How can the Blue Economy Leap-frog by Learning from Traditional Industries?

We explore the similarities and learnings from cross-industry to support the growth of the nascent Blue Economy in Australia & New Zealand

What are the opportunities for investment firms to selectively master key parts of the Data Centre value chain?

We consider the unique aspects of the Digital Infrastructure industry in Australia for selective investment

We view the rapid evolution of AI and particularly AgenticAI evolving towards industry vertical solutions, akin to the early days of ERP then SaaS solutions

Will industry vertical AgenticAI solutions embed itself as the next phase of AI evolution?

Disclaimer: Our White Papers have been written by our team with the help from AI for sourcing, statistical validation and researching examples. We strive for our White Papers to be no more than 20% AI generated thus ensuring authenticity and a focus on our own experience

Synopsis

We explore the similarities and learnings from cross-industry to support the growth of the nascent Blue Economy in our region of Australia, New Zealand and the Pacific Island Nations

Context

How can the Blue Economy Leap-frog by Learning from Traditional Industries ?

Our team includes technology and telecom industry “veterans” (essentially team members pushing towards or past 50 years old) that have seen the Blue Economy “movie” before (not Le Grand Bleu which is indeed a great movie by Luc Besson starring Jean Reno, 1988) – that is, the journey of Blue Economy start-ups and purpose-led organisations that tend to first focus on one ‘spot problem’ before broadening into a broader array of solutions and services

Within Australia, many begin with Blue Tech, being technology-led solutions to help solve specific use cases, such as Carnegie Clean Energy that has patented CETO which is a renewable wave energy technology. Others begin with AquaCulture solutions such as Freemantle Seaweed which, as the name suggests, farms seaweed for agricultural feed such as cattle feed and helps to reduce methane ‘output’ along the way. Yet others are ESG and sustainability focussed, such as Uluu which creates biodegradable plastic alternatives sourced from farmed seaweed

The Australian and New Zealand market has never been short of ideas nor entrepreneurs or start up founders – however, this White Paper explores how the lessons learned from other industries such as Telecoms, Technology and Network-based start-ups can help the nascent Blue Economy ‘leap-frog’ or more rapidly scale, grow and become a continuous sustainable business (in our region and globally)

Defining the Blue Economy in ANZ

Let’s begin with a brief definition of the Blue Economy (this portion is AI generated amalgamating the numerous definitions from respected global organisations in this domain). The Blue Economy generally refers to the sustainable use of ocean resources for economic growth, improved livelihoods and jobs, ways to reduce fossil fuel carbons, while preserving the health of ocean ecosystems

In the Australia, New Zealand and the Pacific Islands – the Blue Economy is strongly tied to the Pacific Ocean and Southern Ocean (Antarctic), with a focus on the following attributes:

• Sustainable resource management: fisheries, aquaculture, and the responsible extraction of marine resources

• Marine renewable energy: Exploring and developing offshore wind, wave, and tidal renewable energy sources

• Maritime transport and logistics: Ensuring sustainable shipping, passenger and freight ferry transport and port operations

• Coastal tourism: Promoting eco-friendly tourism that benefits local communities and protects marine environments

• Marine park protection: Lobbying and gaining government approval for marine parks and/or protected marine zones

• Marine biotechnology: Utilising marine organisms for pharmaceuticals, cosmetics, and other products

• Ocean conservation and research: Investing in scientific research to understand and protect marine ecosystems

• Climate change resilience: Addressing the impacts of climate change on coastal communities and marine environments, including sea-level rise and ocean acidification which impacts Pacific Nations more than most other nations

• Indigenous perspectives: Respecting and incorporating the traditional knowledge and practices of Indigenous peoples in marine resource management

Our region has unique attributes when it comes to the Blue Economy:

• Pacific Island nations: The Blue Economy is particularly crucial for the Pacific Island nations, where it represents a significant portion of their economies and livelihoods

• The Southern Ocean: The unique ecosystems of the Southern Ocean, including Antarctica, require careful and different management and conservation approaches to equatorial oceans and marine ecosystems

• Great Australian Reef Ecosystems: In Australia, the preservation of iconic ecosystems like the Great Barrier Reef is a central focus of the Blue Economy. More recently, the coral bleaching of Ningaloo Marine Park on the western (Indian Ocean) side of Australia has raised concerns about ocean heating across the country

• Aquaculture: Both Australia and New Zealand are expanding their aquaculture industries, with a focus on sustainable practices. This industry has been at the forefront of merging Blue Tech with Blue Economy sustainable practices

• Collaboration: There is a strong emphasis on regional collaboration and cooperation to address shared challenges and opportunities in the Blue Economy.Organisations such as the World Wildlife Fund (WWF), Pacific Islands Forum Fisheries Agency (FFA), the Blue Economy Cooperative Research Centre (CRC) based in Australia and technology companies like Ackama (New Zealand) are trying to facilitate cross border collaboration in their specific Blue Economy focus areas

Cross- Industry Learning

As we outlined from the top, Attain.ai has several team members with deep and multi-decade experience in the telecoms and technology industries. These more established industries have a mix of large corporates but also an established history of new businesses with better ideas, entrepreneurial start-up businesses and VC or PE backed disruptor businesses

