The history of intelligence is concerned with the capacity to understand; the history of application is concerned with bringing knowledge to bear upon a particular purpose. Their combination expresses an important distinction between possessing an intellectual capability and using that capability to produce a considered result. Intelligence may exist as knowledge, analytical power, experience or the capacity to reason. Applied intelligence begins when those resources are directed towards a problem, interpreted in its particular circumstances and used to inform a decision or action. The significance lies not simply in what an individual or institution knows, but in how effectively that knowledge is brought to bear upon the circumstances that require judgement.
The word apply derives through Old French from the Latin applicare, meaning to attach, join or bring into connection. Its English history extends to the late fourteenth century, when it acquired the sense of using or employing something for a particular purpose. The adjective applied, in the familiar distinction between practical and theoretical activity, has been recorded since the seventeenth century. The underlying idea is revealing: to apply is to bring something into contact with the matter to which it is relevant. Applied knowledge is therefore not knowledge in the abstract. It is knowledge placed in relation to a specific task, circumstance or objective.
Intelligence has a different but complementary history. The English word entered use in the late fourteenth century through Old French and Latin forms associated with understanding, knowledge and discernment. Its Latin antecedents, intelligentia and intellegentia, are connected with intelligere, meaning to understand or comprehend. Over time, the word came to describe the faculty of understanding, the interpretation and communication of information and the organised acquisition of knowledge in support of informed decisions. Across these developments, its central significance remains the relationship between evidence and understanding.
The expression applied intelligence brings these histories together, but its precise origin requires care. The constituent words have long histories and the combination has been used in different professional, academic, commercial and technological contexts. It does not possess one universally accepted technical definition comparable to a formally established scientific construct and a definitive first use of the complete phrase cannot be established confidently from the available evidence. It would therefore be inappropriate to attribute its invention to a particular individual or organisation without reliable documentary support.
Its intellectual significance, nevertheless, is readily understood. Applied intelligence concerns the conversion of understanding into purposeful judgement. It is the difference between identifying a possible risk and determining what that risk means for a particular underwriting decision; between detecting a pattern and deciding whether the pattern should affect the terms offered; between possessing historical knowledge and using that knowledge to assess a new exposure. The concept directs attention towards the point at which analysis becomes consequential.
For insurance underwriting, this distinction is fundamental. An underwriting agent may possess substantial specialist knowledge, sophisticated analytical tools and extensive information about the risks being considered. Yet those resources create value only when they improve the selection, pricing, structuring or management of risk. Intelligence that remains disconnected from the decision it is intended to inform may be technically impressive without being practically effective. Applied intelligence concerns the disciplined relationship between what is known, what that knowledge establishes and what the professional should do in consequence.
From knowledge to judgement
The distinction between knowledge and its application has long occupied practical and philosophical thought. Knowledge may be accumulated, classified and preserved, but its usefulness depends upon an ability to recognise when it is relevant, understand the conditions under which it can be relied upon and determine how it should influence conduct. A principle may be sound in general yet inappropriate in a particular situation; a method may work reliably under familiar conditions yet fail when its assumptions are no longer valid. Practical intelligence therefore requires more than the possession of knowledge. It requires discernment concerning its appropriate use.
This distinction is especially important in professions where decisions carry consequences for other people and institutions. Professional judgement requires the application of established knowledge to circumstances that may not conform exactly to previous experience. Rules, models and procedures provide necessary structure, but their effective use depends upon understanding their scope and limitations. A professional must recognise when a familiar method is appropriate, when a different interpretation is required and when the evidence is insufficient to justify a confident conclusion.
The development of applied disciplines across science, engineering, medicine and economics reflects this relationship. Theoretical understanding can establish principles and methods; application tests how those principles operate within particular circumstances. In practice, the relationship is reciprocal. Experience may reveal limitations in an established method, creating new questions for investigation and improving subsequent understanding. Application is therefore not merely the final stage of an intellectual process. It can also become a source of further intelligence.
Insurance underwriting provides a compelling example. Historical claims experience and actuarial analysis may establish a useful basis for evaluating an exposure, but the particular risk may possess characteristics that require a different interpretation. The applicant's operations, contractual arrangements, controls, location, dependencies or changing circumstances may alter the significance of the available evidence. An underwriting model may identify a relevant pattern, but a professional must determine whether that pattern applies in the circumstances under consideration and whether it justifies a particular decision.
