Artificial intelligence may be forcing companies to rethink not only their technology infrastructure but the way boards and senior management govern risk itself. Traditional oversight structures built around committees, periodic reports and relatively static risk assessments can struggle when AI applications are introduced, modified and expanded across an organisation far faster than conventional governance processes can respond. That was the central argument presented at Ai4 2026 in Las Vegas by Brian Allen, Executive Director of the Center for AI Oversight, who argued that organisations should stop thinking primarily about governing AI technology and instead govern the management system surrounding its use.

The distinction is important. A board does not need to understand every model, agent or technical architecture operating inside the business. It needs confidence that management has established a repeatable process for deciding where AI can be used, what risks the organisation is prepared to accept, when problems or missed opportunities must be escalated, and whether the expected economic benefits justify the exposure. Allen framed this as part technology revolution and part management revolution, arguing that companies are introducing a rapidly changing technology into governance structures designed for a much slower environment.

This also changes the relationship between risk and strategy. Traditional cybersecurity discussions have often concentrated on downside: preventing breaches, protecting information and avoiding disruption. AI creates another dimension because moving too slowly can itself become a competitive risk. A company that blocks useful automation while competitors reduce costs, improve margins or deliver services faster may be accepting a different kind of exposure. Governance therefore needs to balance the danger of deploying AI too aggressively against the economic cost of not deploying it quickly enough.

Quarterly reporting illustrates the problem. A board may receive an AI risk update every three months, yet employees and business units can introduce new models, applications and agents in a fraction of that time. By the time a traditional report reaches senior management, the company’s actual AI environment may already have changed materially. Allen therefore challenged what he called the assumption that having an AI committee automatically means an organisation has effective governance. A committee is a governing body, but the more important question is what management system that body is actually governing.

If the committee’s main role is simply to receive periodic information about projects and risks, executives may be well informed without actively governing how the organisation balances AI opportunity and exposure. Effective oversight instead requires a mandate defining responsibility, authority, risk tolerance, escalation procedures and the evidence necessary to determine whether management is operating within those expectations. This broadly reflects the direction of established frameworks such as the NIST AI Risk Management Framework, which treats governance as something extending across the entire lifecycle of an AI system rather than as a one-time technical approval.

Risk tolerance is central to this approach because it gives management something concrete against which investment and behaviour can be measured. Broad statements that a company has a low, moderate or high appetite for AI risk provide limited practical guidance. A more useful model is to establish tolerances around particular applications and consequences. An internal productivity assistant may justify a different tolerance from an AI system influencing lending decisions, recruitment, pricing, customer data or safety-critical infrastructure.

Those tolerances should function as decision points rather than automatic stop signs. If an AI application moves beyond an agreed boundary, management can decide whether to reduce the risk, introduce additional controls, accept greater exposure or discontinue the activity. The same principle can work in the opposite direction. If controls are preventing a low-risk AI initiative from generating measurable savings or competitive benefits, that missed opportunity can also be escalated. Governance then becomes a mechanism for balancing risk and return rather than merely stopping activity.

That makes AI oversight closely connected to resource allocation. Once a company understands the economic importance of a particular use case and the exposure associated with it, management can decide how much investment in security, data quality, monitoring, human oversight and resilience is justified. The objective is not zero risk, which is unrealistic in any business environment. It is to determine how much risk the company is prepared to accept in exchange for a particular strategic benefit.

Allen proposed an oversight model built around several interconnected areas: governance responsibility, a risk-informed decision system, trust and assurance, risk-based strategy, and escalation and disclosure. The precise structure will differ between organisations, but the underlying principle is that these elements need to operate as one programme rather than as a collection of disconnected activities. Many large companies already have much of the required infrastructure through legal, compliance, cybersecurity, audit, data governance and enterprise-risk teams. The challenge is connecting those functions into a repeatable system that can make AI decisions consistently and quickly.

Clear ownership is therefore essential. Companies need to know who has the authority to establish the AI governance programme, who operates it and which matters must reach executive management or the board. Formal mandates and documented responsibilities become especially important when difficult decisions arise because employees cannot rely solely on informal relationships or assumptions that information will automatically move upwards through management. A codified governance structure provides something the organisation can lean on when business pressures become intense.

