https://i.nextmedia.com.au/Assets/HederBG.mp4
If the eye-watering size of the cheques being written to build out AI infrastructure are anything to go by, Australia is reaching a turning point in its AI journey.
The major public cloud hyperscalers are doubling down on local capacity. In April, Microsoft announced a $25 billion investment to expand Azure capacity in Australia by more than 140 percent by the end of 2029, while AWS has committed $20 billion to expand Sydney and Melbourne data centre infrastructure.
CDC Data Centres said this month that it had signed Australia’s largest ever data centre deal to develop 555MW (megawatts of capacity) for a US tech company. That’s equal to around 40 percent of the total operating capacity of all data centres in Australia.
It all makes for good photo opportunities and ribbon-cutting ceremonies for politicians chasing economic gains and productivity improvements.
But it amounts to a multi-billion dollar bet that after several years of pilots, proof of concept activity, and rolling out AI copilots, Australian organisations are ready to scale AI and even become “AI-native” organisations.
Cloud is the infrastructure making that possible, supplying the computing capacity, data platforms and deployment environments needed to move AI from strategy into operations.
Five key takeaways:
- Australia’s AI infrastructure boom is well underway, with surging public cloud and Platform-as-a-Service (PaaS) spend, hybrid-by-default infrastructure, and strong hyperscaler investment.
- But only a minority of firms are yet realising transformative business impact from generative and agentic AI.
- The core blockers are legacy architectures, fragmented and low-quality data, immature governance, and skills shortages, which drive high AI pilot failure rates and stalled GenAI programs.
- Developing modern data platforms, containerisation, and robust DataOps and sovereignty-aware design is essential to lay the foundations for successful AI initiatives.
- Security and regulatory risk are rising sharply with cloud and AI adoption, pushing organisations toward Zero Trust, continuous cloud monitoring, and disciplined operating models.
https://i.nextmedia.com.au/Assets/20260601122840_Cloud_4.mp4
The cloud as scalable AI infrastructure
Australian organisations are forecast to spend more than $33.6 billion on public cloud services in 2026, a healthy 17.9 percent bump year on year, according to Gartner. Infrastructure-as-a-Service is growing fastest, rising 24.1 percent to $7.1 billion, driven largely by demand for GPU-intensive AI workloads.
Yet infrastructure investment doesn’t automatically translate into AI success. Deloitte’s State of AI in the Enterprise report found only 12 percent of Australian leaders say generative AI is already transforming their business, compared with 25 percent globally. Australia’s AI ambition remains strong, but enterprise readiness is still uneven.
Hybrid is the sweet spot
In 2026, cloud infrastructure is no longer separate from AI strategy. Organisations that once viewed cloud mainly as a cost-efficiency tool are now recognising it as the only practical way to scale AI, giving them access to graphics processing unit (GPU) capacity, elastic compute and managed services for model training, fine-tuning and inference.
“There’s a shift toward inference-optimised approaches as organisations fine-tune smaller domain-specific models instead of relying on larger general-purpose LLMs,” said Gartner’s Adrian Wong.
“Many are turning to hybrid cloud architectures to push this processing to the edge, which lowers cloud costs while still supporting automation at scale.”
The market is moving beyond experimentation toward real-time inference and agentic AI, with organisations relying more heavily on Platform-as-a-Service (PaaS) from the likes of Salesforce, Oracle, ServiceNow and ZenDesk to manage autonomous workflows and embed AI in business applications.
PaaS spending in Australia is forecast to grow 20.9 percent in 2026 to almost $10 billion, said Gartner in its May update, underscoring where AI development and orchestration are now concentrated.
Hybrid infrastructure is now the default model. Rather than relying solely on public cloud, many organisations are combining on-premises systems for sensitive workloads, private cloud for compliance-heavy data and public cloud for burst AI capacity. Australia’s hybrid cloud market, valued at US$4.8 billion ($6.7 billion) in 2025, is projected to reach US$18.6 billion ($25.97 billion) by 2034, with AI and edge computing among the main growth drivers.
Hybrid cloud growth is particularly strong in manufacturing, logistics, and agriculture where edge-based applications of AI are common. But companies across the board are seeing the value of taking a hybrid approach when it comes to developing and deploying AI.
“We needed somewhere we could let AI agents act autonomously without data security exposure, and learn what guardrails were actually necessary before deploying agents into our cloud environment,” Ivan Wong, head of data & AI at funds manager LVP, told iTnews.