We think its useful and insightful to consider the affinity or common attributes between the Blue Economy and networks-based businesses such as telecoms and technology. While some may argue that a networks-based industry is seemingly distinct, they do share several key attributes and rely on similar enablers and capabilities:

• Customised and Advanced Technology Dependence:

  • These industry sectors rely heavily on advanced technologies. The Blue Economy utilises technologies like underwater sensors, seafloor imaging, autonomous vehicles, and satellite imaging for monitoring and resource management. Similarly, the telecom industry is built on specialised network technologies, a mixture of IT and OT (operational technologies) to operate and maintain the business, a transition to open sourced systems and custom built solutions for operations

  • Similarities will exist across the infrastructure aspects of the Blue Economy (eg offshore floating wind farms, tidal energy, wave energy) and Telecoms (telecom towers, data centres and exchanges, etc) – in these areas, the Blue Economy can leap-frog with the lessons learned from Telecoms and Infrastructure solutions such as Digital Twin asset management and scenario plannin, geo-spatial solutions, IoT and operational technology/device based solutions, and most importantly, the efficiency gains from technology in the operate & maintain portion of the value chain since ocean based infrastructure is notoriously harder than land-based equivalent infrastructure

  • This shared reliance creates opportunities for collaboration in areas like data collection, analysis, and real/near-real time communication to operations centres, ‘field’ and assurance technicians, proactive and predictive infrastructure and facilities assurance or maintenance, AI and data-driven scenario planning based on a myriad of internal and external data points and many more shared use cases

• Data Collection and Analysis:

  • Both industry sectors require extensive data collection and analysis. The Blue Economy needs macro data on ocean conditions, marine life, and resource availability. It also needs micro data on the specific use cases of the Blue Economy venture such as water quality, coastal erosion trends over time, tidal strength and so on. The telecom and technology industries are the masters at managing data across multiple levels (macro to micro) and understanding causal versus correlated data points when building AI models for things such as predictive maintenance or inhibitors to network optimisation speed and capacity, traffic management, and user behaviour analysis

  • The Blue Economy has an opportunity to improve its understanding of data management and data use from these industries. This shared need drives innovation in data processing, data visibility for decision making, AI, and machine learning models

• Infrastructure Development and Investment Optimisation:

  • As alluded to above, these sectors require significant infrastructure investment. The Blue Economy needs infrastructure investment for mega projects such as offshore energy, aquaculture, and coastal protection. The telecom industry needs infrastructure for networks, data centers, and communication towers. A nascent and emerging Blue Economy doesn’t mean an immature approach to investment planning and development if it can learn from these other industries by finding the growth plays (e.g. Digital Infrastructure such as Data Centres, distribution and logistics facilities, and until recently things as boring but stable like telecom towers and leasing sites). In all these examples the demand far exceeds supply (even as new supply for things like Data Centres are constantly being built)

  • What are the areas of the Blue Economy that are growth plays with areas of demand far exceeding supply? Sustainable seafood with an equivalent of ‘certified organic’ (e.g; no pink dye being injected into salmon nor antibiotics) as well as practices that minimise environmental impact is in demand in our ANZ region. The jury remains out on marine renewable energy supplementing or replacing aging offshore wind turbines but as with all infrastructure as it scales and finds better cost inputs, there is a real possibility of the tidal and wave renewable energy sources becoming viable alternatives. There is a need to revisit the business cases and cost improvement trends of early offshore wind energy to leverage the lessons learned and speed up the viability and feasibility of these emerging marine energy sources

  • The same approach can be said for marine biotechnology and pharmaceuticals. They are emerging and not yet fully exploited in a sustainable way because the current costs remain high. But how to circumvent this with lessons from land-based bio technologies and their commercialisation processes?

  • Our team at Attain.ai have been advising and supporting numerous investment thesis and business cases over the years – for potential transactions, carve-outs and valuing start-up businesses. We strongly believe there can be more cross-industry learning and sharing to be gained by the Blue Economy industry in ANZ. The overlapping need for robust and reliable infrastructure, especially in marine, coastal and offshore environments is just the beginning of the overlap

• Remote Monitoring and Communication:

  • A core need in the Blue Economy is the ability to monitor remote ocean environments. This aligns directly with the telecom industry's expertise in providing reliable communication networks, monitored 24/7 by a network operations centre

  • Over the years, telecom organisations have become experts in highly secured, highly automated, highly accurate predictive assurance and monitoring of their assets and networks. Any small degradation in services, unusual movement in network speed or capacity, or alerts from sensors on large infrastructure assets will be immediately made visible, inspected and rectified. For example, automated correlation of alarms generates the automated creation of tickets of work for site inspection. These practices will help the Blue Economy to leap-frog to better practices in these operate and maintain, monitoring and data sharing domains

  • As it stands, multiple telecom network technologies (satellite, fibre fixed, fixed wireless and mobile cellular networks) enable real-time monitoring of marine activities, environmental conditions, and infrastructure, enhancing safety and efficiency. It stands to reason to leverage the learnings from telecoms in the remote monitoring domain as well

• Emphasis on Sustainability:

  • Increasingly, these industry sectors are focusing on sustainability. The Blue Economy aims to balance economic growth with environmental protection. The telecom industry is working to reduce its carbon footprint through greater renewable energy sources (especially for power hungry data centres) and promote responsible resource use. Technology organisations such as Microsoft promises the ‘100/100/0 ambition’ where 100% of their energy will be from 100% renewable sources and require them to purchase zero carbon credits

  • Collaboration across the industry needs to be promoted and accelerated. Would a business case for a renewable marine energy provider be improved if it had a large data centre anchor client from day zero rather than being drafted in isolation? Could there be shared real estate costs in the landing point for the energy provider with the Telecom owning the nearby Data Centre, or the communications infrastructure such as cables being bundled with the utility cables to the energy generation offshore site?