The board’s role should nevertheless remain different from management’s. Directors do not need to operate AI controls themselves. Their responsibility is oversight: understanding whether an appropriate system exists, whether management is executing it and whether significant risks and opportunities are reaching the board. Operational teams then determine how technical controls, cybersecurity measures, data processes and assessments are implemented. This separation allows directors to remain engaged without attempting to manage technology they are not equipped to operate.

Corporate law provides important context to that discussion, although it should not be interpreted as establishing a specific AI governance standard. Delaware’s Caremark line of cases has progressively examined directors’ and, more recently, corporate officers’ responsibilities for oversight and reporting systems, particularly where risks are central to a company’s operations. The lesson for AI is not that boards automatically face liability whenever an AI system fails. Rather, as AI becomes material to business operations, directors and executives may increasingly need to demonstrate that appropriate information and escalation mechanisms existed and that significant warning signs were not ignored.

This creates another reason for companies to document how AI decisions are made. Governance should produce evidence showing what management knew, which risks were considered, what tolerance was approved and why a particular decision was taken. The standard should not be that every AI decision must ultimately prove correct. Business involves uncertainty. A more realistic objective is demonstrating that material decisions were informed, appropriately authorised and consistent with the organisation’s established governance process.

Explainability illustrates the distinction. AI models are frequently described as black boxes because it can be difficult to reconstruct precisely how a complex model produced an output. That does not necessarily mean a company cannot provide meaningful accountability. Management can document what information entered the system, what it was intended to accomplish, which controls applied, what output was produced, how that output was validated and why the organisation considered it reliable enough for the intended purpose. The emphasis shifts from trying to expose every internal mathematical operation towards demonstrating that the overall decision process is controlled and defensible.

The same principle applies to AI strategy more broadly. Companies cannot know precisely how models, regulation and competitive pressures will evolve over the next several years. Governance therefore needs to accommodate uncertainty rather than pretending it can eliminate it. Organisations should establish boundaries that permit experimentation while creating clear mechanisms for escalating situations where financial, legal, operational or reputational exposure becomes significant.

For property companies and investors, these issues are likely to become increasingly relevant as AI enters investment analysis, valuation, leasing, development, building management, procurement and financing. An investment manager may use AI to screen acquisitions, a lender to analyse borrowers, a developer to assess sites and a property manager to automate tenant interactions. Each application creates a different balance between economic opportunity and risk, which means a single company-wide rule is unlikely to be appropriate for every use case.

Commercial real estate also demonstrates why governance needs to extend beyond the technology department. AI-generated investment analysis may depend on data controlled by asset managers and finance teams. Automated lease analysis involves legal departments. Building-management AI can affect operations and safety, while customer-facing systems create privacy and reputational considerations. Effective oversight therefore requires technology specialists to work with the people responsible for the underlying business processes rather than treating AI as an isolated IT function.

There is also a longer-term organisational issue. As companies automate more analytical and administrative work, they need to consider what happens to the institutional knowledge traditionally developed by employees performing those tasks. Junior professionals often learn by completing repetitive work, observing experienced colleagues and gradually understanding exceptions that are not obvious from formal procedures. If AI removes much of that work, businesses may need alternative ways to develop the judgement required for future senior positions.

This makes AI governance partly a question of organisational design. Companies are not simply deciding which software to buy. They are deciding which decisions remain with people, which can be delegated to machines, what knowledge must remain inside the organisation and how accountability operates when human and automated decision-making become intertwined.

The organisations best positioned for this transition may therefore be those that treat governance as an accelerator rather than a brake. A well-defined programme can allow employees to experiment faster because boundaries, responsibilities and escalation routes are already understood. Instead of requiring every new AI project to begin a fresh debate about risk, management can evaluate it against an established system and make decisions more quickly.

That may ultimately be the most important distinction between traditional technology governance and the emerging AI model. Companies cannot realistically govern every technological change individually at the speed at which AI is developing. What they can govern is the system through which those changes are evaluated and adopted.

For boards, the question is therefore moving away from whether the company has an AI committee or policy. The more important questions are whether management knows which risks it is willing to take, whether those boundaries can adapt as circumstances change, whether opportunities as well as threats are escalated and whether there is evidence that the organisation is actually operating according to those decisions.

