Most PPP training focuses on the front end of a project: appraisal, feasibility, Value for Money and procurement. Those stages matter. But once the contract is signed, the public authority has to live with it for decades.

That is when the questions get hard. Is the operator meeting the service levels? Should a deduction be applied? Who pays when demand falls short, or a new regulation changes the economics? What condition must the asset be in when it is handed back?

Government lawyers sit at the centre of all of these questions. If they only learn PPPs from the procurement angle, they meet the hardest problems for the first time on a live project. Our programme was designed to change that.

Lesson 1

A policy goal such as “a modern, high capacity terminal” sounds clear in a cabinet paper. In a contract, it is almost unenforceable. Who decides what “modern” means in year 17?

The fix is to write output specifications instead of input ones. An input specification tells the private partner how to build or operate something. An output specification states the result the public needs, in a form that can be measured and verified. A simple test is SMART: is the requirement Specific, Measurable, Achievable, Relevant and Time bound?

On the first day of the workshop, participants reviewed a draft specifications package that deliberately mixed good clauses with weak ones. Their job was to find the weak ones and rewrite them. Later in the week, they found out why it mattered: the same clauses came back when they had to monitor performance and resolve a dispute.

Lesson 3

A PPP involves lawyers, economists, engineers, financiers and line ministries. When each group uses PPP terms slightly differently, negotiations slow down and gaps appear in the contract.

That is why the programme opened with the CP3P Foundation, the Certified PPP Professional programme developed by APMG International with the World Bank. Over four days, participants covered the full PPP lifecycle: building a national PPP framework, screening and appraising projects, structuring and tendering contracts, and managing delivery, operation and handback. The week ended with the official CP3P Foundation exam.

Certification does two things. It gives the whole team one internationally recognised vocabulary, and it gives the organisation evidence that the capability is real, not just attended.

Lesson 4

Most workshops use a fresh example for every topic. It keeps sessions tidy, but it hides the most important truth about PPP contracts: every decision is connected.

The second week, Contract Management for PPPs, used a single running case for all four days: Bayview International Airport, a composite PPP built from real project experience. The 40 page case package held draft specifications, a term sheet, financing documents, a quarterly performance report, bid documents and a dispute scenario. The case was built so that choices made early came back later:

A vague clause on Monday became a monitoring headache on Wednesday and a live dispute on Thursday, just as it would on a real project. That experience is much harder to forget than a slide.

The programme ran over two consecutive weeks: the CP3P Foundation course and exam from 28 September to 1 October 2026, then the Contract Management workshop from 5 to 8 October. It was delivered by PPP Alliance, Cybiant’s specialist PPP subsidiary, and led by Ir. Drs. Jan-Willem Middelburg, one of the first trainers worldwide certified under CP3P.

What stood out was how quickly the conversation moved from definitions to judgement. By the final afternoon, groups were debating which dispute mechanism to choose and how to protect the relationship with the private partner- the kind of call these lawyers will make on real projects.

The full story, including the day by day agenda and outcomes, is in our case study:
Equipping a national government legal office to structure and manage PPPs.

We design, build, run and teach. For Public-Private Partnerships, that work is led by PPP Alliance, our specialist subsidiary.

If your organisation drafts, vets or manages PPP contracts, three questions are worth asking:

A PPP is won or lost in the decades after the ceremony. Teams that share a common language, write measurable specifications, allocate risk with care and have practised the hard moments before they happen protect the public interest for the full life of the contract. Automate What Matters.

Four forces have pushed ITAM out of the IT department and into the risk register. Each one lands on a director’s desk as money, liability or reputation.

The common thread is simple: you cannot govern, secure or optimize what you cannot see. Flexera’s own takeaway is that visibility is now a scope problem rather than a tooling problem, because the estate keeps expanding into cloud, SaaS and AI faster than policy can follow.