  • This shared focus pushes these industries to innovate collaboratively on topics such as greener (or bluer?) energy generation, shared technologies and shared infrastructure and real estate use

• Research and Development:

  • These industries require ongoing R&D investment to improve existing technologies and develop new ones. The approach to R&D by technology companies especially is very useful for the Blue Economy to understand. A technology company that pours every dollar of net margin back into R&D (it happens! Especially founder-led organisations) doesn’t have funds for rainy days or an unexpected dip in revenues. Finding the balance of R&D to beat competitors or be first to market versus funding contingencies and containing costs can be a very useful skill for remaining in the long game when a Blue Economy start-up business

  • The Blue Economy approach to research can involve third parties and science organisations, in the same way that telecom and tech companies can partner with third party hardware firms, software firms, chip makers and the like.Again, the management of R&D and other partners and alliances from the telecom and tech industries can benefit the Blue Economy to avoid mistakes such as loss of IP or future IP disagreements

Methods to launch new products

We argue that the area of single greatest learning benefit to the Blue Economy is the approach undertaken by telecoms and technology businesses to new product development, AI and data-led solutions and being rapid to market

We are speaking of course about use case origination, testing use cases for their viability (financial predominately) and feasibility (how implementable it may be predominately), service design methods to ensure its fit for the end customer or user – all supported by Agile-based methods that ensure speed to market, quick learning (fast fails) and other techniques to iterate and speed up the process from Proof of Concept into a scalable, fully operational (and revenue generating) product or service

We have recently been supporting some Blue Economy start-ups, less in the Blue Tech domain, more in the services, infrastructure and aquaculture domains – and there are some gains to be made in the robustness and fidelity of the use case-to-deployment approach and methods.Better practices in this space will ensure a better use of investment funds, a more likely chance of successful Proofs of Concept than failures and thus a better chance of product or services commercialisation

We believe greater cohesion and collaboration with specific cross-industry specialists from telecoms and technology backgrounds will benefit a Blue Economy start-up business immensely

These businesses can glean valuable insights from the successes and failures of start-ups in these other industries. Here are five key conclusions:

1. Importance of Cross-Industry Collaboration:

• Blue Economy Application:

  • Blue economy start-ups should prioritise building teams with diverse skill sets and backgrounds to address the complex challenges of sustainable ocean resource management

2. Emphasis on Scalability and Sustainable Business Models:

• Blue Economy Application:

  • Technology start-ups focus on creating scalable solutions that can reach a wide market AND combining technology with viable business models as the key criteria

  • The Blue Economy start-ups should adopt a similar approach, developing technologies and business models that can be readily replicated and expanded. This is crucial for ensuring the long-term viability of Blue Economy ventures and their ability to make a significant impact on ocean sustainability

3. Data-Driven Decision Making:

• Blue Economy Application:

  • The technology sector thrives on data analysis and informed decision-making. Blue Economy start-ups should leverage data approaches from technology companies to support their ocean monitoring, remote sensing, and other sources to assure their operations and assess their environmental impact

4. Navigating Regulatory Landscapes:

• Blue Economy Application:

  • The telecommunications industry is heavily regulated, and start-ups must navigate complex licensing and compliance requirements. Blue Economy start-ups will also face a range of regulations related to environmental protection and resource management. Early engagement with regulatory bodies and a proactive approach to compliance are essential

5. Importance of Early Investment and Funding:

• Blue Economy Application:

  • Securing adequate funding is a critical challenge for all technology start-ups. Blue Economy start-ups may require significant upfront investment in research and development, as well as infrastructure.The proven approach and networking methods for seeking out venture capital, government grants, and other funding sources is essential

Written by Andrew Lorken, Co-Founder and MD of Attain.ai Advisory Pty Ltd

Summary

Context

Summary of Australian Data Centre Market Growth

Our team have been working with Private Equity and Investment Firms throughout 2024 in the Data Centre domain, unsurprising given the rapid growth and the mega deals such as the Blackstone, along with the Canada Pension Plan Investment Board (CPP Investments), purchase of AirTrunk from Macquarie Asset Management and the Public Sector Pension Investment Board (PSP Investments) in September 2024. The deal was reportedly worth over A$24 billion.

There are numerous good quality papers on Data Centre growth and drivers, even with a focus on the Australian market. Our White Paper summarises the market overview for context however we pose a question that has not been analysed with the same degree of depth:

What is the value chain for Data Centres specifically in the Australian market, and what are the opportunities for investment firms to selectively master key parts of that value chain?