“That sequencing, on-prem first to build conviction and cloud second, has worked better than starting in the cloud and retrofitting controls.”
Real-time inference and agentic AI
Forget AI assistants and copilots. All the hype, and much of the development, is now centred on AI agents. IDC research found 41 percent of Australian organisations are already deploying agentic AI and another 50 percent plan to do so within six months.
The resulting productivity gains are now widely reported. When home loan lending platform Lendi Group implemented numerous agents to streamline loan application processes.
“For customers, you can walk out of an open home, upload a contract and within 60 to 70 seconds have a readable summary and property report,” said Lendi’s chief technology officer, Devesh Maheshwari.
“For brokers, the agentic funnel means richer data and far less back-and-forth. In some areas we’re seeing applications processed 60 percent faster, and in others up to 200 percent improvements in throughput.”
Gartner expects 40 percent of enterprise applications to include task-specific AI agents by the end of 2026, up from less than 5 percent in 2025.
At the same time, inference-optimised architectures are replacing brute-force approaches. Instead of relying on large general-purpose models for every use case, organisations are fine-tuning smaller, domain-specific models that reduce cost and latency. This is encouraging hybrid deployments in which inference runs closer to users or devices while cloud handles training and model management.
The public sector’s cloud-first push
It’s not just corporate Australia leveraging the cloud to enable AI. The public sector is also accelerating. The Australian Government’s Whole-of-Government Cloud Computing Policy takes effect on 1 July 2026 and creates a unified framework for cloud adoption across the public sector.
The government has also signed a five-year Microsoft volume sourcing agreement to support AI and cloud uptake, while the APS AI Plan 2025 establishes chief AI officers in each agency. AI could add up to $142 billion a year to Australia’s GDP by 2030 if adoption accelerates, according to ChatGPT maker OpenAI, which alongside Anthropic has established an Australian presence.
Investment appetite remains strong, but scrutiny is increasing. While 92 percent of Australian data leaders intend to increase GenAI investment, 98 percent say they face significant challenges proving business value, according to Informatica’s CDO Insights 2025 global research. The gap between pilot enthusiasm and production outcomes is now one of the defining issues, not just in Australia, but globally.
Modernising architectures to support AI workloads
A major operational challenge in 2026 is the condition of existing enterprise architecture. Legacy systems, siloed data and rigid application frameworks were not designed for AI workloads or the high volumes of unstructured data they depend on.
IDC and MongoDB research covering 200 Australian organisations found 58 percent say their architecture is too rigid, costly and slow for building new AI applications. Organisations that fail to address technical debt are likely to see AI initiative failure rates rise by 50 percent by 2027.
By contrast, companies running strategic modernisation programmes are generating nearly three times more digital revenue from digital channels, 68 percent versus 24 percent for mainstream peers.
That is driving demand for AI-ready data infrastructure. Data lakehouses are gaining ground as a foundation for enterprise AI, while containerisation is accelerating portability across on-premises, cloud and edge environments.
Andrew Burnet, CTO at a large Australian university, said his institution had evolved its data architecture into a tiered model specially designed for AI/ML activities.
“We… redesigned our AWS landing zones to support greater freedom of development whilst maintaining the required corporate security controls, and built an enterprise AI governance framework,” he said.
Support for AI is now the leading reason for database and application modernisation in Australia, cited by 45 percent of organisations. But 96 percent have experienced some form of failed modernisation effort, with siloed and poor-quality data the most common obstacle. That reinforces a core lesson of 2026: architecture modernisation is a prerequisite for AI success.
“I have the advantage of owning both the architecture and the implementation, so decisions get made on output quality rather than procurement cycles,” said Wong.
“That's a contrast to larger enterprises I've worked in, where existing partnerships, risk committees and governance layers slow things down considerably.”
Q&A with Avron Welgemoed, Strategic AI Advisor and AI Practice Lead, Fujifilm
AI has moved beyond the hype, but many organisations are still unsure how to turn it into measurable business value. In this Q&A, FUJIFILM Business Innovation Australia explores why scaling AI starts with clearer use cases, stronger data foundations and a more practical approach to Microsoft Copilot.
Read more.
https://i.nextmedia.com.au/Assets/Cloud_3.mp4
Preparing and securing data for AI and machine learning
If architecture is the skeleton of enterprise AI, data quality and governance are the control system. The readiness gap in Australian data estates is a key reason many AI pilots are not reaching production.
“The biggest challenge for us is data quality and consistency across disparate data sources,” said Burnet, a problem made worse by the organic growth of an older organisation.