The International Association of IT Asset Managers (IAITAM) has set the training and certification standard for the profession since 2002. Its advanced courses are built on the IAITAM Best Practice Library (IBPL), which organizes ITAM into 12 Key Process Areas covering everything from acquisition and contracts to disposal and security.

IAITAM offers a role-based path, so organizations can certify the right people for the right job. Cybiant delivers four core IAITAM certifications as instructor-led virtual courses:

Two design choices make IAITAM credentials especially credible at board level. First, every certification must be renewed annually, so a certificate reflects current practice rather than a one-off exam. Second, the exams are demanding: most require a score of 85% to pass.

Certified people set the standard; automation keeps it running at scale. No team can manually track thousands of SaaS subscriptions, cloud licenses and AI agents that change every week.

Cybiant brings both halves together. As the author of the global standard for agentic service automation and an IAITAM Accredited Training Organization, Cybiant can design, build, run and teach ITAM end to end.

IT assets are now among the largest and least visible investments on the balance sheet. Organizations that pair IAITAM-certified people with intelligent automation will spend less, pass audits with confidence and govern AI before it governs them.

• Flexera 2026 State of ITAM Report (June 2026, 512 respondents)
• Flexera: How governance gaps are driving costly software audits
• IAITAM Training and Certifications
• IAITAM Frequently Asked Questions
• DORA Regulation Article 8, Identification
• Cybiant IAITAM Training

Private investment in infrastructure across low and middle income countries reached $100.7 billion in 2024, according to the World Bank’s Private Participation in Infrastructure (PPI) Database. The 2024 PPI Annual Report notes this is the first time the $100 billion mark has been passed since the COVID-19 pandemic began, and that emerging infrastructure subsegments are attracting more of that capital. For comparison, 2023 saw $86 billion across 322 projects in 68 countries.

AI is one of the forces pulling demand into new asset classes. The World Bank describes AI readiness in terms of “four Cs”. Where those foundations are missing, it observes, promising AI pilots rarely scale responsibly (World Bank, 2026). Every one of those four Cs is a candidate for public-private collaboration.

The public sector is also adopting AI in its own operations. The OECD’s 2025 report Governing with Artificial Intelligence reviews around 200 use cases across 11 core government functions, including public procurement, which accounts for roughly 13% of GDP in OECD countries. PPP units and contracting authorities are part of that shift whether they plan for it or not.

The APMG PPP Certification Program was developed on behalf of the Asian Development Bank, EBRD, the Inter-American Development Bank, the Islamic Development Bank, the Multilateral Investment Fund and the World Bank Group, with funding support from PPIAF. Its PPP Guide is the shared body of knowledge for practitioners, and it is the foundation for the Certified PPP Professional (CP3P) credential.

The Guide treats PPPs as a contractual way to deliver public assets and public services. It distinguishes infrastructure PPPs (building new assets or substantially upgrading existing ones) from service PPPs (where a private partner manages existing infrastructure or operates a public service). Both matter in an AI context: a government data centre is an infrastructure PPP, while an outsourced AI-enabled citizen service looks much more like a service PPP.

Three principles from the Guide are especially relevant when technology moves fast:

None of these principles change because AI is involved. What changes is the speed, the data available at each phase, and the type of risks that have to be allocated.

The second shift is that AI infrastructure is itself becoming a PPP asset class. The World Bank Group’s approach shows how the public and private roles fit together (World Bank, 2026):

In Romania, IBRD support is helping the government migrate up to 60 applications, including procurement and payment systems, to the cloud under a new regulatory framework. A joint World Bank Group study has assessed market and financing conditions for cloud and data centre development across 15 priority countries.

Partnerships are also taking new shapes. In October 2025, Google and the World Bank Group announced a collaboration to help governments deploy AI-powered digital public infrastructure for sectors such as agriculture, health and skills, accessible in more than 40 languages on simple devices. Not every arrangement like this is a PPP in the APMG sense, but the same questions apply: who owns the data, who carries performance risk, and what happens at the end of the term.

