Large scale organisations are under pressure to undergo a fundamental transformation of processes, functions, and models through the integration of AI-driven capabilities.
Unlike incremental improvements, we are seeing a significant shift in how businesses operate, make decisions, and create value. This moves beyond automation and low-value uses of AI to simplify a process or increase a certain efficiency. This is a move to redefine workflows, customer experiences, and industry dynamics.
Businesses are primarily playing in the Process Reinvention stage, focussing on how to do things faster and cheaper, on opportunities that are prime for automation. Here, AI is used for specific tasks such as fraud detection, information retrieval, customer service chatbots, and optimising marketing campaigns.
A few early adopters are transitioning into Function Reinvention, where AI is orchestrating multiple processes within key business functions – and it is these organisations who are deriving huge value.
For example, we’re working with insurers who are completely reimagining the underwriting process, deploying AI to improve the efficiency and effectiveness at every stage of the value chain. It is at this stage that we’re seeing Agentic AI come to the fore – autonomous agents, working together, to provide more than just a single output and providing a new level of value across multiple complex workflows.
However, the full potential of AI-driven reinvention—enterprise-wide transformation, market evolution, and economic restructuring—is still in its early stages.
Agentic AI represents a step change in AI development, moving beyond traditional Generative AI models. The hype of the last few years has been driven by Generative AI and LLMs, but these approaches operate within a simplistic framework with a single input and a single output.
But the real-world problems and processes that enterprises are trying to optimise with AI are more nuanced and complex than this model allows. The autonomous agents that comprise Agentic AI can make independent decisions, collaborate with other AI agents, interact dynamically with their environment and work toward goals without predefined paths.
Agentic AI extends beyond Gen AI, using other data and AI capabilities and allowing the integration of different models to achieve more complex outcomes. This autonomy enables businesses to build intelligent, goal-driven systems that operate with minimal human intervention, optimising complex processes and workflows.
Currently, we know the dominant perception among our customers is that Agentic AI is a means for low-impact use cases, primarily focused on process optimisation and cost reduction in back-office functions. But there are a number of benefits we can see for our customers as the technology develops:
This also requires reinventing the justification for different use cases, moving beyond the cost savings and efficiency benefits of optimising back-office tasks, to considering how to enhance core capabilities that drive revenue growth. For our insurance example, imagine a series of agents working together across underwriting, pricing, policy execution and claims, ensuring the right products are delivered to the customer in a faster time.
This is an opportunity for organisations to completely reinvent how they reach strategic objectives.
There are two considerations for the success of Agentic AI and how organisations use it in their reinvention journeys. Agent based systems will place even more emphasis on the data foundations of enterprises, specifically on the quality and accessibility of data. As with Generative AI, deploying these complex technologies on low quality data will result in the whole endeavour falling at the first hurdle, and an organisation’s appetite for the technology diminished.
A second consideration is around how solutions are built. The successful adoption of sophisticated AI capabilities extends beyond the solution itself. It relies on us devising the right strategy that aligns with business objectives and a sophisticated solution architecture that ensures the long-term success of any AI capability. This will ensure customers don’t fall into the trap of repeated POCs and endless siloed innovation.
We should also remember that agent-based systems are more than chatbots performing tasks; they are autonomous intelligence capable of interacting with and influencing business operations.
A number of our customers have focused on using LLM for information retrieval, analysis and chat, however these solutions are likely to be commoditised. Now, we are already starting to apply agent-based systems for some of our customers, making them early adopters, who are already seeing the value.
Agentic AI is being used to generate personalised market commentaries. This is currently a costly manual process which doesn’t allow for much personalisation and distracts advisors from higher value tasks.
Mesh-AI has developed a solution that allows wealth managers to generate market commentaries automatically based on various trusted news feeds, allowing for a deeper level of personalisation based on the assets owned by the customer. In this workflow, agents summarise articles, critique content, ensure accuracy, and refine outputs, whereby a human in the loop validates and refines the commentary before sending to the customer. This approach speeds up the process and enhances customer satisfaction, leading to increased client retention and potential upsell opportunities worth millions annually.
As it stands, this example falls within the “process reinvention” category mentioned above. But we can imagine expanding the scope of what this Agent does to more complex interactions for customers - for instance, proactively suggesting portfolio rebalancing based on signals from the market which are then carried out automatically once the customer has given consent. By automating this process, firms can not only reduce the costs involved, but enable advisors to focus more on revenue-generating activities such as client acquisition and portfolio growth.
Mesh-AI is building an agent-based bid response system to enhance efficiency in bid writing, increasing speed, reducing costs, and improving content quality.
The system utilises information from across the organisation and enhances bid responses, making them more efficient to produce and elevating the quality of responses. Agents specialised in both generating and then validating the responses provide an initial first draft needed by technical authors, who can then produce bids more efficiently and at a higher quality.
By reducing the human oversight, this will uplift productivity by 30% and, with increased job satisfaction, reduce the multi million pound employee attrition costs.
Agentic AI is set to redefine business reinvention, moving organisations beyond process optimisation into strategic transformation. As businesses progress through the phases of reinvention, Agentic AI will unlock new levels of autonomy, intelligence, and value creation. The companies that successfully adopt and integrate Agentic AI into their operations will gain a competitive edge in an increasingly AI-driven economy.