As much as we don’t want to think or acknowledge it, we can spend as much time eating up our precious time and mental energy and putting ourselves in the boring, repetitive tasks that prevent us from focusing on true strategic work, but Agent AI has changed the equation and we find this technology is rapidly abolished in Australia and the new zealand. According to a survey by YouGov and Salesforce, 69% of ANZ C-Suite executives, who prioritize AI, will focus on implementing agent AI over the next 12 months, with 38% saying they already implementing technology.
Agent AI is considered by many as a new frontier in AI innovation. This is because these agents can automate boring or repeating processes without having to prompt directly from human users, allowing a wide range of possible applications to be opened. For example, AI agents can provide professional-level advice to their customers, perform financial or HR management tasks, perform complex data analytics, and perform potential use cases. However, to adopt AI agents safely and efficiently, ANZ and later organizations need to do more to ensure and optimize the data that drives the agent tools. Without strong data security and governance, agents will not function effectively or securely.
What is Agent AI? Set the record straight
What is an AI agent? Microsoft defines it as (A) automate and execute business processes, and (a) act as a digital colleague to assist or perform tasks on behalf of a user or team.” Meanwhile, Salesforce calls it “Automatic Intelligence (AI) the type of artificial intelligence (AI) that can work independently, make decisions, and perform tasks without human intervention,” while IBM calls it “an artificial intelligence system that can achieve certain goals with limited oversight.”
While these definitions may not be completely identical (and there have definitely been some healthy debates in the industry!), the core concepts are consistent. An AI agent is an AI system that can act intelligently and autonomously without direct, continuous prompts from humans. What really stands out from AI assistants like ChatGpt, Google Gemini, and Microsoft 365 Copilot is this autonomy and advanced reasoning power.
Think about it like this. The assistant will help you write. This opens up a world of possibilities, including expert-level customer advice, automated management tasks for finance or HR, or performing complex data analytics on its own. For example, this week I asked AI agents to compile a report to compare the functionality of software products with international standards and provide suggestions for additional functionality. This saved about three days of research and allowed me to spend that precious time analyzing the results.
The reason for enhancing data governance is to bring safer, more secure AI agents
Agent AI has unique benefits, but also presents unique risks, and as more organizations adopt Agent AI, they discover robust data governance for policies, roles, and protections to manage and protect an organization’s data assets. Therefore, a recent survey from Drexel University shows that 71% of organizations have data governance programs, compared to 60% in 2023.
Effective governance is increasing as it helps address critical AI-related security and productivity issues, such as preventing data breaches and reducing AI-related errors. Without a strong data governance measure, agents may incorrectly disclose sensitive information or make autonomous decisions. Powerful data governance measures allow organizations to actively protect their data by implementing comprehensive governance policies and deploying technology to monitor AI runtime environments. This not only enhances security, but also ensures that agent AI tools work best and provide significant value with minimal risk.
The key elements of this approach are:
• Protect data without humans in loops: Agents rely on the data they consume, and often do not have humans in the mix so that data is consumed and distributed correctly. This means it is important to accurately classify this data, ensure relevance and mitigate risk. If humans are not in the loop, powerful data governance measures can intervene to enable AI agents to access or repeat sensitive data.
• Preventing errors and violations: A robust governance framework helps agents avoid “hatography” (AI generates misinformation) and protect sensitive content from accidental exposure by improving the quality of AI data. This significantly reduces the likelihood that autonomous agents will make harmful decisions.
To tackle these and other AI-related challenges, Gartner recommends that organizations apply the AI TRISM (trust, risk, and security management) framework to their data environments. Data and information governance, along with AI governance and AI runtime inspection and enforcement technology, are an important part of this framework. The very existence of this new framework highlights the immeasurable risks as well as the immeasurable possibilities of agent AI.
Securing the future with AI
The future of work is here, and it is driven by Agent AI. While the wave of adoption is clearly built across the ANZ, organizations need to prioritize robust data security and governance. This is not just about managing risk. These powerful tools are to optimize data to facilitate and ensure that they work effectively and safely. Organizations cannot afford to be left behind, so more needs to be done to ensure risk is managed and this powerful tool is effective.