For broadband and connectivity, the World Bank already promotes gap financing, risk mitigation instruments, PPPs and infrastructure sharing to make investment viable in underserved areas (World Bank).

For the public sector

For the private sector

Five issues come up repeatedly:

The OECD recommends proportionate, risk-based guardrails rather than blanket rules, so innovation and safety can move together. That fits the PPP mindset: allocate each risk deliberately, document it, and manage it for the life of the contract.

Technology changes quickly; good PPP practice changes slowly. The organisations that benefit most from AI-era partnerships will be those whose people understand both. On the PPP side, the CP3P credential gives public officials, advisers, lenders and contractors a shared, internationally recognised language. Candidates earn it by passing three exams based on the APMG PPP Guide.

Cybiant is an APMG accredited training organisation for the CP3P programme and advises governments and companies on PPPs, data and AI, and intelligent automation. That combination matters: the questions in this playbook sit exactly where PPP structuring meets digital delivery.

• ADB, EBRD, IDB, IsDB and WBG. The APMG Public-Private Partnership (PPP) Certification Guide. Washington, DC: World Bank Group. ppp-certification.com
• World Bank PPP Resource Center. PPP Certification
• World Bank. Private Participation in Infrastructure (PPI) 2024 Annual Report
• World Bank. Private Sector Pumps $86B into Infrastructure in Low to Middle Income Nations (May 2024)
• World Bank. Building Data Infrastructure for AI Readiness (May 2026)
• World Bank. Benchmarking Infrastructure Development 2023
• World Bank PPP Resource Center. Disruption and PPPs / PPP Contracts in an Age of Disruption
• World Bank PPP Resource Center. Guidelines and Tools
• World Bank. Digital Infrastructure: Broadband Infrastructure, Access and Use
• OECD (2025). Governing with Artificial Intelligence: AI in public procurement
• Policy Edge (2025). Summary of OECD Governing with Artificial Intelligence
• Google (2025). Google and the World Bank Group AI collaboration

ITIL 4 gave organisations a durable foundation. Its service value system connects guiding principles, governance, the service value chain, management practices and continual improvement into a single model for creating value. It also introduced four dimensions of service management: organisations and people, information and technology, partners and suppliers, and value streams and processes.

Crucially, ITIL 4 put automation on the agenda. One of its seven guiding principles is Optimize and automate: use people, often the scarcest resource, and automation effectively, so teams stop spending time on simple, repetitive tasks and focus on complex decisions, creative work and problem solving.

The principle was right. The problem is how most organisations operationalised it. In practice, traditional ITSM still tends to look like this:

Technology is rarely the bottleneck. Operating models, governance and skills are.

A new framework version does not upgrade anything on its own. Here is what separates organisations that get real value from automation from those that accumulate expensive bots.

ITIL describes what good service management looks like. What it does not do, by design, is prescribe how to architect, build and run automation at enterprise scale. That is the gap the Service Automation Framework (SAF) was created to fill.

Cybiant authored SAF, the global standard for agentic service automation. It gives organisations a structured method to identify the right opportunities, design services around both human and digital workers, and govern automation as an ongoing capability rather than a one-off project.

Before writing a roadmap, decide how the CoE will actually operate. There are three common models, and most mature organisations land on a version of the third.

Pick the model deliberately, rather than letting it emerge by accident, because it determines almost everything that follows: who you hire, what tooling you standardise on, and how quickly you can scale.

A functioning CoE, even a small one, needs clarity on who does what. The specific titles vary, but the responsibilities do not:

• Executive sponsor — keeps the CoE aligned to business strategy and unblocks resourcing and cross-department disputes.
• CoE lead — owns the roadmap, the governance model, and the relationship with the rest of the business.
• Automation architects and developers — design and build the automations and agents, and maintain the shared component library.
• Governance and risk owner — defines the controls an automation must pass before deployment, particularly important once agentic AI is in scope.
• Business analysts or process owners — translate real operational knowledge into requirements the technical team can build against.
• Automation champions — embedded advocates who surface opportunities, support adoption, and act as first point of contact before issues escalate.