Applied intelligence consequently requires a continuous relationship between knowledge and action. The decision must be informed by evidence, the evidence must be interpreted in context and the outcome must remain open to evaluation. Where new information emerges, the original judgement may need to be reconsidered. Where a decision produces an unexpected result, the institution should examine whether its reasoning, assumptions or methods require improvement. The objective is not action for its own sake, but action supported by a defensible understanding of the matter in hand.
The modern development of applied intelligence
The twentieth century brought increasingly systematic methods for applying analytical knowledge to complex institutional problems. Statistics, operational research, computing, systems analysis and decision theory enabled organisations to examine larger bodies of evidence, evaluate alternatives and improve the consistency of decisions. These disciplines developed in different contexts, but they shared an interest in connecting information and analytical methods with practical objectives.
The development of artificial intelligence introduced new possibilities into that relationship. Early research explored the extent to which computational systems could perform tasks associated with reasoning, problem-solving and learning. The 1955 proposal for the Dartmouth Summer Research Project on Artificial Intelligence, followed by the 1956 workshop, helped establish artificial intelligence as a distinct field of research. Subsequent developments expanded the ability of machines to classify information, recognise patterns, process language, make predictions and support decisions across a growing range of applications.
Yet the ability to produce an analytical output does not establish that the output has been applied intelligently. A model may identify a statistical relationship that is irrelevant to the decision. A system may generate a plausible explanation without sufficient evidence. An automated process may increase speed while reproducing an inherited error. Computational performance and practical judgement remain distinct, even when they reinforce one another.
The development of more capable artificial intelligence therefore makes the concept of applied intelligence increasingly important. As machines become capable of performing more information-intensive tasks, the central question shifts from whether a task can be automated towards whether the resulting capability improves the decision for which it is intended. This requires an understanding of the problem, a clear objective, appropriate evidence, reliable methods and a process through which the outcome can be evaluated.
Applied intelligence is consequently not a synonym for artificial intelligence, nor is it confined to any particular technology. It concerns the purposeful use of intelligence, whether human, machine-based or institutional, in the circumstances where it can contribute to a better decision. Artificial intelligence may greatly extend the resources available to a professional. Applied intelligence describes the disciplined process through which those resources are converted into professional capability.
The distinctive position of the underwriting agent
Insurance underwriting agents occupy a distinctive position within the insurance ecosystem. They may possess specialist knowledge of particular classes of business, markets, industries or forms of exposure and may operate under delegated authority granted by an insurer or other capacity provider. Their work can combine the assessment of individual risks with the management of a portfolio, the application of underwriting appetite, the operation of referral procedures and the maintenance of relationships with brokers, insurers and other market participants.
This position makes the quality of applied intelligence especially consequential. An underwriting agent must make decisions where information is often incomplete, where specialist knowledge may be necessary to understand the exposure and where the characteristics of the individual risk must remain consistent with the broader objectives and constraints of the portfolio. The agent must consider not only whether a risk appears attractive in isolation, but whether it is adequately understood, appropriately priced, within authority and consistent with the intended accumulation of exposure.
The professional title Chartered Insurance Underwriting Agent has a defined status within the United Kingdom's Chartered Insurance Institute framework. The title recognises appropriately qualified professionals working in underwriting agency roles, with the CII identifying specialist knowledge, professional standards and ethical conduct as central elements of the designation. The title is therefore not simply a general description of someone who underwrites insurance. It denotes a recognised professional standing within the insurance sector.
Applied intelligence does not replace the technical or ethical standards of that profession. It concerns how the intelligence available to the professional can be organised and used to strengthen underwriting decisions. Its focus is the point at which specialist knowledge meets a particular exposure: the interpretation of a submission, the assessment of uncertainty, the calibration of price and terms, the exercise of delegated authority and the management of the resulting portfolio.
The central problem is not merely that underwriting agents need more information. They need the right information, in the right context, at the point where it can influence a decision. A detailed analysis delivered after a risk has been bound may be of limited value to the original selection decision. A model output that cannot be interpreted by the underwriter may add little to the quality of judgement. A portfolio report that identifies a concentration without connecting it to the relevant underwriting authority may fail to produce a useful response.