AI governance at that level becomes much closer to strategy than compliance. It is a mechanism for deciding how quickly an organisation wants to move, where it is prepared to take risk and where stronger controls are necessary. As AI spreads across corporate operations, the ability to make those decisions repeatedly, transparently and at speed may become as important as the technology itself.

Source: CIJ.World Research & Analysis Team

The banking industry’s artificial-intelligence debate is moving from how much money institutions are investing towards a more difficult question: where is the return actually appearing? At Ai4 2026 in Las Vegas, BMO Financial Group’s Chief AI & Quantum Officer Kristin Milchanowski argued that AI is entering a new stage in financial services in which access to technology itself will become less differentiating. As banks increasingly obtain similar foundation models and cloud infrastructure, competitive advantage will depend more heavily on how effectively those capabilities are embedded into business processes, client relationships and decision-making.

That represents an important change from the first phase of generative AI adoption. Banks spent heavily on experimentation, employee tools, pilots and infrastructure, but a successful demonstration is not the same as an application capable of producing repeatable value across a large institution. Milchanowski described one of BMO’s central tests as whether AI has been connected to a workflow that genuinely matters to the business. If employees constantly need to be reminded that a particular system is using AI, the technology may not yet be sufficiently integrated. The objective is for AI increasingly to disappear into everyday banking processes rather than remain a separate product employees or customers have to consciously operate.

BMO has formalised much of this strategy through its Institute for Applied Artificial Intelligence & Quantum, created in 2026 as an enterprise-wide centre covering implementation, governance and emerging quantum capabilities. Milchanowski’s role therefore combines two responsibilities that are sometimes separated inside large organisations: building technology and establishing the framework under which that technology can operate. The economic challenge is that most banks do not have an accounting line showing AI revenue or AI profit. AI contributes to other activities, meaning the return has to be identified inside broader business outcomes such as faster credit processing, stronger client engagement, improved win rates or lower operating costs.

That makes baseline measurement critical. A bank needs to know how long a process took, what it cost and how effectively it performed before AI was introduced. Without that starting point, it becomes difficult to demonstrate whether the technology actually created value. Milchanowski described BMO’s work in commercial credit as one example, where generative AI is being used to collect information and help bankers prepare credit analysis more quickly. The bank has also developed Aura, a GenAI-powered assistant supporting commercial underwriting by retrieving information from approved systems and helping prepare parts of credit submissions while human underwriters remain responsible for the final assessment.

The underlying challenge is familiar across financial services. Large banks have accumulated technology over decades, leaving employees working across numerous applications and databases to complete a single task. A banker or financial-crime specialist may need to search several systems, review multiple documents and assemble the results into a memo before making a decision. Generative AI and agents can increasingly act as an information layer across those fragmented environments, retrieving relevant material and bringing it together for the employee. The human may continue making the consequential decision, but considerably less time is spent locating, comparing and reorganising information.

This is why many of the most credible banking applications currently involve document interpretation, information retrieval and the restructuring of content from large bodies of unstructured material. AI can locate clauses in legal documents, identify risk factors, compare information between reports or assemble material scattered across internal systems. These applications may appear less dramatic than autonomous banking, but they address workflows that consume significant employee time across lending, compliance, legal, operations and risk and can therefore produce measurable operational value.

The larger economic opportunity, however, may not come from efficiency alone. Milchanowski cautioned against treating AI as another cost-reduction programme. A strategy concentrated exclusively on doing the same amount of work with fewer resources could overlook the greater opportunity to generate revenue. Her argument was that banks should concentrate on improving client experience, increasing adoption and strengthening relationships, with efficiency emerging partly as a consequence of better-designed processes rather than becoming the sole objective.

That distinction matters because efficiency gains can quickly become part of the industry’s new cost base. If every major bank reduces the cost of producing a credit memo or answering a routine customer question, competitors will eventually be forced to make similar investments. The more durable advantage could come from using AI to improve how quickly institutions identify opportunities, respond to clients and make decisions. Milchanowski described this as decision velocity: reducing the administrative distance between information and the people responsible for acting on it.

Large organisations still rely heavily on periodic reporting processes in which information moves through multiple departments before senior management receives an aggregated picture. Forecasting can require several teams to prepare spreadsheets, reconcile assumptions and roll information upwards before a final view reaches leadership. If AI can continuously access approved information from core systems and assemble it into useful analysis, management could move towards a more continuous operating model. The value would not come from allowing AI to determine strategy independently, but from enabling people to make informed decisions sooner.