Smaller organisations often combine several of these into one or two people at the start. What matters is that every responsibility is assigned to somebody, rather than assumed to be covered informally.

The fundamentals of building a CoE have not changed much over the past several years: sponsorship, governance, prioritisation, and a realistic phased roadmap still matter as much as they always did. What has changed is the scope of what the CoE now needs to govern. Agentic AI adds a layer of autonomous decision-making on top of the deterministic automation most CoEs were originally built to manage, and that layer needs its own governance model rather than an assumption that existing RPA controls will simply cover it.

Organisations that treat this as an extension of a disciplined automation practice, rather than a separate initiative bolted on beside it, are the ones scaling agentic AI successfully in 2026. The ones treating it as a fresh start, disconnected from the governance and lessons already built into their existing CoE, tend to be relearning the same hard lessons from scratch.

None of this replaces human judgement. It replaces the manual assembly work that used to sit between a decision and its execution.

Agentic AI didn’t appear overnight; the underlying research has been building for years. What changed is that several enabling forces converged at once: reasoning capability in the underlying models improved meaningfully, context windows grew large enough to hold real operational data, integration protocols standardised so agents could reliably connect to enterprise tools, orchestration platforms matured enough to coordinate multiple agents at once, and organisations finally have enough deployed use cases to point to measurable return on investment.

The result is a visible shift in enterprise posture. Analyst projections now put task-specific AI agents inside a substantial share of enterprise applications within the next few years, up from near-zero adoption only recently. That’s a signal that this has moved decisively from experimentation into operational deployment.

The question for 2026 isn’t whether an enterprise can deploy agents. It’s whether it can do so with the accountability, auditability, and oversight that autonomous action demands. That distinction is where most agentic AI initiatives quietly stall.

Cybiant is the world’s only partner that has authored the global standard for service automation, and can design, build, run, and teach it, end to end. That statement isn’t a tagline; it describes a specific and rare position in the market. Most organisations offering “agentic AI services” today are doing one of two things: reselling a platform’s technology, or advising on strategy from the outside. Very few can do both, and even fewer can do it against a framework they wrote themselves.

If your organisation is at the point of asking “should we deploy agentic AI,” the honest answer is rarely “not yet.” It’s “not without the right foundation first.” A few practical starting questions worth asking internally:

Agentic automation is not the next version of RPA, and it’s not generative AI with extra steps. It’s a genuinely new capability, one that requires new governance, new architecture, and new expertise to deploy safely and profitably.

The enterprises that will lead their industries by the end of this decade won’t be the ones who moved fastest into agentic AI. They’ll be the ones who moved with a standard, a framework, and a partner who has done this before.

That’s the work Cybiant does. Design it. Build it. Run it. Teach it.

The Enterprise Big Data Framework is a vendor-neutral capability model developed by DASCIN. Its full methodology is set out in Jan-Willem Middelburg’s book, The Enterprise Big Data Framework: Building Critical Capabilities to Win in the Data Economy, which introduces the framework and its six capabilities in depth. DASCIN’s Enterprise Big Data certifications, accredited by APMG International, run from the entry-level Enterprise Big Data Professional credential through to the Architect and Scientist specialisations — all built around the six capabilities the framework defines, rather than around any single vendor’s platform or toolset.

Rather than prescribing specific technologies, the framework identifies six capabilities that every enterprise needs to build out over time: strategy, architecture, algorithms, processes, functions, and artificial intelligence. Most organisations already own the technology needed to collect and store data at scale. What actually separates the businesses that extract value from their data from the ones that do not is whether these six capabilities are developed deliberately, in a sensible order, and with enough organisational backing behind them.

The book frames these six capabilities as a progression that any organisation can use to measure its own maturity over time, rather than as a one-off project with a fixed end date. That distinction matters for how a business plans its data investment: it reframes the work as an ongoing capability to be maintained, not a system to be switched on once and left alone.

