to enhance the workflow,” said Anthony Abbatiello, partner and workforce transformation leader at PwC. “[AI agents] actually have the thinking and learning models to make decisions that mimic human behavior and human decision-making, and can learn and continue to process, versus chatbots, which are automating the cycle.”
Users can query an AI copilot to help answer a question. It will seek the answer within the resources it’s trained on, and then quickly reply with an answer or advice or insights. AI copilots are really useful for a lot of rote ER processes or content creation but they “sit in that one loop.”
Quick-to-read HR news & insights
From recruiting and retention to company culture and the latest in HR tech, HR Brew delivers up-to-date industry news and tips to help HR pros stay nimble in today’s fast-changing business environment.
The transition from assistive AI tools to agents upends how businesses leverage the technology. Instead of waiting for prompts, agents can identify issues and opportunities and respond in real time. Agents carry out tasks and manage workflows on their own, learning how to improve the process along the way without needing to retrain the model.
Impacts to HR. Agentic AI will likely change how companies and HR teams manage the workforce and productivity, talent acquisition (TA), learning and development (L&D), and even employee engagement.
HCM platforms, such as Oracle and Workday last month or SAP late last year, have announced product developments and changes that incorporate agentic systems inside their platforms.
“Everything about the web and the web architectures of all of these products is about to get disrupted. Everything,” said HR industry analyst Josh Bersin. “The AI technology can make sense of a huge amount of data and act on your behalf. None of these systems could before.”
Take TA and recruiting, for example. Agentic AI can autonomously source candidates, assess résumés, schedule interviews, and even conduct initial screenings—all while ensuring legal compliance and adherence to company hiring and diversity policies. Unlike traditional AI-powered ATS tools, which require recruiter intervention, an agentic AI hiring system could refine job descriptions, adjust outreach strategies based on candidate response rates, and handle logistics end-to-end.
“LinkedIn created something called the LinkedIn Hiring Assistant,” Bersin said. “It’s actually great, but all it really does is it uses LinkedIn for you, in a little bit of a faster way. It doesn’t do anything that LinkedIn couldn’t do before. It just does it in a more autonomous way.”
LinkedIn last fall released its first agent in its Recruiter platform. TA pros can upload job descriptions, intake notes, and job postings to the assistant and the tool will “translate that information into role qualifications and build a pipeline of qualified candidates,” according to LinkedIn.
These agentic tools come amid a tough environment for TA pros and can support thin-spread recruiters, identifying and courting the right candidate without much human oversight.
“It can now actually start to interact with the person in the same way a human might,” Abbatiello said. “Whereas the chatbots and copilots have good machine learning behind them, it doesn’t mimic the decision-support and thought-process and then actually bring someone to a new place.”