Before technology is introduced, however, BMO’s approach places responsibility on business teams to simplify existing processes. Milchanowski’s warning was straightforward: automating a poorly designed workflow simply allows the organisation to perform unnecessary work more efficiently. Banks therefore need to remove obsolete approvals, duplicated steps and bottlenecks before applying AI. Technology should then be used to redesign the remaining workflow rather than reproduce decades of accumulated complexity in automated form.

This principle also affects how AI programmes are organised and funded. Instead of building hundreds of unrelated applications, central technology teams can develop reusable capabilities that individual business lines adapt for their own products and workflows. BMO’s model combines enterprise-wide technology and governance with authority retained by the individual profit-and-loss businesses, which remain responsible for identifying client needs and allocating investment. Common infrastructure can therefore be developed centrally without requiring every business unit to create its own AI platform.

Governance becomes particularly important as banks move from assistants towards agents capable of performing actions. Milchanowski identified identity, orchestration and accountability as some of the main challenges. An institution needs to know who created an agent, why it exists, what information it used, which systems it may access and who is responsible for the results. As multiple agents begin interacting, banks will also need mechanisms determining which systems can communicate and under what circumstances.

Well-designed governance can eventually accelerate deployment rather than simply slow it. When responsibilities, thresholds and escalation procedures have already been defined, teams do not need to reconsider the same fundamental questions every time a new application is proposed. The institution can move more quickly because the boundaries are clearer. This also means employees need sufficient AI literacy to understand how tools are being used and remain accountable for decisions made with their assistance.

The technology strategy reflects a similar desire to avoid adding another fragmented layer to an already complicated banking architecture. Rather than buying large numbers of disconnected applications, AI increasingly needs to sit inside established infrastructure and data environments if banks want to improve decision speed across the organisation. Buying a platform does not eliminate development work because the technology still needs to be integrated with the bank’s own data, controls and operating processes.

A related shift may eventually favour smaller and more specialised language models. Milchanowski predicted that banks could increasingly use smaller models where their capabilities are sufficient because running the largest available model for every task can make little economic sense. A bank may need advanced reasoning for some activities but not for routine extraction, classification or internal Q&A. The return on AI can therefore improve partly through matching model capability and cost to the complexity of individual workloads.

Quantum computing occupies a much earlier position in this development cycle. BMO is researching quantum applications through its work with IBM and has explored areas including optimisation, risk and environmental modelling. One example involves earthquake forecasting, where the bank is researching ways to process extremely large and complex datasets. The broader logic is not that earthquake prediction will suddenly become a banking product, but that methods developed in demanding scientific problems could eventually have applications in capital markets and financial risk.

For financial institutions, the important contrast is therefore between today’s AI investment cycle and the much earlier quantum research cycle. Banks are already being asked to demonstrate measurable returns from AI, while quantum investments can still reasonably be justified through research, readiness and long-term capability building. Confusing those stages could lead organisations either to demand commercial returns too early from quantum research or to accept experimentation for too long in mature AI programmes.

The broader message from the Ai4 discussion is that banking’s AI race may ultimately become less about access to technology. Major financial institutions can increasingly obtain similar models, cloud platforms and development tools. What they cannot purchase as easily is the operating discipline required to redesign workflows, integrate proprietary data, establish governance, train employees and connect AI investment to commercially meaningful outcomes.

That is why AI may increasingly become a margin and execution issue rather than simply a technology issue. Banks using the same underlying models could still produce very different financial results depending on how those models are deployed. The winners may not be the institutions with the largest number of AI pilots, the highest token consumption or even the most advanced individual models. They are more likely to be the banks capable of turning intelligence into faster decisions, stronger client relationships and repeatable economic value across the organisation.

Source: CIJ.World Research & Analysis Team

Artificial intelligence is beginning to change the economics of space, moving the industry beyond satellites that simply collect and transmit information towards increasingly software-defined infrastructure capable of analysing data, managing missions and potentially providing computing capacity in orbit. Speaking at Ai4 2026 in Las Vegas, Heather Pringle, CEO of Space Foundation and a retired U.S. Air Force major general, described AI as increasingly embedded across the space industry, from Earth observation and spacecraft design to mission management, collision avoidance and onboard processing.