“Data quality, lineage, shared definitions – they’re all part of the day‑to‑day rhythm now, because without that, you can’t really be an AI‑native organisation,” said Lendi’s Maheshwari.
Informatica’s survey found that in 70 percent of Australian organisations, fewer than half of generative AI pilots have successfully moved into production. As a result, 53 percent of Australian organisations, the highest proportion globally, have slowed or halted GenAI initiatives over the past year. The leading data-related challenges are data quality, ethical use of AI, and data privacy and protection. Eighty percent of Australian data leaders also identify data reliability as a major issue.
This is pushing organisations to treat data preparation as a dedicated AI workstream. That includes building data quality pipelines, master data management, data catalogues, lineage controls and governance policies that define how data can be used in training and inference.
Cloud providers are helping through managed data preparation, feature engineering and MLOps services that lower the engineering burden of building AI-ready pipelines. At the same time, data sovereignty remains a major Australian consideration.
“I think we'll see more hybrid setups going forward, where on-premise hardware running local models handles the confidential workloads and cloud is reserved for everything else,” said Wong.
Requirements around data residency and compliance shaped by the Privacy Act 1988, APRA CPS 234 and the Security of Critical Infrastructure Act, are highlighting the importance of design decisions for local cloud regions and in-country data processing.
Managing risk with cloud-native security
As cloud and AI adoption rise, so does the attack surface. Australian organisations are now dealing with AI-specific vulnerabilities on top of conventional cyber risk.
CyberCX’s Cloud & AI Security report, based on more than 230 Australian and New Zealand organisations, identified five main obstacles: pressure to use AI without adequate security planning, data security and privacy risks, skills shortages, identity and access complexity, and changing regulatory requirements.
“Security hasn’t gone away, it’s become more granular,” Maheshwari told iTnews.
“We now secure at the workflow and agent level.”
Shadow AI is another growing concern and regulators are sending stronger signals about data breaches.
“The governance question isn't whether to experiment, it's how to do it without creating data exposure or compliance issues,” Wong said.
“We try to make the safe path the easy path, so people aren't tempted to route around it.”
In February, a Federal Court ordered FIIG Securities to pay a $2.5 million penalty for cybersecurity failures after sensitive data was stolen and leaked onto the dark web. The ACCC has warned that companies remain responsible for the outputs of AI systems and chatbots.
The cloud-native security response centres on Zero Trust architecture, Cloud Security Posture Management (CSPM) and AI-enabled monitoring. For organisations running AI pipelines against sensitive data, these are becoming baseline operational requirements rather than optional controls.
Q&A Paul Heaton, Co-Founder and CEO at Cubesys
AI success is no longer just about choosing the right tool. In this Q&A, Cubesys explains why organisations need to rethink how workflows, where AI can act as a genuine partner, and how governance, adoption and data foundations must come together before AI can scale.
Read more.
https://i.nextmedia.com.au/Assets/Cloud_2.mp4
AI on the edge for real-time insights
Cloud and edge computing are becoming increasingly interdependent in Australian AI deployments. Cloud provides the environment for model training, orchestration and management, while edge delivers low-latency inference where real-world operations happen.
Across mines, farms, factories and logistics warehouses, sensors generate large volumes of data in remote or latency-sensitive settings. Processing everything in the cloud is often impractical, so edge nodes analyse data locally and send only results or model updates back to central systems.
The rollout of 5G in major cities has also helping enterprises deploy edge nodes closer to users and connected devices. Hybrid cloud is becoming the practical model that ties these layers together, with edge supporting inference and cloud handling training, analytics and lifecycle management.
The hardest parts – finding talent and change management
The defining challenge for Australian organisations in 2026 is not access to AI technology but execution maturity. We have work to do on that front.
Poor-quality data or the inadequacies of legacy systems are issues that can be dealt with. But the bigger constraint in Australia is talent, Wong said.
“There's a real shortage of people who can actually build and deploy production AI systems, and that bottleneck shows up in more conversations with CTOs and CIOs than data quality does.”
Cloud platforms have made advanced AI capabilities widely available, but the differentiator is whether organisations have the architecture, data, governance and security needed to deploy them reliably at scale.
Those that align cloud strategy with AI strategy, modernise data platforms, adopt hybrid infrastructure and build secure-by-design deployments are best placed to capture AI-driven productivity and revenue gains. Those that move ahead without those foundations risk more stalled pilots, governance failures and rising infrastructure costs.
