I believe that artificial intelligence has reached its inflection point. If 2024 was the year of AI adoption, 2025 has been the year of acceleration. Now comes the next chapter…autonomy.
AI is moving beyond copilots and chat interfaces into what analysts call agentic transformation: intelligent systems that can reason, decide, and act on our behalf. It’s an extraordinary opportunity and a growing risk. Because the future of AI won’t be defined by how fast organisations adopt it, but by how responsibly they govern it.
From Automation To Autonomy
For the last decade, digital transformation has been about automation: streamlining processes, reducing manual effort, integrating systems. Agentic transformation is different. It embeds adaptive intelligence into those same workflows so decisions happen in real time.
According to IDC Asia/Pacific, nearly 70% of enterprises in Asia-Pacific expect agentic AI to reshape their business models within 18 months.1 That shift is particularly visible across Southeast Asia, where industries from financial services in Singapore to public healthcare in Malaysia are already piloting agentic systems to automate decisions and workflows.
In Australia and New Zealand, boards are also leaning in. Industry research suggests that a majority of ASX 200 companies now include AI and technology as recurring board agenda items. This is a stark shift from just three years ago, when most treated AI as experimental.
These systems are no longer tools; they’re teammates. But autonomy without oversight isn’t efficiency, it’s exposure. Governance is what stands between a future of chaos and a future of sustainable, AI-driven growth.
The Governance Gap Becomes an Enterprise Risk
A report by Boomi found that only 2% of AI agents deployed today are fully accountable under always-on governance, while 80% of organisations admit they lack full control.2 That’s a governance gap waiting to erupt and it’s most acute in Asia-Pacific, where regulation and maturity are advancing unevenly. Singapore leads the region with initiatives such as AI Verify, while Malaysia’s AIGE Guidelines have become a model for responsible AI too.
Meanwhile, MIT Project NANDA’s recent report revealed that 95% of enterprises see no ROI from AI, not because their models are weak, but because their data foundations and integration are.3 AI fails when it’s disconnected, when models operate on incomplete, siloed, or ungoverned data. As I’ve said in previous discussions, “AI doesn’t fail because of bad models. It fails because of bad data, poor integration, and a lack of trust.”
Three Pillars of Trust for the Agentic Era
Across the many AI use cases I’ve discussed with business and technology leaders this year, success comes down to three imperatives: visibility, control, and guardrails.
Visibility
You can’t govern what you can’t see. Organisations need a complete inventory of their AI agents including what data they access, what actions they take, and how they behave. Governance platforms like Boomi’s Agent Control Tower make that visibility possible, providing observability and early intervention when agent behaviour drifts from policy.
Control
Governance isn’t bureaucracy; it’s dynamic oversight. Role-based access, audit trails, and kill-switch capabilities ensure that autonomy stays accountable. As Steve Lucas writes in his book Digital Impact, “Automation that isn’t integrated equals a legacy company that will underperform.”4 Integration isn’t just technical plumbing; it’s the nervous system that keeps AI in check.
Guardrails
Operational guardrails define how far AI can go on its own. They prevent bias, enforce compliance, and ensure ethical behaviour. This is where policies meet platform — providers must combine architecture, governance, and human oversight to make autonomy safe by design.
Research from Deloitte shows that companies with mature governance structures see 28% higher AI adoption and 5% greater revenue growth.5 Yet across Asia-Pacific, nine in ten organisations remain at foundational governance maturity. Trust isn’t a compliance requirement, it’s a competitive advantage.
From AI Islands To Agentic Ecosystems
Across industries, AI is evolving into agent ecosystems with multiple agents collaborating across sales, finance, operations, and customer service. They’re already transforming healthcare decision-making, claims management, and supply chain optimisation.
In Asia-Pacific, we’re seeing this evolution most vividly in healthcare, insurance, and education. In Singapore, the Ministry of Health has committed S$200 million to scale AI across national healthcare systems. In Malaysia, insurers are deploying AI agents to accelerate claims verification. Across Australian universities, AI-driven integration is improving student visibility and retention. This is a reminder that every AI success begins with unified data.
