How Agentic AI will revolutionize business operations – are you ready?

Generative artificial intelligence is evolving. Knowledge-based applications such as AI chatbots and co-pilots are being replaced by autonomous agents that can reason and carry out complex, multi-step workflows. These are powered by what is known as agency AI. This latest advancement in artificial intelligence is poised to transform the way businesses operate by understanding context, setting goals, and adapting actions to changing conditions.

With these capabilities, agency AI can perform a range of tasks previously thought impossible for a machine to perform, such as setting sales targets and pitching, analyzing and optimizing supply chains, or acting as a personal assistant to manage employees’ time.

AmazonThe recent partnership with agency AI specialist Adept signals a growing recognition of the systems’ potential to automate diverse, high-complexity use cases across business functions. But to take full advantage of this technology, organizations must first face several challenges with underlying data, including latency issues, data silos, and inconsistent data.

Rahul Pradhan

Rahul Pradhan, VP of Product and Strategy, Couchbase.

Three foundations of agency AI

For its complex functions to operate successfully, agency AI needs three key components: a plan to work on, large language models (LLMs)and access to intact memory.

A plan allows the agent to execute complex, multi-step tasks. For example, handling a customer complaint may include a predefined plan for verifying identity, collecting details, providing solutions, and approving the resolution.

To follow this blueprint, an AI agent can use multiple Masters of Science (LLM) to break down problems and perform subtasks. In context customer In services, the agent can call an LLM to summarize the current conversation with the customer, creating a working memory that the agent can refer to. A second LLM can then plan subsequent actions and a third can evaluate the quality of these actions. A fourth LLM can then generate the final response seen by the user, informing them of potential solutions to their problem.

And just like humans, agentic AI systems cannot make informed decisions without using memory. Consider an AI healthcare assistant with access to a patient’s medical history. medical recordsand past consultations. Remembering and leveraging this data allows AI to deliver personalized and accurate information, explaining to the patient why treatment was changed or reminding them of test results and doctor notes.

Both short- and long-term memory are needed for tasks that require immediate attention and to create an understanding of context that AI can rely on for future inferences. But herein lies one of the biggest hurdles to optimizing agency AI today: often businesses databases These memory systems are not advanced enough to support them, limiting the potential of AI to deliver accurate and personalized information.

The data architecture needed to support AI agents

The dominant approach to meeting memory system requirements is the use of special-purpose, stand-alone memories. database management systems for various data workflows. However, this practice of using a complex network of independent databases can harm the performance of AI in several ways.

Latency issues arise when the different databases used each have different response times, causing delays that can disrupt AI operations. Additionally, data silos, where information is isolated in separate databases, prevent the AI from having a unified view, preventing comprehensive analysis, which causes the agent to lose connections and provide incomplete results. At a more fundamental level, inconsistent data due to differences in quality, formatting, or accuracy can also lead to errors and skewed analysis, leading to poor decision making. Using multiple single-purpose database solutions also creates data sprawl, complexity, and risk, making it difficult to trace the source of AI hallucinations and debug incorrect variables.

Many databases are not suitable for the speed and scalability required by artificial intelligence systems. Its limitations become more apparent in multi-agent environments where rapid access to large volumes of data (e.g. via Masters) is required. In fact, only 25% of businesses have high-performance databases that can handle unstructured data at high speed, and only 31% have combined their database architectures into a unified model. These databases will struggle to meet the demands of GenAI, let alone support any kind of unconstrained AI growth.

As GenAI evolves and agency AI becomes more prevalent, unified data platforms will become central to organizations’ successful AI implementations. Updated data architectures provide benefits by reducing latency with edge technology, efficiently managing structured and unstructured data, streamlining access, and scaling on demand. This will be a significant advance in creating compatible, interoperable and resilient memory infrastructures and allowing businesses to eventually benefit from the automation, precision and adaptability that agency AI offers.

Embracing the AI revolution

Agent AI is ushering in a new era in which AI agents act collaboratively and innovatively, fundamentally changing the way humans interact with technology. Once businesses overcome the challenges of disparate data sources and optimized memory systems, they will pave the way for widespread use of tools that can think and learn like humans with unprecedented levels of efficiency, insight, and automation.

We have presented the best online SQL courses.

This article was produced as part of TechRadarPro’s Expert Insights channel, where we highlight the best and brightest minds in today’s tech industry. The views expressed here are those of the author and not necessarily those of TechRadarPro or Future plc. If you’d like to contribute, learn more here: