The Year of the Fire Horse: signals change and action for AI’s next chapter?
By Jay Tuseth (pictured), Vice President and General Manager, Asia Pacific & Japan, Nutanix
The Lunar New Year has commenced, bringing in the year of the Fire Horse. It’s the first in 60 years and it’s traditionally thought to bring rapid change, action, boldness, and freedom.
If the last few years have taught us anything, it’s to expect the unexpected. Since the uptake of Artificial Intelligence (AI) has accelerated, the world feels like it’s moving faster than ever before.
In 2025, AI moved from possibility to priority almost overnight. What began as some enterprises tentatively experimenting with AI, soon shifted to a tool embedded in most day-to-day operations.
The downside of pace is the risk of moving too quickly for the foundations and infrastructure to keep up. Time and again, AI pilots have exposed gaps in infrastructure, operations, and governance.
The Year of the Fire Horse will force decision makers to look hard truths in the eye, and they will need to make a choice. Do you sit back and let the whirlwind of AI dictate your risk profile, or do you brace for change by building the infrastructure needed to turn AI pilots into sustained momentum?
Some may argue the change has already arrived – and they’d be right. In Australia, the latest Nutanix ECI report found 61 per cent of local organisations were actively implementing a GenAI strategy, yet 83 per cent said their current infrastructure was not fit to support these workloads.
Cart, meet Fire Horse.
Breaking in AI: From pilot to practical
In the past year, the approach to AI followed more of an “act first, think later” model. This year, enterprises will start to rein themselves in, entering a more disciplined phase of AI adoption.
Breaking in AI requires a shift from proof of concept to operational practicality. Rather than questioning what it can achieve, we should focus on whether it can be sustained without compromising core business functions.
For a smooth transition, enterprises must prioritise use cases with clear business outcomes. After all, you can only achieve goals if you know what they are. Enterprises should seek consistency in the deployment phase, which will result in a steadier, more reliable shift into production.
This means creating workloads with scalability as the foundation. Workloads need to be portable, reliable, and easy to manage across environments. Without that consistency, even the most promising AI initiatives can run amok once they pick up speed.
Technologies such as containerisation can provide control, reducing friction and allowing AI services to scale without constant re-engineering. While AI has previously been measured in speed, success is now measured in endurance.
Juggling jurisdictions
As AI becomes woven into operations, infrastructure strategies spread accordingly, following data rather than adhering to a singular stream. Enterprises must now perform a juggling act between enterprise jurisdictions – public cloud, private data centres, and the edge – to meet demands.
So far, training has stayed cloud-centric, yet evidence suggests AI inference benefits from environments closer to where data is generated. Predictable costs, lower latency, and tighter governance are pushing more AI workloads toward on-premises and edge deployments. For regulated and real-time use cases, this is particularly true.
The edge is not far-off pasture. It’s a sovereign layer of enterprise infrastructure – one that is centrally managed, yet locally autonomous – capable of supporting even the most complicated AI programs, while meeting the new age of data residency concerns and regulatory requirements.
Durability through maturity
Durability is the goal for any new project, but of course, it’s easier said than done. While initial deployment might be more exciting, maintenance of AI services over time and across multiple environments requires far more effort. Model refreshes, security updates, compliance controls, and coordination across teams and locations should be embedded in daily operations.
To learn from the lessons of the Fire Horse, to embrace change, operational stamina matters. In the enterprise world, this means a unified foundation that delivers flexibility, consistency, and strength across environments. As a result, platform architecture is emerging as one of the most critical choices IT leaders have to make. Cloud-native, modular designs enable teams to adapt to change by letting services evolve independently, without disrupting the wider system.
Orchestration platforms create a unified operating model across hybrid environments, allowing AI and traditional applications to run side by side instead of being managed in isolation.
AI reaches its potential when infrastructure is resilient, dependable and well-governed, this is when potential will shift to tangible value across the organisation. Only with reliable infrastructure will the promises of the AI evolution – productivity, automated decision making, and accelerated processes – be realised.