The data economy across Asia Pacific is growing faster than most organisations’ ability to absorb it, and the numbers behind that growth are worth sitting with for a moment.

IDC forecasts that AI spending in Asia Pacific, excluding Japan, will grow at 1.7 times the rate of overall digital technology spending, generating an economic impact of more than 1.6 trillion US dollars by the end of 2027 — a figure that keeps compounding through the back half of the decade as the region’s digital economy matures further. Combined with IDC’s more recent projection that AI will drive half of all new economic value created by digital businesses across the region by 2030, the direction is clear: the return on AI investment is shifting away from simple cost savings and toward new revenue streams and entirely new markets. Organisations without the underlying data capability to support that shift will find the opportunity difficult to capture, no matter how much they spend on AI tools themselves.

That opportunity is being met with real capital, not just intent. IDC reports that Asia Pacific, including Japan and China, already accounts for roughly a third of worldwide digital transformation spending — a share close to that of the United States and larger than that of Europe. This is not a future commitment sitting in a strategy document; it reflects capital already being deployed across cloud infrastructure, data platforms, and AI tooling by regional competitors who are not waiting for a perfect strategy before they start building. For any single organisation, that scale of surrounding investment changes the competitive calculus. Falling behind on data capability increasingly means falling behind a market that is moving as a whole, not just a handful of early adopters within it.

At the same time, confidence rarely matches actual readiness. A global data maturity study covering more than 1,100 business leaders across 21 countries, including 203 respondents based in Asia Pacific, found that more than 80% believed their organisation was more data mature than its competitors. Yet less than half of the same respondents actually met recognised best practices for data quality once assessed against a structured standard.

For most APAC organisations, the shortage is not ambition. Investment appetite and awareness of AI’s potential are both high across the region, arguably higher than at any point before. The shortage sits in the underlying capability:

• Architecture built without governance
• Algorithms built without a connected business case
• Strategy set without the organisational functions in place to carry it out

Each of these gaps is exactly what the Enterprise Big Data Framework’s capability model is designed to expose before it turns into a wasted investment — giving an organisation a structured way to check its own maturity against an external, vendor-neutral standard rather than against internal assumptions that may not survive contact with reality.

DASCIN developed the Enterprise Big Data Framework, and its founder Jan-Willem Middelburg wrote The Enterprise Big Data Framework: Building Critical Capabilities to Win in the Data Economy, the published reference that introduces the framework’s six capabilities and its capability maturity model to a general business audience. Cybiant is an accredited training organisation delivering the Enterprise Big Data certification pathway, from Enterprise Big Data Professional through to Architect and Scientist level credentials, built on the same framework.

For an APAC business trying to close the gap between where it believes it stands on data and where it actually stands, that combination of a credible standard and accredited training is a practical place to start.

Sources:  IDC, IDC Predicts AI Spending Will Grow 1.7x Faster Than Overall Digital Tech Investments, Driving an Asia Pacific Economic Impact of Over $1.6 Trillion by 2027; IDC, IDC FutureScape 2026 Predictions: AI to Drive 50% of New Economic Value from Digital Businesses in APJ by 2030; IDC, Worldwide Digital Transformation Spending Forecast (regional spending share); Forvis Mazars, The Race to Data Maturity in APAC; AI Malaysia (National AI Office), National AI Action Plan 2026–2030; MyDIGITAL Corporation, Malaysia Digital Economy Blueprint, Phase 3 (2026–2030); Singapore Ministry of Digital Development and Information, Update to Singapore’s National AI Strategy (May 2026) and National AI R&D Plan (January 2026); ASEAN Secretariat, Digital Economy Framework Agreement (negotiations concluded May 2026); Middelburg, J-W., The Enterprise Big Data Framework: Building Critical Capabilities to Win in the Data Economy; DASCIN, Enterprise Big Data Framework white paper (2025).