What are the opportunities for investment firms to selectively own key parts of the data centre value chain?

The Australian data centre market is experiencing significant growth, driven by the increasing demand for digital services and cloud computing. According to a December 2024 report by CBRE, "the digital economy is growing exponentially as there is a greater dependency on technology across all facets of our lives". This growth is expected to continue in the coming years, with the market projected to more than double in size by 2030.

The Australian data centre market is currently dominated by a few key Tier 1 players, including AirTrunk, Amazon Web Services, CDC (Canberra Data Centres), Microsoft, Equinix and NEXTDC. There are also a plethora of Tier 2 data centre operators in the fragmented Australian market such as Macquarie Telecom, Telstra, Fujitsu and Global Switch and many other Australian and global players.

These companies are investing heavily in new data centre capacity to meet the growing demand for their services. For example, AWS announced in August 2024 plans to invest A$13 billion over the next three years to expand its hyperscale cloud computing and AI infrastructure in Australia.

The growth of the Australian data centre market is being driven by several factors, including:

1. The increasing adoption of cloud computing by businesses and consumers

2. The growth of the digital economy

3. The increasing demand for data storage and processing

4. The development of new technologies with exponential compute and storage requirements, such as artificial intelligence and machine learning

   The first three factors typically drive Enterprise Co-Location Data Centres whilst the mega trend of the fourth factor, being AI, is driving the need for GPU-based large Data Centres with much greater computing power and storage.

The Drivers to Data Centre Growth

There are a number of key drivers that contribute to the growth of the Australian data centre market:

• Hyperscaler demand: There is a surge in demand for data centre capacity from hyperscale cloud providers like AWS, Microsoft, and Google. This is driven by factors such as:

  • Increasing cloud adoption: Businesses are increasingly migrating their IT infrastructure to the cloud.

  • Growing demand for AI: AI applications require significant computing power, which is driving demand for high-performance data centres. AI workloads are anticipated to contribute 20% absorption by 2028, based on global workload benchmarks.

  • AI workload is set to grow at a compound annual growth rate (CAGR) of ~33% over the next 5 years, compared to traditional workload's 8%.

  • Government investment: The Australian government is investing heavily in digital infrastructure, which is creating new opportunities for data centre operators.

• Edge computing: The growth of edge computing is also driving demand for data centres in Australia. Edge computing involves processing data closer to the source, which can improve performance and reduce latency.

• Data sovereignty: There is a growing trend for businesses to keep their data within Australia. This is due to concerns about data privacy and security.

• Sustainability: Data centre operators are increasingly focused on sustainability. This is due to both customer demand and government regulations.

What are our 10 Key Insights to Data Centre Growth in Australia?

When discussing the rapid growth with investment firms and data centre operators, we have found the 10 key insights regarding data centre growth in Australia can be summarised as:

1. Market Overview: The investment firms tend to classify the Australian Data Center market with Sydney and Melbourne identified as Tier 1 cities and other major cities like Brisbane and Perth as Tier 2. Regional cities, where there are also great opportunities are classed as Tier 3 sites. The overall national contracted utilization rate in 2024 is estimated at 85-90%.

2. AI-led growth: The increasing adoption of AI is expected to significantly boost the revenue growth of Hyperscale Data Centers in our region’s operational markets (Australia, Japan, Singapore, Malaysia), with the range across numerous market reports suggesting from 6% to 9% YoY.

3. AI Investment: AI-led investments are on the rise in Australia, mirroring trends in the US, although Australia is considered to be in the early stages of AI investment compared to the US.

4. Capacity Demand Drivers: The demand for Hyperscale Data Center capacity is primarily driven by cloud migration and the increasing need for GPUs and High-Performance Computing (HPC) racks to support AI workloads (typically 100KW per rack and greater).

5. Computing Power Demand: The demand for computing power is rapidly increasing, driven by AI and other emerging technologies.

6. Future Data Centers: Future Data Center design will be integrated computing clusters providing diversified computing power, greener energy solutions, and meeting the needs of latency-sensitive applications.

7. Computing Network: China and the US is promoting the concept of a "computing network" to connect Data Centers nationwide, requiring improvements in computing power transmission efficiency and high-bandwidth, low-latency connections. There will be implications for regional and edge data centres in Australia given the size of our country.

8. Key Technical Features: Future Data Centers will be characterised by diversity and ubiquity, security and intelligence, zero carbon and energy conservation, and flexible resources.

9. AI-ready Data Centers: There's a growing need for Data Centers that can handle AI workloads, which require higher power densities and more advanced cooling solutions. Within Australia, the retrofit and renovation of brownfield Co-Lo data centres to AI-ready, including racks capable of 100KW or more, liquid cooling and guaranteed high power - could be 80-90% of the cost for a greenfield AI data centre.

10. Acquisition Considerations: Key considerations for data centre acquirers include AI-ready facility infrastructure, deployment speed for new data centres (e.g. proprietary construction methods), CAPEX optimization, power optimisation, Hyperscaler pricing, and future Data Center service offerings.

The Key Challenges to Data Centre Growth

Despite the strong growth prospects for the Australian data centre market, there are a number of challenges that need to be addressed:

• Power: Data centres require a significant amount of power to operate. This can be a challenge in Australia, where there is already a shortage of power.

