BEIJING – JANUARY 19: A McDonald’s employee hands out food as she waits on customers at their new … (+)
Shortcuts litter software history. We’ve been working to try and provide intelligent assistants, macros, shortcuts and workaround automations for a very long time. Notwithstanding the abomination that was Clippythe technology industry’s latest work in this space is of course focused on AI across the swathe of generative intelligence development. As we know, this work now dovetails with the “employment” of agentic AI functions designed to provide human-like complex problem-solving with little or no user intervention.
That term employment might be more important than we think i.e. at what point are we going to start talking about software copilots actually taking over workforce roles (at least in part, if not wholly) and really starting to contribute to an organization’s bottom line?
There will be much talk of autonomous workers this year. To get to the crux of the matter, we will need to start analyzing which human workflow tasks are the ripest for digital automation and (alongside the pressing need to enforce security and compliance over these new elements) where we can bring these new worker robots into teams to perform tasks above simple chatbot functionalities.
With an opinionated view on this topic drawn from witnessing the hype cycle in this space at its most fervent, CEO of agentic AI startup Integrail Peter Guagenti says that over this past year, specific and specialized AI agents have emerged to take over common business functions, automating everything from sales and marketing activities to software engineering tasks.
EmployAI Of The Month
“A new generation of agentic AI platforms are now extending these capabilities, providing companies with a ‘starting point template’ for their own AI workers but adding the ability to tailor each to their unique ways of working to create an ‘employee of the month’ for any given assignment,” stated Guagenti, speaking to press in London this month. “These platforms also introduce connections to a vast array of business systems and data sources, allowing AI to move across systems and to have context awareness of the business, ensuring that the AI worker is able to perform like a fully trained employee from day one.”
In short, with these new agentic AI platforms, we are seeing AI workers able to be brought to life through plain language instructions and just a handful of clicks. If someone can describe a work process, define requirements and delegate a task, they can now create AI agents that complete the most complex tasks.
Insourcing Your Outsourcing
“With agentic AI, businesses benefit from specialized AI agents tailored to specific functions and tasks that allow for streamlining work, eliminating time-consuming or low-value tasks and insourcing jobs that currently reside in consulting or offshore (outsourcing-based)
teams,” said Guagenti. “Within the next three to five years, AI workers will become just part of our world, our workforce. Managers will oversee teams of both human and digital workers and we’ll delegate tasks to AI like we would any worker.”
How fast will this unfold? Let’s remember that the rise of cloud computing Software-as-a-Service platforms gave companies incremental productivity improvements and expanded business capabilities, which compounded slowly over the course of nearly two decades. Guagenti suggests that agentic AI stands to deliver an order of magnitude more impact than SaaS and this change will emerge in less than half of that time.
We can say that these statements are somewhat underscored by analyst house McKinsey in a recent report that states, “Agentic AI is acting autonomously.” McKinsey notes that “In 2023, an AI bot could support callcenter representatives by synthesizing and summarizing large volumes of data, including voice messages, text and technical specifications in order to suggest responses to customer queries. In 2025, an AI agent can converse with a customer and plan the actions it will take afterwards. So for example, processing a payment, checking for fraud and completing a shipping action.”
In terms of broader economics, ongoing efforts to improve corporate productivity and reduce operational costs have taken their toll. Since the start of post-pandemic belt-tightening, businesses have struggled to do more with less while competitive pressures continue to rise in nearly every category, leaving many organizations understaffed to address it. At the same time, business leaders expect AI to reduce their cost to operate and to make their products and services dramatically better than their competitors.
As of now, though, only a few big tech brands or AI startups have the volume of data scientists and AI engineers to pull this off. Businesses see a clear path forward, but the vast majority of companies are struggling with the complexity of this new technology. Furthermore, they don’t fully understand what is possible with agentic AI and where or how they can find immediate value. What’s more, they lack the expertise to deploy and manage AI infrastructure on their own.
Why AI Is Overwhelming
“Setting up a large language model, curating the data required to provide context to those models through retrieval augmented generation and building out the integrations to the business systems used every day all are overwhelming tasks for many enterprises,” detailed Guagenti, in an effort to explain just how tough it has been to bring new intelligence online.
The answer perhaps lies in the idea that enterprises are looking for platform-level solutions that help them identify opportunities for AI and convert ideas into reality quickly. This is what Integrail has set as its mission i.e. a no-code and no-integrations toolset designed that offers guidance and packaged services to create agentic AI functions.
There will be much more to come in this space and the amount of white noise already being generated around agentic AI itself has already reached an ear-shattering cacophony.
Building these functions into the workforce via no-code and no-integrations is admirable, but the longer-term (almost human-cultural) impact of these changes is yet to be seen. In an era when we’re still talking about back-to-office initiatives, the rise of digital nomads and a massive proliferation of smart city connections driven by edge computing designed to further automate our lives, it remains to be seen just exactly how a new AI-compliant status quo will settle across the business landscape.
AI is with us at work yes, but has it yet to embarrass itself at the staff party and learned to get the coffee order in for everyone? A dose of realism may need to surface.