The Green IT Framework is a structured model, developed by DASCIN, for managing the environmental impact of IT operations across seven components: Green IT Strategy, Hardware, Software, Processes, Practices, Data Storage, and the enabling disciplines of Measurement and Reporting and Continual Improvement.

Hardware gets most of the attention because its impact is easy to see: a server rack, a laptop fleet, a bin of retired equipment. But hardware is only one lever, and it is not necessarily the one with the most room left to improve. The framework treats Software, Processes, and Data Storage as domains of equal weight, because software does not release emissions on its own. It determines how much work the hardware underneath has to do, and at enterprise scale that difference adds up quickly.

The framework’s central point is that these domains reinforce one another. A hardware refresh without software optimisation, or a cloud migration without proper resource governance, tends to fall short of the savings it was supposed to deliver.

Green IT is shifting from a sustainability team initiative to something CIOs are directly measured on. Four data points, drawn from separate research firms, show why.

DASCIN developed the Green IT Framework, and Cybiant is the accredited training organisation delivering the Green IT and Sustainability pathway that equips IT and sustainability practitioners to apply it, covering the measurement methodology, the business case, and domain-by-domain implementation guidance across hardware, software, processes, practices, and data storage.

This is not a checklist dressed up as a course. It follows the same approach Cybiant takes with its other accredited pathways: take a rigorously developed standard and build the organisational capability to apply it, whether that means a government structuring its first public-private partnership, an enterprise governing its automation programme, or an IT function building the measurement discipline needed to make its sustainability figures credible to a board.

Sources: Gartner, Gartner Predicts 75% of Organizations Will Have Implemented a Data Center Infrastructure Sustainability Program by 2027; International Energy Agency, Energy and AI and Key Questions on Energy and AI; McKinsey & Company, Scaling Bigger, Faster, Cheaper Data Centers With Smarter Designs and The Cost of Compute: A $7 Trillion Race to Scale Data Centers; World Economic Forum, How Data Centres Can Avoid Doubling Their Energy Use by 2030; DASCIN, Implementing Green IT: A Practitioner Guide (2026).

CP3P (Certified Public-Private Partnerships Professional) is the world’s leading professional certification for PPP practitioners. It was developed by APMG International in collaboration with some of the most influential institutions in global development finance, including the World Bank, the Asian Development Bank (ADB), the European Bank for Reconstruction and Development (EBRD), the Inter-American Development Bank (IDB), the Islamic Development Bank (IsDB), the African Development Bank (AfDB), and the Public-Private Infrastructure Advisory Facility (PPIAF).

In other words, CP3P is not a private training company’s branded course. It is a global standard, co-authored by the institutions that fund and oversee the majority of the world’s infrastructure PPPs. The certification is built around the PPP Certification Program Guide, a free, peer-reviewed body of knowledge organized into eight modules that mirror the real lifecycle of a PPP project, from identification and appraisal, through procurement and contracting, to construction and operational management.

Preparation and Execution must follow, in sequence, building a practitioner’s competence step by step, much like a PPP project itself, which only succeeds when each phase is built on a solid foundation.

PPPs sit at the intersection of public accountability and private-sector discipline. When government teams lack a shared standard for how to evaluate, structure, and manage these deals, the consequences land squarely on citizens and budgets. CP3P matters to government leaders for four concrete reasons.

Cybiant is an Accredited Training Organization for CP3P and a founding member of the PPP Alliance, delivering the official Foundation, Preparation, and Execution certification courses to government officials and private-sector practitioners across Asia, Africa, the Middle East, and the Pacific.

This is not theoretical training. Cybiant has delivered the full CP3P pathway for national and provincial governments, often funded directly by institutions like the ADB. These are real ministries building real institutional capability to manage real infrastructure pipelines.

It is a natural extension of how Cybiant works: as the partner that has helped author and operationalize global best-practice frameworks, then trains organizations, public and private, to apply them with confidence. Whether the framework governs how a government structures a PPP or how an enterprise governs its automation program, the underlying philosophy is the same. Do not just point to a standard. Build the capability to run it.