• Land: Data centres require a large amount of land. This is a challenge in Australia metro regions, where land is expensive and there are competing demands for land use.

• Skilled labor: The data centre industry requires a skilled workforce. There is currently a shortage of skilled labor in Australia.

• Sustainability: Data centres can have a significant environmental impact. The industry needs to find ways to reduce its environmental footprint.

• Approvals: The process for obtaining approvals for new data centres can be complex and time-consuming.

Opportunity Areas for Data Centre Investment in Australia

Despite the challenges, there are a number of significant opportunities for data centre investment in Australia:

• Hyperscale data centres: Globally, it is estimated that only 17-20% of data centres are AI capable with racks over 100KW. Hyperscale data centres will continue to be built in Australia as firms scout for the right land and site, access to power, utilities and connectivity. Computational Demand: The time taken for computational power to double has decreased from 21 months to 5 months in recent years.

• Edge data centres: Edge data centres are smaller data centres that are located closer to the end user. This can help to improve latency and performance. Given the size of Australia and vast distances between major cities, edge data centres and regional data centres can be an attractive investment.

• Co-location data centres: Co-location data centres provide space and infrastructure for businesses to house their own servers. These traditional data centres will continue to grow especially in Tier 2 and Tier 3 cities of Australia.

• Data centre services: There is a growing demand for data centre services, such as managed services, extremely intricate cleaning and maintenance services, security, and consulting.

Attributes of the Data Center Market in Australia:

• Total Live IT Supply: Currently stands at >1GW with overall contracted national utilisation at 85-90%.

• Key Demand Indicators: Contracted IT Demand: 932 MW, Live IT Supply: 1,061 MW, Power Utilization: 83-93%.

• AI Investment: AI-led investment is picking up (8-9x) in Australia for the past 2 years, mirroring US trends.

• Workload Contribution: AI workloads are expected to contribute 20% absorption by 2028 based on global workload benchmarks.

• Private Investments in AI: Australia's private investments in AI appear to be in early stages compared to the US, roughly 4-5 years behind.

• Data Center Capacity: Melbourne and Sydney accounts for ~50% of Australia's data center capacity.

• Co-Lo Market Growth: The co-location market continues to forecast significant supply growth, with analysts projecting annual growth rates of up to 20% between 2025 and 2028.

• Co-Lo Data Center Hub: Melbourne is Australia’s largest data centre hubs, with more than 230 co-location data centres.

• Investment: Investment in Australian data centre capacity is set to exceed AUD $26 billion by 2030.

• Energy Demand: Data centre electricity use is set to double by 2028.

• Global Data Centers: An estimated 4000 data centres (DCs) operate globally and Australia is in the top 5 locations with a similar capacity to London.

• Market Size: The mega DC market, valued at USD 23.23 billion ($37 billion) in 2023, is projected to expand to USD 29.34 billion ($44 billion) by 2028, growing at a compound annual growth rate (CAGR) of 4.78 per cent during the forecast period.

• Moody's Growth Forecast: Moody's estimate the Asia-Pacific data centre capacity will grow at a compound average of 20 per cent or so through 2028 which is in line with Australia.

• Investment: In dollar terms, the expansion involves investment of over US$564 billion to 2028, with a pipeline of over 4400 MW under construction in the APAC region.

• Australia Data Center Market Size: The Australia Data Center Market size is estimated at 2.18 thousand MW in 2025, and is expected to reach 4.07 thousand MW by 2030, growing at a CAGR of 13.35%.

• Colocation Revenue: The market is expected to generate colocation revenue of USD 2,501.8 Million in 2025 and is projected to reach USD 5,318.5 Million by 2030, growing at a CAGR of 16.28% during the forecast period (2025-2030).

• Cloud Adoption: As of 2022, approximately 42% of businesses in Australia reported using cloud computing services. By 2028 this is forecasted to be ~60%

Value Chain for the Data Centre Business in Australia

There does not seem to be an agreed value chain for the data centre business in Australia. Based on our conversations, we have summarised the Data Centre Value Chain as follows:

1. Site Selection: involves the procurement of land (including land banking for future sites) with consideration to power supply, water supply and telecom networks. Increasingly, access to renewable energy sources are a key site selection criterion.

2. Raw Materials: involves the procurement of materials required for the construction of data centres, such as steel, concrete, and glass. An increasing differentiator for Data Centre constructors and owners is access to strategic suppliers of prefabricated materials for the build.

3. Design, Approvals and Construct: building of data centre facilities. An increasing differentiator for Data Centre constructors and owners is the use of proprietary methods (often called modular DC) to truncate the design-to-construct process to approx. 12 months or less which is being achieved by some US based constructors. In Australia, the average design-to-build process (including approvals which varies State by State) can range from 18-26 months.

4. Equipment Installation: installation of IT hardware, including servers, storage, and networking equipment. Again, the larger Data Centre owners are finding the strategic alliances and ensuring access to the latest and greatest equipment.

5. Connectivity: providing network connectivity to data centres. Australia has a competitive market for connectivity providers but can be restricted by access to the connection points given the vast size of the country.