But every new agent adds complexity. Without a unified data fabric and consistent governance, these agents can’t see the full picture. They make faster decisions, but not necessarily better ones. That’s why integration and automation have become the backbone of the agentic enterprise. As I’ve said previously in interviews, “APIs are the eyes and ears of AI. They give agents the context they need to make intelligent, trustworthy decisions.”
When organisations achieve that level of connected intelligence, the impact is immediate. According to a Total Economic Impact™ study by Forrester Consulting, commissioned by Boomi, organisations using the Boomi Enterprise Platform achieved a 347% ROI and payback in under six months6, proving that the right integration layer doesn’t just connect systems; it accelerates outcomes.
The Human Role in the Self-Driving Enterprise
Autonomy doesn’t mean absence of humans. It means elevated human oversight. The CIO of the future is becoming a kind of Chief People Officer for digital workers who would be responsible for hiring, training, and offboarding AI agents just like human employees. That idea might sound futuristic, but it’s already here. Many organizations are appointing a Head of Enterprise AI to lead governance, ethics, and internal activation.
Across Asia-Pacific, governments are also investing heavily in AI upskilling. From the Philippine National AI Upskilling Plan to Singapore’s SkillsFuture AI programs, ensuring that the next generation of workers is prepared to collaborate with intelligent systems.
As Steve Lucas reminds us in Digital Impact, “AI will make us superhuman but only if it’s built on a foundation of trust and truth.” The next generation of work will be defined not by human versus AI, but by human + AI, combining computational scale with empathy, context, and creativity.
From Readiness To Responsibility
In 2025, many organisations reached AI readiness: modern data stacks, pilot projects, and proof-of-concept ROI. The next step is responsibility by governing AI as a core enterprise discipline.
The winners of this next phase will be those who:
- Treat governance as a growth enabler, not a brake
- Build integrated, API-driven architectures that keep AI connected and contextual
- Maintain human-in/on-the-loop oversight for every critical decision
The result? AI that’s not just faster, but fairer and not just automated, but accountable.
The Road Ahead
The self-driving enterprise won’t arrive overnight. But the road to it is already visible: clean data, integrated systems, governed autonomy, and empowered people. Asia-Pacific is poised to lead this transformation. From Singapore’s various AI frameworks to Malaysia’s ethical AI initiatives and Australia’s renewed public-sector digital modernisation, the region is building the foundation for trusted autonomy.
Gartner also predicts that by 2028, enterprises using AI-governance platforms will achieve 30% higher customer-trust ratings, 25% better regulatory compliance, and 40% fewer AI-related ethical incidents than competitors.7 That’s not just an IT outcome; that’s a business differentiator. As Steve Lucas writes in Digital Impact, “Change only happens at the speed of trust.” That’s the real measure of AI maturity: not how fast it acts, but how confidently humans can trust it.
Read our Agentic Transformation Playbook for practical advice on steps to take to succeed in this new era of AI.
—–
Sources:
- IDC: Around 70% of Asia/Pacific Organizations Expect Agentic AI to Disrupt Business Models Within the Next 18 months, March 24, 2025
- “Navigating the AI Agent Governance Gap”, a study conducted by FT Longitude on behalf of Boomi
- “The GenAI Divide: STATE OF AI IN BUSINESS 2025”, MIT NANDA
- “Digital Impact: The Human Element of AI-Driven Transformation,” Steve Lucas
- Deloitte Access Economics: “AI at a crossroads: Building trust as the path to scale”, Deloitte Asia Pacific | AI Institute, December 2, 2024
- Forrester Consulting, “The Total Economic Impact™ Of The Boomi Enterprise Platform”, August 2025
- Gartner Research, “Signature Series: Top Strategic Predictions for 2026 and Beyond”, Daryl Plummer, September 9, 2025