The shift is happening alongside rapid expansion of the commercial space economy. Launch costs have fallen, reusable rockets have increased deployment frequency and satellite constellations have become much larger. Governments are also relying more heavily on commercial companies rather than developing every capability internally. Pringle argued that this is helping move space away from a hardware-dominated model towards one that is increasingly based on software and services, because some capabilities can continue evolving after launch rather than remaining fixed for the lifetime of a satellite.

One of the clearest examples comes from Earth observation. Satellites already collect enormous amounts of imagery covering agriculture, weather, infrastructure, environmental change and natural disasters. Historically, much of that information has been transmitted to Earth before being processed. NASA demonstrated a different model in 2026 when researchers successfully deployed the NASA-IBM Prithvi geospatial foundation model aboard two orbital platforms, allowing AI analysis to take place in orbit before all of the underlying data was sent back to terrestrial systems.

The significance goes beyond speed. Satellite communications bandwidth is limited, particularly when large quantities of high-resolution imagery are involved. If AI can identify relevant information in orbit, satellites may only need to transmit the data that matters most. A system monitoring wildfires could prioritise areas where conditions have changed, agricultural monitoring could identify unusual crop patterns and disaster-response satellites could detect flooding or fire damage before the full dataset reaches Earth.

This begins to change the relationship between satellites and terrestrial data centres. Instead of space functioning mainly as a source of raw information for computing infrastructure on Earth, part of the analysis can increasingly happen close to the sensors themselves. The concept resembles edge computing on Earth, where information is processed near the point at which it is generated rather than continually transmitted to a central cloud facility.

That becomes increasingly important as satellite constellations grow. Large networks generate huge volumes of telemetry and imagery while also requiring constant monitoring of satellite condition, orbital position and surrounding traffic. AI can help identify anomalies, predict equipment failures and support decisions about how satellites should manoeuvre. Collision avoidance is becoming particularly important because low-Earth orbit is increasingly crowded with active satellites, debris and planned future constellations.

The implications extend beyond satellite operations because commercialisation is also changing how governments procure space capabilities. Pringle said U.S. agencies increasingly seek to use commercial services where practical rather than attempting to match the speed of private-sector development internally. NASA, the U.S. Space Force and other agencies can therefore become major customers of privately developed satellite communications, imagery, launch capacity and digital services.

This represents a significant shift in the structure of the space economy. Public agencies remain major sources of research funding and mission demand, but private companies increasingly own and operate the infrastructure. That commercial model is already visible in communications and Earth observation and is beginning to extend into lunar services, private space stations, manufacturing and potentially computing.

Among the most ambitious ideas discussed at Ai4 was the development of orbital data centres. The concept has moved from speculative discussion towards actual corporate proposals. SpaceX filed with U.S. regulators in early 2026 for permission to develop an orbital data-centre system involving as many as one million satellites, while other companies have proposed computing constellations involving tens of thousands of spacecraft. These numbers represent proposed systems rather than deployed infrastructure, and their economics, engineering and regulatory feasibility remain uncertain.

Nevertheless, the scale of the proposals shows how seriously some companies are beginning to examine space-based computing. The attraction is closely connected with the terrestrial AI infrastructure boom. AI requires enormous quantities of electricity and increasingly large data-centre campuses, putting pressure on electricity grids, development land, cooling systems and planning processes in major markets.

Space theoretically offers a different resource environment. Solar energy can be available for long periods in orbit, while computing infrastructure would not compete directly with terrestrial uses for development land. But those advantages are offset by major engineering difficulties, particularly cooling. Data centres on Earth remove heat using air or liquid systems. In space there is no atmosphere to support conventional cooling, meaning heat must primarily be rejected through radiation, which creates important limits on how densely computing equipment can operate.

Launch cost, reliability, maintenance and hardware replacement are additional challenges. Terrestrial data centres can replace failed servers continuously, while repairing or upgrading computing infrastructure in orbit is far more complicated. Communications represent another constraint because orbital data centres would only be useful if enormous amounts of information could move efficiently between satellites, ground infrastructure and end users.