6. Data Centre Operations: managing the day-to-day operations of data centres, including power, cooling, maintenance and cleaning, and security.

7. Customer Management: managing and contracting the businesses and individuals who use data centre services. In Australia the margins on Co-Lo remain slightly better than the margins for hosting hyperscalers given the sheer volume and negotiation power of the Hyperscalers.

Construction Companies in Australia

Increasingly, the discussion with investment firms has moved beyond the ‘why’ and ‘where’ for data centre investment. The discussion has moved to ‘whom’ should we be using for the design, approvals and construct in order to minimise the risk of construction delays and thius revenue generation delays. In our discussions, we have heard the following independent construction companies mentioned positively:

• Hansen Yuncken is a privately owned construction company in Australia.

• Lendlease is an international property and infrastructure group with a significant presence in Australia.

• Multiplex is a global construction company with a significant presence in Australia.

• PCL Construction is a global construction company with a significant presence in Australia.

• Vaughan Constructions is a privately owned construction company in Australia.

Opportunities for Investment Firms in Data Centres in Australia

We believe the following areas are some of the ‘spot opportunities’ across the value chain for investment firms in data centres in Australia:

• Retrofit conversion (Brownfield DC to Greenfield): The Australian market remains expensive for DC conversion from Co-Lo to AI-ready. The opportunities are significant given the site locations and established utilities connectivity, however Australian labour and materials costs means the conversion can be 80-90% the cost of a new greenfield. An investment firm that can fund and support a retrofit constructor with proprietary methods to retrofit may find significant market opportunities.

• Growing and retaining a skilled workforce: The challenge for construction firms to train and retain a skilled workforce for data centre build and retrofit is one of the industry’s largest challenges. An investment firm that has a PortCo or experience with construction workforces and field technicians could invest in Data Centre design and construction businesses with the right remuneration, incentives, training and career progression ladders that retain the most experienced workforce in this industry.

• Development of data centre construction businesses: Investment firms can fund and develop data centre design and construction businesses and peripheral businesses that import and supply the internal equipment. A rapid mover towards proprietary modular design and construct (for greenfield and brownfield conversion) will obtain significant capex benefits in comparison to the market as well as faster time to revenue.

• Strategic Location: At a macro level, Australia's position in the Asia-Pacific region makes it a crucial hub for data connectivity. This is particularly important for businesses seeking to serve the growing Asian market. At a micro level, we see opportunities to minimise latency (a key demand for Hyperscalers) through careful Tier 2 and 3 site acquisitions in regional areas and smaller cities.

• Increased Demand for Edge Computing: The rise of video-based streaming, IoT and real-time applications as well AI services is driving demand for edge data centers, which process data closer to the source. This creates opportunities for investments in strategically located regional data centers and micro sites in the metro areas.

• Focus on Sustainability:The growing emphasis on sustainable data center operations, with increasing demand for renewable energy sources. Microsoft has published its aspiration for 100/100/0 where 100% of its data centre energy will be 100% renewable thus allowing for zero carbon credits.This presents opportunities for investments in the renewable energy generation itself as well as data centers powered by renewable energy, aligning with Hyperscaler ESG goals.

Sources:

• Revolutionising Digital Infrastructure Deployment

• Australian-Infrastructure-Investment-Monitor-2024_WEB_2.pdf

• Huawei Data Centre 2030 Forecast

• 2024 ACIF ACMR

Written by Andrew Lorken, Co-Founder and MD of Attain.ai Advisory Pty Ltd

AgenticAI evolves towards industry vertical solutions

We see the AgenticAI development industry evolving towards the industry-verticalised solutions much like the days of ERP and SaaS. Do you?

The initial wave of AI adoption has seen a broad, horizontal spread of its use, with industries across the board implementing general-purpose AI tools. The results have been patchy and sporadic with only a smaller percentage of use cases creating lasting, sustainable benefits – however, these are the early days of its adoption and embedment. The future remains exciting.

As AI technology matures, a distinct shift is underway: the rise of industry-vertical AI solution creation and adoption. This evolution is driven by the fundamental limitations of "one-size-fits-all" AI and the growing recognition that true AI value lies in deep specialisation – either by industry specialisation which happens to be the “go-to-market” we are used to with SaaS and platform providers, or, through competency-areas such as leading practices in e-commerce retailing or predictive network assurance in industries such as utilities and telecoms.

The drivers moving us towards industry-specific AI:

• The necessity of nuance:

  • Each industry possesses unique data sets, operational complexities, and regulatory landscapes. General AI often lacks the contextual understanding required to navigate these intricacies effectively.

  • The highly confidential data used in healthcare is very different than the data used in public transport.

• The importance of proprietary data:

  • Each industry possesses uniqueness, and each organisation within the industry must protect and leverage its own data in a proprietary manner to gain advantage, for example, the better use of 1st Party data by a media company to create targeted advertising will enable a competitive advantage until others can follow the same formula.

  • Whilst that data remains proprietary, nonetheless, over time, the whole industry becomes better at the capability of targeted advertising with inhouse or third party providers supplying the industry specific solutions to power the protected data sets and AI agents constantly looking for better advertising matching to the end user.