Applied intelligence therefore begins with the professional decision and works backwards towards the evidence, analysis and capabilities necessary to support it. This distinguishes a purposeful intelligence framework from technology deployment undertaken for its own sake.
Intelligence at the point of decision
The practical value of intelligence depends upon its relationship with the decision it is intended to inform. In underwriting, that relationship may involve accepting or declining a risk, determining a price, establishing terms and conditions, referring a case for further consideration, setting limits, applying a deductible or reviewing an existing underwriting position. Each decision requires a different combination of evidence and professional judgement.
The first requirement is to establish precisely what must be decided. An underwriter assessing a new submission needs to understand the nature of the exposure, the reliability of the information provided, the factors that materially influence the probability or severity of loss and the extent to which the risk falls within the intended appetite. A portfolio manager may need to determine whether a series of individually acceptable risks is producing an unintended concentration. A delegated-authority manager may need to establish whether the business being written remains consistent with the agreed parameters and the capacity provider's expectations.
Artificial intelligence can assist by assembling relevant information, examining documents, identifying inconsistencies, comparing risks and highlighting evidence that merits further investigation. Its contribution is meaningful when those capabilities help the professional answer the question more rigorously or efficiently. The presence of technology alone does not establish that the underwriting decision has improved.
The discipline is therefore to connect analysis directly to a decision. What does the system reveal that was not previously apparent? Which assumption does it challenge? What uncertainty does it help resolve? What remains unknown? Is the resulting information material to the risk and what professional action, if any, is justified? These questions help distinguish intelligence that supports underwriting from intelligence that merely adds another layer of reporting.
The objective is not to ensure that every analytical finding produces an intervention. Sometimes the appropriate response is to change a price or term; sometimes it is to seek further information, refer the risk or decline it. In other circumstances, the evidence may confirm that the existing decision remains appropriate. Applied intelligence is demonstrated by the quality of the response, including the decision not to act when the evidence does not justify change.
The application of intelligence to risk selection
Risk selection is one of the most consequential functions of underwriting because the quality of the portfolio is shaped by the decisions through which individual exposures are accepted. A technically proficient pricing process cannot fully compensate for consistently selecting risks that are poorly understood or inconsistent with the intended appetite. Conversely, a disciplined selection process can improve the quality of a portfolio before claims experience provides definitive evidence of its performance.
Applied intelligence seeks to strengthen selection by helping underwriters evaluate individual risks within a wider analytical context. Artificial intelligence may assist in comparing submission information with relevant historical cases, examining the characteristics of peer risks, identifying gaps in documentation, detecting contradictory statements and surfacing combinations of factors that warrant additional scrutiny. It can also help identify where a risk falls outside familiar experience, making it easier to direct specialist attention towards cases in which conventional rules may be insufficient.
These capabilities should support, rather than replace, professional evaluation. A risk that differs from previous cases is not necessarily unacceptable; it may represent an opportunity to apply specialist knowledge in a new area. Similarly, a risk that resembles a previously profitable exposure may possess an important difference that makes the comparison misleading. The task is to understand the significance of similarities and differences, not to treat historical classifications as definitive answers.
The underwriter must therefore retain the responsibility to determine whether the evidence is credible, whether the comparison is appropriate and whether the risk can be accepted on terms consistent with the portfolio's objectives. Artificial intelligence can enlarge the evidential field and make relevant comparisons more accessible. It cannot remove the need to decide what those comparisons mean in the circumstances.
A mature application of intelligence seeks to make risk selection more discriminating, not simply more automated. It should help the organisation recognise where ordinary processes are adequate, where additional investigation is required and where the available evidence does not justify acceptance. The objective is better selection supported by a stronger understanding of the exposure.
Pricing, terms and the expression of uncertainty
Insurance pricing is an exercise in translating an assessment of risk into a commercial decision. The price must be considered in relation to expected losses, uncertainty, expenses, capital, reinsurance where relevant, market conditions and the strategy of the underwriting portfolio. Policy terms, exclusions, limits and deductibles also affect the risk transferred. Price and terms therefore form part of a single assessment of the relationship between the exposure and the contractual protection offered.
