Organisations have begun to examine the role of artificial intelligence in business workflows.
)
Artificial intelligence (AI) certainly was the most talked about topic at Forrester’s Technology & Innovation Summit. However, it appears that the conversation has moved on. In preparation for an era of agentic AI organisations are beginning to consider where employees should fit, where to use contractors or external service providers, and what tasks AI can be used to enable internally.
The AI-enabled tasks are a
Mark Moccia, Forrester’s research director, wrote a blog post in conjunction with the event. He discussed how a third CIOs would adopt “gig worker protocol”, where IT teams will include AI agents, gig workers, and employees who have multiple jobs. Manuel Geitz, Forrester’s principal analyst, spoke at the Technology & Innovation Summit about how business and tech leaders can prepare for this change. “You need to understand which expertise is needed to drive your business model,” Geitz told the audience.
Geitz believes that IT leaders can prepare their organisations for workflows which may be split among internal staff, contractors, and AI agents, by capturing knowledge using structured data. He suggested that delegates could then experiment with business models to monetise the expertise on demand. Eventually, he said, they can build a platform in which AI agents are the frontline for knowledge delivery supported by humans.
The idea that AI agents can be used to create a gig economy is gaining popularity among industry commentators. Jessica Apotheker (Managing Director and Chief Marketing Officer at Boston Consulting Group) discussed with Computer Weekly how the marketing function – which draws on both internal and external expertise – could evolve using an agentic AI workflow. She gave the example of the content production workflow.
She said, “There are a lot of external people who work in the content group.” “There are creative agencies, production companies, and localisation agencies. There are internal people, local marketers and tech people. All these people must come together to reinvent.”
Apotheker says that AI has the ability to change the workflow process for content. IT and business decision makers need to rethink what parts of the process are theirs and what can be automated or outsourced to service providers who may use AI and automation.
Pricing value
According to BCG research, organisations that see significant business benefits by deploying AI are AI-first. This means that business leaders rethink the role of people in a process or workflow where some aspects will be automated using AI.
Apotheker said, “Think about an AI-first workflow.” “You have to rethink the things you buy and make. It’s not clear that your current buy or make strategy is what you need. You want your contractor to automate their part of the workflow.”
Prem Ananthakrishnan discussed in a recent Accenture podcast how the use AI and agentic AI is changing the way people think about software.
He said, “There’s a fundamental shift from thinking of software as just a tool to software as a collaborator who drives an outcome for the company.”
The next step in software licensing is to move beyond consumption-based pricing. Ananthakrishnan echoed the remarks made by the Forrester analysts, and BCG’s Apotheker: “We still view software purchases as tools. We need to think of procuring a vehicle for collaboration. I believe that IT buyers must evolve their thinking from procurement to performance- and design-driven thinking. Do not think about purchasing software anymore. Ananthakrishnan says that these digital teammates are going to be paid on the basis of outcomes, using “value-based pricing”.
It’s a big shift in mindset, but IT and business leaders can start by evaluating which parts of a process are strategic and should remain within the company. In conversations Apotheker had with organisations who are considering a radical overhaul of their workflows, she said that: “Either you choose a BPO and fully outsource, hoping that they will transform the processes with AI and reward on outcomes, OR you reshape them internally.” The more an AI-based service is used, then the more tokens and credits are consumed. He said that these consumption-based models are a bridge leading to an outcome based pricing model, where organisations hire AI agents for work.
Ananthakrishnan suggested that IT leaders begin implementing business impact metrics such as linking return-on-investment to an AI credit. They could also consider a hybrid pricing model based on an upfront AI credit where the supplier receives a bonus for achieving a specific outcome.
As working practices adapt to include AI-enhanced tasks, there are many things to consider. But, regardless of whether AI-enhanced tasks are achieved internally or through an external service provider – value-based pricing will be coming. And people in IT leadership, procurement, and other departments, will need to assess the risk versus return when the product or services being procured are in a probabilistic rather than a deterministic environment.
Read More on Artificial Intelligence, Automation and Robotics
-
By Cliff Saran
Google introduces array of technical agents across its apps
By: Esther Shittu
ServiceNow targets “data hell” and BI play