• The demand for precision:

  • Organisations are increasingly seeking AI solutions that deliver precise, targeted outcomes, such as the targeted advertising use case above. This necessitates AI models that are fine-tuned to the specific requirements of their respective industries and the use cases within them.

• The pursuit of sustainable ROI:

  • Many AI and data solutions to date, to the disappointment of the CFO, can be classed as short-term improvements, or, the hope of efficiency gain by aggregating often disparate time savings along a complex set of activities. Industry-vertical AI offers a more direct path to measurable return on investment that is sustainable over the longer term. By addressing specific pain points and optimising core processes, these tailored solutions deliver sustainable business value (in either revenue growth or cost take-out).

The transition from horizontal to vertical AI reflects a natural progression in technological adoption. As we move beyond the initial excitement of general AI, we are entering an era of specialised AI, where deep domain expertise and targeted solutions will drive the next wave of innovation. The trend is shifting from "one-size-fits-all" AI to specialised solutions that address the unique requirements of different industries.

The emergence of industry-based AI solutions:

From our discussions with numerous stakeholders (investors, industry leaders, research and knowledge base partners, heads of data and AI within mid-to-top tier organisations) there's a strong trend and belief that the AI industry is increasingly moving towards industry-based verticals and specialties.

Unique challenges driving vertical solutions:

• While general-purpose AI is powerful, many industries have unique challenges and data requirements that require specialised solutions, often with nuanced and specialised data.

• "Vertical AI" focuses on optimising processes within a specific industry or competency domain, leading to more effective outcomes.

• Enhanced data processing capabilities: Advancements in AI allow for the analysis of complex, industry-specific data, patterns and trends.

• Growing investment in AI technologies: More resources are being directed towards developing specialised AI applications by the industry players themselves for their own unique needs.

Key industries and AI Specialisation:

It's very clear that the AI industry is rapidly specialising, with numerous sectors already implementing vertical AI solutions. Some key industries and how they're utilising these specialized AI applications are outlined below:

• Healthcare:

  • AI is revolutionising diagnostics, drug discovery, and personalised medicine.

  • Specific applications include:

    • Analysing medical images for early disease detection.

    • Predicting patient outcomes and optimising treatment plans.

    • Accelerating drug development through AI-powered simulations.

• Financial Services:

  • AI is crucial for fraud detection, risk management, and personalised financial services.

  • AI is increasingly being used for share market and financial trading

  • Examples include:

    • Real-time fraud detection through transaction analysis.

    • AI-driven risk assessment for loan applications.

    • Personalised investment recommendations.

• Manufacturing:

  • AI is enhancing efficiency and quality control in manufacturing processes.

  • Applications include:

    • Predictive maintenance to prevent equipment failures.

    • AI-powered quality inspection for defect detection.

    • Optimization of production processes.

• Retail:

  • AI is used to personalise customer experiences and optimise supply chains.

  • Examples include:

    • Personalised product recommendations based on customer data.

    • AI-driven inventory management and demand forecasting.

    • Highly effective chatbots for customer service support.

• Legal:

  • AI is used to analyse large amounts of legal documentation, for things like e-discovery, and contract review.

The emerging attributes of vertical AI:

• The rise of "vertical AI agents" designed for specific tasks within industries is increasingly becoming a speciality of AI and data service providers.

• Startups are beginning to specialise in niche AI solutions for particular industry sectors.

• Moving from the horizontal or broad application of AI by adopting AI toolsets for automation of repetitive tasks – towards the use of AI solving industry specific process challenges that have always previously been too challenging for automation. One example, in our industry speciality of Telecoms, is predictive maintenance and assurance of a telecom network and services on top of that network. Until recently, millions of investment dollars were required into network diagnostic systems, relying on very accurate network inventory assets, in order to correlate network alarms and predict network issues that will impact end users. Today, a telecom organisation can use its own data with specialised AI toolsets to predict when and where the network or service degradation or outage could occur, and to proactively intervene to mitigate for it.

Example #1: Telecom Industry – Vertical industry AgenticAI providers

Continuing with the industry example of Telecoms, this section provides some examples of industry-specific AgenticAI players and how they are disrupting a very traditional industry. Firstly, some of the key applications of AgenticAI within Telecoms includes:

• Autonomous Network Management:

  • Agentic AI supports networks to self-heal, self-optimise, and dynamically adapt to changing traffic patterns.

  • It facilitates real-time network planning and design, allowing for on-demand adjustments.

  • It supports closed-loop assurance and zero-touch operations, minimising downtime and improving network reliability.

• Enhanced Customer Service:

  • Agentic AI powers advanced virtual assistants that can handle complex customer inquiries and resolve issues autonomously.

  • It enables personalised customer journeys, adapting to individual preferences and providing proactive support.

  • It improves first-contact resolution rates and reduces call handling times.

• Predictive Maintenance:

  • Agentic AI analyses network data to predict potential failures and trigger proactive maintenance, preventing disruptions.

  • It optimises resource allocation and reduces maintenance costs.

• Network Security:

  • Agentic AI enhances network security by detecting and responding to threats in real-time.

  • It can dynamically adjust security protocols based on identified risks.

• Contract Management:

  • Agentic AI can be used to analyse large numbers of vendor contracts, to extract key information, and provide real time reporting. This allows for better vendor management, and cost control.