Pringle therefore described orbital computing as a technology that still requires considerable development rather than an immediately mature alternative to data centres on Earth. She suggested that meaningful commercial progress could potentially emerge within roughly five to seven years, although this remains her own assessment rather than an established industry forecast.

What is clearer is that the economics are beginning to attract capital. The AI boom has dramatically increased the value of access to power, computing and connectivity, encouraging companies to examine infrastructure concepts that would previously have appeared economically unrealistic. Space infrastructure may therefore begin competing for investment within the wider digital-infrastructure sector.

Investors who traditionally examined fibre networks, terrestrial data centres, towers and cloud infrastructure may increasingly encounter opportunities involving satellite communications, orbital connectivity and space-based computing. The boundaries between these sectors are already becoming less clear because satellite networks ultimately depend on terrestrial fibre, cloud computing, ground stations and physical data centres.

Space infrastructure therefore does not replace terrestrial digital infrastructure. It extends it. Ground stations still require suitable locations, power and connectivity. Launch facilities require large areas of specialised industrial infrastructure. Satellite manufacturing requires advanced production facilities, while space companies also need laboratories, research centres and testing environments. As the number of commercial missions increases, this supporting property ecosystem could expand with it.

The development of private space stations could create another layer of commercial activity. Several companies are working on privately operated orbital platforms intended eventually to supplement or replace some of the functions currently provided by the International Space Station. These facilities could support scientific research, manufacturing and commercial experimentation, including specialised pharmaceutical and advanced-material processes.

AI could make such facilities more practical by reducing the number of tasks requiring continuous human control. Autonomous monitoring, robotics and intelligent mission-management systems could allow orbital platforms to operate with fewer personnel while handling more experiments. This illustrates a broader relationship between AI and the economics of space: because deploying people in orbit remains exceptionally expensive, technologies that allow machines to make more decisions independently can create proportionally greater value.

The same principle applies to exploration. The further spacecraft travel from Earth, the less practical continuous human control becomes because communications delays increase. Autonomous systems therefore become increasingly important for missions to the Moon, Mars and beyond. AI can help spacecraft interpret sensor information, identify hazards, plan movements and respond to unexpected conditions without waiting for instructions from Earth.

This does not mean humans disappear from mission control. It means decisions increasingly have to be divided between those requiring human judgement and those machines can execute safely within predetermined boundaries.

The Artemis programme illustrates the increasingly international nature of major space missions. Artemis II carried astronauts around the Moon in 2026 with significant contributions from international partners, including Europe and Canada. Future space infrastructure is therefore unlikely to be developed by individual countries operating completely independently.

International partnerships can provide both economic scale and operational resilience. That resilience is becoming increasingly important because satellite infrastructure now supports communications, navigation, financial transactions, aviation, logistics, weather forecasting and many other parts of the global economy.

Governments consequently view access to space infrastructure as a strategic issue as well as a commercial one. Satellite networks increasingly need to be designed to continue operating if individual spacecraft fail or are damaged. Large constellations provide one form of resilience because the overall network can potentially continue functioning after losing individual satellites, while multiple orbital layers and links between national and commercial systems provide further redundancy.

The growing strategic importance of orbital infrastructure also creates geopolitical complications. Major economies are investing more heavily in satellite communications, Earth observation and defence-related space capabilities, making resilience, sovereignty and allied cooperation increasingly important parts of infrastructure planning.

The AI race and the space race are therefore beginning to overlap. AI needs enormous quantities of infrastructure, while space increasingly needs AI to manage the complexity created by larger constellations, greater data volumes and more autonomous operations.

That relationship could eventually produce an entirely new category of digital infrastructure extending from terrestrial data centres through fibre networks and ground stations into orbit. Much of that vision remains early, and orbital data centres in particular still face serious technical and economic obstacles. Announcements involving enormous future constellations should not be confused with completed infrastructure.

But the direction of travel is becoming clearer. Space is no longer simply a destination for scientific missions or communications satellites. It is gradually becoming another layer of the digital economy.

For real-estate and infrastructure investors, that means the AI infrastructure story may eventually extend considerably beyond the enormous data-centre campuses currently being built on Earth. The next phase could involve the physical infrastructure connecting terrestrial computing, ground stations, satellite networks and eventually computing systems operating in orbit.

If that happens, the boundary between aerospace and digital infrastructure will become increasingly difficult to define.