The Government, tech vendors and economists all agree that AI will prove to be a massive boon for productivity and economic development. But the evidence on the ground suggests that there’s plenty of work ahead to secure that rosy picture of an AI-augmented future.
Q&A with Aaron Cunnington, Managing Director at Antares Solutions
AI pilots are easy to start, but much harder to scale. The real challenge is turning experimentation into secure, practical tools that change how people work and deliver measurable value.
Read more
https://i.nextmedia.com.au/Assets/Cloud_5.mp4
CASE STUDY: Lendi Group
Lendi Group, the Australian home loan platform formed from the merger of Lendi and Aussie, has spent the past few years turning a traditional broking operation into an AI-native, end-to-end property and finance platform.
Founded around 14 years ago to digitise what was a mostly manual broker experience, Lendi now integrates with more than 30 lenders plus its own white-label products, and supports a national network of around 1,300 mortgage brokers.
Its strategic expansion has gone hand in hand with a deep re-architecture of its cloud and data foundations to support a new generation of AI agents.
When CTO Devesh Maheshwari joined a little over two years ago, the brief from Lendi Group’s founders was to become an AI-first company.
“That meant we had to stop thinking in terms of one-off AI use cases and start designing production architecture that could safely scale AI across the business,” he told iTnews.
The Lendi–Aussie merger had left a patchwork of microservices, databases and integration patterns, which was fine for traditional workloads but a bottleneck for AI agents that need to reason over consistent, connected data.
Building an AI-ready cloud and data spine
Lendi Group now runs almost entirely on AWS, with 96–97 percent of workloads there and only minor use of other public cloud platforms. Rather than betting everything on a single, managed AI stack such as Bedrock, the team has built its own agent orchestration layer using the Amazon Elastic Kubernetes Service, and open source frameworks, with a strong focus on avoiding lock-in.
Maheshwari admits that Lendi was taking a bet in doing so.
“We borrowed the AWS principle around one-way door and two-way door decisions and were very deliberate about understanding what decisions we had to make and in applying the first principles of technology,” he explained.
A self-hosted LiteLLM proxy allows Lendi to route traffic to different model providers – Bedrock, Anthropic, Google’s Vertex AI or OpenAI – and even fail over between them within seconds when capacity is constrained. That “ultimate flexibility”, as Maheshwari calls it, is designed so the architecture can be deployed into new markets, without re-engineering core AI services.
On the data side, Maheshwari’s team has been consolidating and standardising what was previously a “fragmented” environment. Lendi has moved to MongoDB as an operational data layer that can represent complex lending and property objects closely aligned to business processes, while streaming into a Databricks lakehouse for analytics and AI.
Lendi’s data backbone now supports 14.5 million property records and enables creation of “digital twins” of each customer, used to model the impact of interest rate changes, equity shifts and future scenarios in near real time via the mobile app.
Agentic AI at scale – without the bill shock
Lendi’s AI initiatives range from customer-facing experiences to internal productivity tools. An agent-driven “Guardian” experience guides customers through the home loan process, orchestrating a swarm of agents behind the scenes to collect richer, more complete application data. Applications coming through this agentic funnel are processed up to 60 percent faster. On some metrics, Lendi is seeing up to 200% improvement in processing efficiency.
Brokers who previously handled three applications a day can now complete six or seven, because follow-ups and data gaps have been dramatically reduced. Importantly, brokers remain “in the loop” at the tail end, reviewing high-quality, AI-prepared files rather than being disintermediated.
Another flagship use case is contract-of-sale analysis. Customers or brokers can upload a complex, mixed-format contract and, within about a minute, receive a plain-language summary and a property report that outlines key risks and negotiating points. Maheshwari describes the initial temptation to solve this with GPU-heavy infrastructure, before a cost–benefit spike showed it would not scale economically for a free customer service.
The team went back to the drawing board and redesigned the architecture.
“With the contract-of-sale analysis we walked away from that design, spent three more weeks rethinking the architecture, and ended up with the same outcome at roughly one-tenth of the cost,” Maheshwari said.
Guardrails, risk and operating model
Because Lendi deals with sensitive financial data and regulated credit advice, security and compliance have had to evolve alongside AI. Traditional cloud controls around identity, access, misconfiguration, and data loss remain, but Maheshwari’s team has added finer-grained controls at the workflow and agent level. Agents inherit role-based permissions so that the same agent behaves differently depending on who invokes it, limiting potential damage from rogue access or misconfiguration. Centralised evaluation guardrails and an “LLM-as-a-judge” evaluation framework continuously score conversations in near real time.
