Why Agentic AI needs quality data to deliver benefits to Australian businesses
By Peter Philipp, General Manager – ANZ at Neo4j
As Artificial Intelligence (AI) continues to evolve rapidly, its latest capabilities are gaining increasing attention from Australian businesses.
Dubbed ‘Agentic AI’, the advancement involves the creation of self-directed ‘agents’ that can make informed decisions by drawing on multimodal data and underlying algorithms. They can then ‘learn’ from the outcomes of those decisions.
The key strength of Agentic AI, compared to earlier versions of the technology, lies in its ability to plan, operate, and adapt independently, executing tasks from start to finish.
For example, in a supply chain, AI agents can analyse market trends and historical demand to anticipate stock requirements. This can then help to avoid out-of-stock scenarios by automating restocking processes.
These AI agents can automatically respond to fluctuating market conditions and adjust their behaviour to better support supply chain optimisation. It’s no surprise, then, that more than one in four leaders (26%) say their organisations are beginning to define strategic roadmaps for the deployment of Agentic AI.
However, as promising as the technology sounds, it also calls for careful oversight. Given all its autonomous power, how can the actions and outputs of AI agents be fully trusted?
In the same way the human brain uses observation and extra inputs to draw conclusions, AI agents require extensive external sources and signals to enhance their reasoning capabilities. Therefore, organisations need solutions and platforms that efficiently collect and present data in a way that is both accessible and retrievable by the AI agents.
Understanding how Agentic AI works
Agentic AI is designed to operate autonomously, and the complexity of the tasks handled by agents often requires access to dynamic external sources. As a result, the risk of something going wrong is increased.
For example, a staff member may trust a chatbot to provide them with an update on the status of a claim or refund, but would they feel as trusting when giving an AI agent their credit card details to book a flight?
Task-based agents can also plan and change actions depending on the context they’re given. They delegate subtasks to the various tools available through a process known as ‘chaining’. This is where the output of one action becomes the input for the next.
It means that queries (or tasks) can be broken down into smaller tasks, with each requiring access to data in real-time, and processed iteratively. The chain effect (in which decisions are made) is informed by the environment and the sources of data that are being monitored.
As a result, explainable and accurate data retrieval is required at each step of the chain for two reasons. Firstly, users need to know why the AI agent has made a particular decision and have visibility of the data source on which it is based. Secondly, they need to be able to optimise the process to get the best possible result each time.
The need to make reliable enterprise data available to agents is vital. This is why businesses are increasingly recognising the power of graph database technology for the broad range of retrieval strategies it offers.
How knowledge graphs unlock data
For Agentic AI to make informed, data-driven decisions, the underlying insights must be accurate, explainable, and grounded in context. These are capabilities that graph databases are purpose-built to deliver.
Gartner identifies knowledge graphs as an essential capability for GenAI applications, as GraphRAG (Retrieval Augmented Generation), where the retrieval path includes a knowledge graph, can vastly improve the accuracy of outputs.
The unique structure of knowledge graphs, made up of nodes and edges, is where higher-quality responses can be derived. Nodes represent existing entities in a graph (such as a person or place), and edges represent the relationship between those entities.
In this type of structure, the bigger and more complex the data, the more previously hidden insights can be revealed. These characteristics are invaluable in presenting the data in a way that makes it easier for AI agents to complete tasks in a more reliable and useful way.
Users have found that, with GraphRAG, not only are the answers more accurate, but they are also richer, more complete, and more useful. For example, an AI agent addressing customer service queries could offer a specific discounted package for a broadband service based on a comprehensive understanding of the customer.
As a result of using GraphRAG to connect disparate information about them, the agent can identify how long the customer has been with the company, the services they are currently using, and whether they have previously filed any complaints. A fragmented and dispersed view of the data could lead the agent to offer a discounted package when it was not due, leading to cost implications for the business.
Preparing for an agentic future
The capabilities of AI agents are shaped not only by the models themselves but also by the robustness of the data environment in which they operate.
It is vital, therefore, that organisations focus on ensuring their data is rich, interconnected, and contextually aware. This will unlock the full potential of the technology allowing agents to deliver the best possible benefits for Australian businesses.
