As AI adoption accelerates across Asia Pacific, Adhil Badat, Managing Director, Asia Pacific and Japan, Rackspace Technology, says businesses must balance innovation with responsibility by embedding governance, transparency, and human oversight into every stage of AI deployment.
As AI solidifies its role at the cornerstone of Digital Transformation across Asia Pacific, the race to implement AI solutions must be balanced with the commitment to responsible governance.
The development of AI initiatives remains a key focus for IT teams, with 88% of Singapore respondents reporting that they have taken steps to integrate AI and cloud strategies in Rackspace Technology’s 2025 State of Cloud Report. When asked about the business outcomes driving the integration of AI and cloud, 49% reported that their goal is to enhance operational efficiency while 40% are leveraging AI to improve the insights generated by data analytics.
Another Rackspace Technology report on AI indicated that 75% of global respondents would unconditionally trust AI-generated answers, but only 20% believe these outputs should always involve human validation. Nearly 40% expressed concern that their organisations lack adequate safeguards to ensure responsible AI use. These disparities underscore a critical gap between confidence in AI capabilities and the implementation of necessary precautions against potential misuse.
Below are some essential guardrails organisations can establish to help ensure that AI solutions are managed appropriately and leveraged responsibly.
Human oversight is crucial
Despite advancements in AI, particularly in large language models (LLMs), challenges remain—especially in complex, industry-specific use cases. In sectors like healthcare and finance, where accuracy and compliance are paramount, substantial human oversight is indispensable. For instance, Explainable AI (XAI) is now a regulatory expectation. The Monetary Authority of Singapore (MAS) has established the FEAT principles (Fairness, Ethics, Accountability and Transparency) to guide the responsible use of AI in the financial sector. These principles emphasise the importance of AI systems being transparent and explainable. Additionally, MAS’s Artificial Intelligence Model Risk Management (AI MRM) paper outlines best practices for managing AI-related risks, highlighting the need for robust governance and model explainability.
While AI, including agentic AI applications, can boost productivity of various tasks, truly autonomous AI systems are still an exception rather than the norm, highlighting the need for human oversight. IDC predicts that by 2027, half of IT buyers in Asia Pacific are expected to work exclusively with vendors that meet social, environmental and governance-related responsible AI criteria, underscoring the increasing importance of ethical AI practices.
Promote transparency
As AI deployment extends into core business areas including marketing, sales, HR, finance and engineering, transparency becomes essential. Singapore has taken a proactive stance with the launch of AI Verify, one of the first testing toolkits and governance frameworks for responsible AI in the region. Developed by the Infocomm Media Development Authority (IMDA), AI Verify provides companies with a structured way to validate and demonstrate that their AI systems are fair, explainable and aligned with ethical guidelines.
Effective AI management requires clear procedures, compliance with legal requirements, and transparency around data usage and privacy implications. Equally important are accountability and openness supported by proper mechanisms to audit AI decisions, assess impact on stakeholders and report issues and complaints.
Ensure accountability and data integrity With the rapid adoption of generative AI, concerns around copyright violations and unauthorised data use have grown significantly. To build trust, AI systems should incorporate rigorous data validation processes. This includes verifying data sources, maintaining data provenance transparency and ensuring data used in training and inferencing is accurately tagged and managed.
Address hallucinations, data validation and security
Most people have heard the term AI hallucinations, where LLMs generate responses based on non-existent data. These incidents raise concerns about the protocols followed by companies building AI models, underscoring the importance of reinforcing validation procedures.
Many LLMs now provide options to trace the lineage and sources of their data. As more AI projects are implemented, it is essential to incorporate thorough data validation into each solution to guarantee accuracy of outcomes. This includes implementing robust guardrails, tagging datasets with appropriate metadata for use during inference and creating processes for generative AI workloads.
A key challenge in AI adoption is that security issues arise from AI models’ complexity and the vast data they handle. These concerns necessitate advanced security protocols and enhanced threat detection mechanisms. The lack of skilled personnel exacerbates these challenges, underscoring the urgent need for upskilling initiatives to close the knowledge gap in AI technologies.
Securing the future of AI with robust guardrails
As organisations continue to adopt AI for more use cases, the need for robust, well-defined guardrails has never been more critical. It is crucial to maintain human oversight, promote transparency and ensure accountability and data integrity across all AI deployments. These measures are not just best practices; they are essential safeguards that protect organisations and the public from the potential pitfalls of unverified AI.