Some of the leading organisations in this domain include the following:

• Network Optimisation and Automation:

  • Net AI: This company focuses on providing AI-driven solutions for network optimisation, offering granular visibility into service usage and traffic flows. Their AI-led technology helps telecom providers enhance network efficiency and reduce costs.

  • Avanseus: Avanseus provides AI-based predictive analytics for network hardware, helping to monitor and predict maintenance faults and network failures, thus minimising downtime.

  • XenonStack: This company focuses on autonomous operations and 5G network management. They utilise generative AI and predictive analytics to help telecom companies improve customer satisfaction, optimise network performance, and mitigate security threats.

• Customer Experience and Service:

  • Success.ai: This startup develops conversational AI platforms, offering chatbot solutions that automate customer service and improve response times.

  • Numerous companies that produce AI driven RPA (Robotic Process Automation) are also heavily involved in the telecoms industry, by automating repetitive tasks. Salesforce, ServiceNow and many other legacy brands play heavily in this space.

• Data Analytics and Predictive Maintenance:

  • Predicty: Predicty provide AI-based predictive analytics for customer churn, helping telecom providers identify and retain at-risk customers.

  • MindTitan: This company develops fault grouping solutions, using machine learning to monitor networks and group faults based on interdependencies.

Example #2: Fintech Industry – Vertical industry AgenticAI providers

The intersection of AI and fintech is a hotbed of innovation, with numerous startups pushing the boundaries. Agentic AI is transforming areas like fraud detection, risk assessment, and personalized financial advice. AI agents can monitor transactions in real-time, identify anomalies, and provide proactive alerts. Automating complex compliance and risk management tasks is a major focus. Some examples of startup companies that are specialising in the fintech industry, with a focus on AI's role include:

• AI-Driven Financial Analysis and Investment:

  • Axyon AI: This company provides AI-powered investment management platforms, helping asset managers make better decisions.

  • Token Metrics: Focuses on AI-driven analysis of the cryptocurrency market, providing data and insights to investors.

• Fraud Detection and Risk Management:

  • ComplyAdvantage: Uses machine learning to help financial institutions detect and prevent financial crime.

  • Many companies are working on AI driven fraud detection, as this is a high growth area.

• Personalized Financial Services and Chatbots:

  • Kasisto: Develops conversational AI platforms for financial institutions, enabling personalized customer interactions.

  • Many companies are working to improve customer service in the fintech area, using AI driven chatbots.

• Data Analytics and Lending:

  • Perfios: Provides data analytics platforms for financial institutions, helping with lending decisions and fraud prevention.

  • Lendio: uses AI to connect small businesses with lenders, and to automate loan decisioning.

The successful implementation of industry vertical AI hinges on several critical factors.

The key elements for success include:

1. High-quality, industry-specific data:

• Data is the fuel:

  • AI models, especially those designed for specific industries, require vast amounts of relevant and accurate data to learn and perform effectively.

  • The quality of the data directly impacts the accuracy and reliability of the AI's output.

  • Industry players will protect and guard their proprietary data to ensure their competitive advantage is protected for as long as possible.

• Data Governance:

  • Establishing robust data governance practices is crucial to ensure data quality, security, and compliance.

2. Deep industry domain expertise:

• Understanding the nuances:

  • AgenticAI developers must possess a deep understanding of the specific industry's challenges, workflows, and regulatory requirements.

  • This expertise is essential for designing AgenticAI solutions that address the use cases for real viability and fesibility – that is, to ensure the benefits of the deployment of AI and actually sustainable and realised for the long term.

• Collaboration:

  • Close collaboration between AI developers and industry experts is vital for successful implementation.

  • Leading AgenticAI providers will have the balanced mixture of subject matter experts and AgenticAI developers working together through the use case origination, viability and feasibility testing and the service design through implementation phases.

3. Clear business objectives and ROI:

• Defining the goals:

  • Organisations must clearly define the business objectives they want to achieve with AgenticAI.

  • Measuring the return on investment (ROI) that is sustained over 2-3 years is essential for demonstrating the value of AI solutions.

• Use-Case identification:

  • It is very important to find the specific use cases that will return the most value. Our method of testing use cases for Viability and Feasibility has proven worthwhile for the investment cases of our clients.

4. Seamless integration and scalability:

• Integration with existing systems:

  • AI solutions must seamlessly integrate with existing systems and workflows to minimise disruption. This is easier said, than done, but the alternative investment case is usually for traditional, expensive legacy systems instead of more nimble AgenticAI solutions.

  • Scalability is crucial for handling increasing data volumes and user demands.

• Adaptability:

  • AI systems should be designed to adapt to evolving industry trends and changing business needs.

5. Focus on user experience:

• User-friendly interfaces:

  • AI solutions should have user-friendly interfaces that are easy to navigate and understand.

  • Providing adequate training and support is crucial for successful user adoption.

  • Our methodology ensures a focus on service design, for the end user, is the driving force to AgenticAI solution development.

Whilst still a nascent and emerging trend, AgenticAI with an industry vertical focus is the likely next phase of AI development through to 2030. We look forward to being part of that journey.

Written by Andrew Lorken, Attain.ai Co-Founder.