“If it looks like the agent is breaching its boundaries, the guardrails kick in,” Maheshwari said. A risk team member is embedded into the AI design process, making “compliance by design” part of the operating rhythm rather than an after-the-fact audit.
Maheshwari is clear that the transformation is only “about a quarter to halfway” complete. But by treating AI enablement as an architecture and operating model problem, not just a tooling problem, Lendi Group has laid a cloud and data foundation that can support continuous AI experimentation without losing sight of cost, compliance or customer outcomes.
Maheshwari concluded: “Ultimately, this is not just a technology exercise. It’s an architecture and operating‑model challenge. Data quality, lineage, shared definitions – they’re all part of the day‑to‑day rhythm now, because without that, you can’t really be an AI‑native organisation.”
Lendi Group
- $80 billion mortgage book
- 600 head office staff and 1,300 brokers
- Formed from the merger of Lendi and Aussie Home Loans
https://i.nextmedia.com.au/Assets/Cloud_1_1.mp4
CASE STUDY: LVP
LVP, a sustainability-focussed investment fund with $1.6 billion in assets under management, is re‑engineering its investment process around AI – but only after rebuilding its cloud and data foundations.
Ivan Wong, Head of Data & AI, said LVP, which has invested in everything from cleantech startups to affordable housing schemes, is “moving aggressively on AI”.
“We see it as a way to free up our people for higher-value work, and we think organisations that haven’t deeply integrated AI into their workflows over the next two to three years face an existential problem,” he told iTnews.
AI agents are used to support LVP’s investment team by “automating research, surfacing context on past deals, and helping analysts produce first-draft memos,” said Wong.
“This isn't just AI as a coding assistant for engineers. It's first-draft financial models, presentation decks, and legal contracts for every knowledge worker. The underlying architecture uses Claude via Anthropic's Application Programming Interface (API), with an agentic orchestration layer that I've built in-house to connect our data and tools via a self-hosted Model Context Protocol (MCP) server,” Wong added.
That stance also feeds directly into LVP’s investment decisions: “We want to see management teams with a coherent AI strategy, not just AI talking points,” Wong added.
OpenClaw, on‑prem first, then cloud
A central architectural choice at LVP has been how to let AI agents act autonomously without creating new security risks.
“The biggest shift this year was standing up an on-prem sandbox after the OpenClaw release in February,” Wong explained.
“We needed somewhere we could let AI agents act autonomously without data security exposure, and learn what guardrails were actually necessary before deploying agents into our cloud environment.”
In his view, sequencing “on-prem first to build conviction and cloud second”, has worked better than starting in the cloud and “retrofitting controls”.
This hybrid stance is also how LVP is thinking about sovereignty and confidential workloads.
“This is tricky if everything you run sits on cloud, which is part of why we went on-prem with OpenClaw,” Wong said.
He quickly saw the potential of the open-source AI agent upon its release in February and set about experimenting to explore its potential to automate workflows at LVP.
“I think we’ll see more hybrid setups going forward, where on-premise hardware running local models handles the confidential workloads and cloud is reserved for everything else.”
The difficult part then becomes a skills issue.
“The question that creates is whether your workforce has the capability to actually secure that hardware and your internal network. That’s where most organisations will get stuck,” Wong said.
Standardising on Claude, keeping the stack flexible
On the model side, LVP has narrowed in on a preferred large language model – Anthropic’s Claude.
“We benchmarked it against other frontier models across multiple use cases and Claude consistently came out ahead for our work,” said Wong.
At the same time, LVP is avoiding model lock‑in: “Our orchestration layer and agents are model-agnostic by design, so if another provider pulls ahead, or Claude’s performance doesn’t hold up, we can switch without rebuilding.”
Wong also directly links architecture control to AI economics.
“I have the advantage of owning both the architecture and the implementation, so decisions get made on output quality rather than procurement cycles,” he noted, contrasting this with larger enterprises where “existing partnerships, risk committees and governance layers slow things down considerably”.
His view on cost is blunt: “If you’re optimising for revenue or outcomes, compromising on model quality to save on inference cost is a false economy.” He added that many failed AI pilots come from “using cheaper, less capable models to hit a budget target, and then wondering why the outputs don’t justify the spend. The cost question is real, but it’s downstream of getting the model choice right first.”
People, security and governance
For LVP, the biggest AI challenge is not data plumbing.