Source: CIJ.World Research & Analysis Team

Commercial credit underwriting is beginning to change from a largely sequential process built around documents, spreadsheets and separate software systems into a more integrated operating model in which AI can analyse information continuously, verify calculations and help lenders reach decisions faster. Speaking at Ai4 2026 in Las Vegas, Pranjal Daga, Co-Founder and CEO of Accend, argued that the next stage of commercial lending will depend less on adding individual AI tools and more on connecting the entire underwriting process through a common intelligence layer.

The distinction is significant because commercial credit remains highly fragmented at many financial institutions. A borrower may submit financial statements through one portal, while another system handles document extraction. Analysts then move information into spreadsheets, calculate ratios, retrieve credit information, prepare credit memoranda and transfer the results into a loan-origination platform, while portfolio monitoring may sit somewhere else entirely. Each component may work adequately on its own, but information has to move repeatedly between systems. In many institutions, the analyst effectively becomes the integration layer, reconciling formats, checking figures and manually carrying information across applications.

That creates both cost and risk. Daga presented examples from Accend’s own work in which financial information changed meaning during migrations or manual processing. These examples are company findings rather than industry-wide statistics, but they illustrate a familiar problem in credit analysis: an error introduced near the beginning of a workflow can affect ratios, credit memoranda and ultimately the lending decision itself. AI could reduce some of this friction by allowing information extracted at the beginning of the process to remain connected throughout underwriting.

Instead of document intake, financial spreading, credit analysis, memo creation and portfolio monitoring operating independently, the information could remain inside a shared borrower context. A financial figure used in a credit decision could potentially be traced back to the original statement, page or spreadsheet cell from which it was extracted. If an analyst corrects that figure, the change could then flow through subsequent calculations rather than being amended separately across several systems. This is particularly relevant to commercial real-estate lending, where underwriting can involve financial statements, rent rolls, property operating information, borrower accounts and multiple legal entities. The quality of the decision depends not only on analysing each source correctly but on preserving the relationships between them.

Daga described this broader architecture as an AI credit operating system. The concept combines document collection, financial spreading, credit analysis, memo preparation, portfolio management and scoring within one connected workflow. Specialised AI agents perform different parts of the process, while an orchestration layer determines which tasks should be handled and in what sequence. One agent may extract financial information from documents, another classify individual line items and another prepare a draft credit memorandum. Separate verification processes can then check whether financial statements reconcile or calculations remain internally consistent.

The important point is that AI is not expected to make every decision independently. Daga’s model separates activities where automation is relatively safe from those requiring judgement. Routine document processing, classification and reconciliation can be highly automated, while analysts continue to review exceptions and make decisions where experience or risk judgement matters. This could fundamentally change the work of credit analysts, particularly financial spreading, where considerable time is traditionally spent taking information from company accounts and placing it into standardised financial models before meaningful analysis can begin.

As AI increasingly handles that preparation, the analyst’s role can move towards reviewing information, investigating unusual results and deciding whether the underlying borrower represents an acceptable risk. This reflects a broader pattern emerging across enterprise AI: automation removes some of the mechanical preparation while increasing the importance of judgement, verification and accountability.

For banks and other lenders, the more profound change could come after the loan has been approved. Traditional commercial credit analysis often provides a snapshot of a borrower at a particular point in time. Financial statements are collected, ratios calculated and a credit score or internal rating assigned, followed by another formal review months later. AI combined with transaction information and other frequently updated data potentially moves credit monitoring towards a more continuous process.

Instead of relying primarily on annual or periodic reviews, lenders could monitor changes in payment behaviour, cash flows, financial information and other indicators throughout the life of a loan. A borrower’s risk assessment could therefore change as the underlying business changes. That has important implications for commercial real estate, where banks could potentially identify deterioration in a property or borrower earlier by combining conventional financial reporting with rent collections, operating expenses, debt-service performance and other available signals. The same approach could identify improving credit conditions earlier.

Rather than simply determining whether a loan should have been approved at origination, AI could therefore help lenders continuously assess whether the assumptions supporting the original decision remain valid. This could become increasingly important for portfolios containing offices, retail properties and other assets where occupancy, rents, costs and valuations can change materially between formal annual reviews.

