Huma Siraj
According to the current state of affairs, agentic AI is promising. with exclusions.
According to a recent PwC poll, 66% of the 300 Senior executives implementing AI agents believe they are increasing productivity. Let’s face it, though: every system produces some level of output. The extra edge that provides severe competitive difference is what executives require.
A report by the authors of the PwC research, not many AI Agents are now “transforming how work gets done.”
A lot of workers are utilizing the agentic functionalities included in Workplace apps to expedite repetitive operations, such as answering inquiries, updating records, and surfacing insights. Although there is a noticeable increase in production, transformation is not achieved.
The PwC writers come to the conclusion that “mindset, change readiness, and workforce engagement are the biggest barriers,” not technology.
The CEO and co-founder of Phenom, which provides agents for HR duties, Mahe Bayireddi, concurs that this is the difficult part. During Phenom’s recent user conference in Philadelphia, I got the chance to speak with Bayireddi, and he made the point that context is crucial for AI agents.
“I believe that this entire process involves a lot of learning,” he stated. “No experts are dynamically stating how they can effectively handle AI agents.”
If agents use the data in an engaging format, manage change well, and use it appropriately, they can increase productivity by about 20–30%, Bayireddi added.
“How they manage the Change Management and make it fly is the key.”
AI agents and the data they use must be domain-specific and will differ from business to business and industry to industry. “In reality, the data at the universal level is complex,” he stated. “The subtleties of context and personalization are crucial for AI to function. It must not be very generic.
As agents become more prevalent, Generative AI becomes increasingly useful. According to him, agents can be “baked into the workflows” once they are implemented. Everyone has required to visit ChatGPT up to this point in order to ask a question and receive a response. It’s not how humans operate.
The focus must be on handling the subtleties of the tasks and procedures that agents are to automate. “That needs to show up in a context-driven, effective format,” he stated.
“Only when an agent is productive in a department can that occur.”
Although Agents will alter the nature of occupations, Bayireddi does not regard them as a danger to employment.
As a result of agents, new jobs will become available. Additionally, a new project will emerge as a result of agents. While having the right skills is important, jobs and work will also change.
When it comes to AI agents, the PwC authors cautioned against settling for too little. Competitors who are prepared to rethink the way work is done will quickly surpass companies that limit themselves to trial initiatives. Few businesses are defining the future early on, creating new operating models that coordinate and integrate several AI agents. Less than half are revamping processes around AI agents (42%), or radically rethinking operating models and how work is done (45%).