“The hardest part isn’t technical, it’s the human transformation,” Wong said.
“People are naturally hesitant to change how they work, and getting non-technical teams to move from familiar BAU processes to AI-driven ones is harder than any data quality or integration problem I’ve dealt with.”
With enough in‑house AI capability, “the change management work is what’ll actually slow you down.”
Security and vendor risk are also front of mind as cloud platforms and SaaS products add AI features.
“The category that worries me most is AI features being bolted onto existing SaaS products and new AI-native startups shipping fast without mature security practices,” he said.
“The attack surface from these integrations is often wider than the vendor has matured into. We’re spending a lot more time on vendor due diligence as a result, auditing the underlying tech rather than taking marketing claims at face value.”
On governance, Wong’s position is that standing still is not an option. “If you’re not experimenting with AI, you’re falling behind people who are,” he said, even if “Australian-market disruption is slower to land.”
For most roles, “the governance question isn’t whether to experiment, it’s how to do it without creating data exposure or compliance issues.”
LVP’s response is to “make the safe path the easy path, so people aren’t tempted to route around it.”
LVP
- Sydney-based with 36 staff
- $1.6 billion under management
- Manages private equity, growth, and real assets funds
https://i.nextmedia.com.au/Assets/Cloud_2.mp4
CASE STUDY: Australian University
How does a university with tens of thousands of students move at speed to deliver AI services that enable innovation in education delivery?
For Andrew Burnet, chief technology officer at a large Australian university and an experienced digital transformation leader with previous senior roles at Virgin Australia and Origin Energy, it’s all about architecture, high-quality data, and leveraging cloud platforms in smart ways.
Guided by Burnet, the university is using hyperscale cloud platforms as the foundation for governed, production-grade services.
“We have evolved our data architecture into a tiered model specially designed for AI/ML activities, redesigned our AWS landing zones to support greater freedom of development whilst maintaining the required corporate security controls, and built an enterprise AI governance framework,” Burnet said.
The challenge: maintaining consistency and cohesion
That combination is designed to give teams room to experiment while ensuring enterprise standards are maintained. Yet the messiness of large institutions remains a challenge.
“What hasn’t worked and remains the biggest challenge of managing a complex IT landscape, is maintaining a sense of consistency and technology cohesion as new pockets of AI appear consciously and through product evolution and/or acquisition,” Burnet said.
The university is also trying to keep AI costs from running ahead of value. Rather than letting model usage expand unchecked, it has introduced fair-use policies and governance platforms to manage access to different models.
Burnet said this has been coupled with the introduction of AI FinOps - managing and optimising the running of AI models and infrastructure.
“This cost transparency drives conversations on specific value, benefits and prioritisation across the organisation,” he told iTnews.
A hybrid platform model
In practice, that means AI investment is being assessed as a strategic portfolio decision, not just another line item in the cloud bill. A hybrid platform model has emerged. Some AI capabilities are centralised, but faculties and teams still need flexibility to pursue specialised use cases.
“Our organisation will always have both due to the nature of the workloads,” Burnet said. The priority, he added, is “foundational governance and secure enabling infrastructure and platforms that allow freedom without compromising data protection or our core cyber posture.”
One teaching and learning project shows what this approach can deliver. A university-wide AI platform for learning and teaching built in collaboration with other prominent Australian academic institutions connects learning management system data to Azure-hosted AI agents via Learning Tools Interoperability (LTI) integration.
The data moves through “structured pipelines with content filtering, moderation controls and data loss prevention (DLP) scans within the platform,” said Burnet. The result has been “decreased student inquiries to academics”.
Data readiness remains the hardest problem. Burnet said the university faces the same issues as many large institutions, beginning with “data discovery, then governance and ownership, which then builds into context and data quality.”
“Data sovereignty is not optional for us,” he said. Every new system must be designed to ensure the university is “not pushing data offshore.”
Cloud-native security
Meanwhile, security is evolving. Burnet said the institution’s security stack is adapting to “AI emerging everywhere”, but the response still starts with fundamentals: vulnerability management, secure-by-design architecture, data protection and strong network controls.
“Cloud-native security for us is grounded in zero trust architecture with the appropriate network segmentation and management,” he said, with policy-as-code part of an ongoing maturity uplift”.
The university’s goal is not to slow experimentation, but to make it safer and more consistent.
“Our business is built on experimentation,” Burnet concluded.
“Therefore, everything we implement needs to support a frictionless end-user experience, whilst ensuring the appropriate governance and guardrails exist.”