Greater automation, however, creates a higher requirement for verification. One of Daga’s central arguments was that lenders should pay less attention to the accuracy of an individual AI model and more attention to the reliability of the complete system surrounding it. A powerful language model alone is not sufficient for high-stakes credit decisions. The architecture also needs deterministic calculations, reconciliation checks, testing, evaluation, source traceability, appropriate context and human review.

This distinction matters because financial analysis contains many areas where conventional software remains preferable to generative AI. Adding numbers, calculating ratios or confirming that assets equal liabilities does not require probabilistic reasoning. These operations can be performed through deterministic rules, while AI is used for tasks such as interpreting documents, understanding context or preparing narrative analysis. The result is likely to be a hybrid underwriting architecture rather than one large model making autonomous lending decisions: AI agents handle selected tasks, traditional software performs exact calculations, verification systems check outputs and humans remain responsible for judgement and exceptions.

This also raises an important competitive question for financial institutions. As major foundation models become widely available, access to the underlying AI itself may become less differentiated. Banks, fintech companies and specialist lenders can increasingly use many of the same commercially available models. The competitive advantage may therefore move towards proprietary data, workflow design and the historical knowledge accumulated through previous credit decisions.

A lender that retains information about how thousands of previous loans were analysed could potentially create a richer context for future underwriting. The system may understand not simply what a financial statement says, but how the institution historically treated similar businesses, which adjustments its analysts normally make and which risk signals have proved most important. Each completed transaction can potentially add information that improves future analysis, provided the system captures the corrections, decisions and outcomes generated by experienced analysts. In that sense, the valuable asset may increasingly be the accumulated institutional knowledge surrounding the AI rather than the model itself.

Auditability will consequently be critical. Every significant figure entering a credit analysis should ideally remain traceable to its original source. Lenders must be able to establish whether a number came from a financial statement, bank record or another data source, whether it was subsequently modified and whether a human approved the final result. This requirement is particularly important in regulated financial institutions, where faster underwriting cannot come at the expense of explainability.

The same principle applies directly to commercial real-estate lending. If an AI system calculates debt-service coverage, property income or borrower liquidity, the lender should be able to identify exactly which financial information produced the result. Automation without traceability could simply create faster errors.

The commercial impact of faster underwriting could nevertheless be substantial. Financing decisions can take weeks when lenders need to collect documents, spread financial statements and circulate credit memoranda internally. Reducing that process could become a competitive advantage for banks, particularly in commercial property where acquisition and refinancing timetables frequently determine whether transactions proceed. A lender capable of providing greater certainty more quickly may therefore compete not only through pricing but through execution speed.

Faster and cheaper underwriting could also influence which transactions lenders are prepared to consider. If analysts can process more applications without proportionally increasing staffing, financial institutions may be able to evaluate opportunities that were previously too expensive to underwrite relative to their potential revenue. This could expand sophisticated credit analysis further into smaller business loans and mid-sized commercial-property transactions.

There are also significant workforce implications. Daga predicted that financial spreading could effectively disappear as a standalone job as manual preparation becomes increasingly automated. The underlying activity will remain necessary, but credit professionals could spend more time supervising systems, reviewing exceptions, interpreting risk and exercising judgement. Expertise therefore does not necessarily become less valuable. It may become more important because experienced professionals will need to recognise when an automated conclusion is technically consistent but economically misleading.

The emerging credit operating model reflects changes occurring across finance more broadly. AI is gradually moving from isolated productivity tools into the underlying architecture through which companies make decisions. For commercial lenders, the next competitive divide may not be between institutions that use AI and those that do not, because most will eventually have access to similar models. The more important distinction could be between institutions that simply attach AI to existing fragmented workflows and those that redesign underwriting around connected information, continuous monitoring, verification and human judgement.

For commercial real estate, that transition could affect how quickly loans are approved, how borrowers and properties are monitored and how early lenders identify changes in asset performance. It could also make financing processes more responsive to changing market conditions by replacing periodic snapshots with a continuously updated understanding of borrower and asset risk.

The next era of underwriting may therefore be defined less by whether AI participates in credit decisions and more by how effectively financial institutions integrate it into the complete life cycle of a loan. The most valuable system will not necessarily be the one that produces the fastest answer, but the one capable of producing it faster while preserving the context, evidence and accountability required to trust the decision.

Source: CIJ.World Research & Analysis Team