A price may be calculated precisely while still resting upon an inadequate understanding of the risk. Historical data may be poorly matched to the submission; a material exposure may be missing from the analysis; a relationship assumed to be stable may have changed; or the model may not adequately reflect the circumstances of the particular risk. Applied intelligence seeks to improve the evidence and reasoning that inform pricing, rather than treating the pricing output as an end in itself.
Artificial intelligence can assist in analysing historical experience, comparing relevant exposures, examining changes in risk characteristics and investigating alternative scenarios. It may help identify where a proposed price depends heavily upon a particular assumption, or where emerging information suggests that a familiar comparator deserves reconsideration. Such findings can strengthen the underwriter's understanding of the relationship between the risk, the price and the conditions under which the price remains defensible.
Nevertheless, the output of a model cannot determine the commercial decision by itself. The appropriate price may also depend upon the terms offered, the quality of risk controls, the capacity available, the insurer's appetite and the wider portfolio context. A model that estimates a loss distribution does not automatically establish what price should be charged and the existence of a competitive market does not make an inadequate price technically sound.
Applied intelligence provides a means of connecting analytical evidence with these broader considerations. Its purpose is to make pricing and terms more informed, better calibrated and more open to scrutiny, while preserving the professional judgement and commercial responsibility inherent in underwriting.
Applied intelligence and delegated authority
Delegated underwriting authority gives applied intelligence an additional institutional dimension. Authority is granted within defined parameters and its effective operation depends upon the ability of the relevant parties to establish that decisions remain within those parameters, the intended risk appetite and applicable requirements. This is not simply a matter of measuring performance after the event. It involves understanding the character of the business being written, the way the portfolio is developing and whether the arrangements continue to operate as intended.
Guidance published by Lloyd's on oversight and monitoring of delegated authority emphasises proportionate oversight, adequate expertise, meaningful management information, continuous monitoring, escalation of material issues and accountable decision-making. The guidance recognises that oversight must reflect the nature, scale and complexity of the arrangement and that responsibility cannot be removed merely because activities are undertaken by third parties.
Applied intelligence is relevant to this setting because it concerns the conversion of information into effective oversight. Artificial intelligence may help identify deviations from expected portfolio composition, changes in referral patterns, deteriorating performance, data inconsistencies or exposures that warrant further examination. It can support the comparison of underwriting activity with delegated parameters and help direct professional attention towards matters where the evidence suggests that review is necessary.
The purpose is not to allow a machine to redefine the terms of a delegation or exercise authority outside the agreed governance arrangements. Rather, it is to improve the visibility of the information required by those responsible for exercising, supervising or reviewing that authority. The relevant parties must still determine whether a deviation is material, whether it reflects a legitimate change in the business and what action is appropriate under the contractual and regulatory framework.
This distinction is fundamental. Delegated authority is both a commercial arrangement and an allocation of decision-making responsibility. Better information can strengthen the arrangement, but it cannot substitute for clear authority, adequate expertise, appropriate controls and accountable judgement. Applied intelligence is valuable when it makes those responsibilities easier to discharge effectively.
Portfolio intelligence and the significance of accumulation
Individual underwriting decisions acquire their full significance when considered collectively. A series of risks may each appear acceptable while producing an aggregate exposure that is inconsistent with the intended portfolio. Concentration may emerge through a geography, industry, supplier, technology, peril, counterparty or economic dependency. The relevant relationship may not be visible within any one submission and its importance may change as the external environment develops.
Applied intelligence extends the underwriting perspective beyond the individual risk towards the structure of the portfolio. Artificial intelligence can assist in identifying common characteristics across exposures, examining concentration, comparing changes over time and connecting risks that conventional classifications may place in separate categories. It can also help bring portfolio findings back to the underwriter, allowing individual decisions to be considered against the broader consequences of the aggregate book.
The challenge is to distinguish a genuine accumulation from a superficial similarity. Two insureds may operate in the same industry while possessing very different risk characteristics; two apparently unrelated businesses may share a vulnerability through a common supplier or infrastructure. An analytical model may identify an association, but professional assessment is required to determine whether that association creates a material exposure and how it should affect underwriting decisions.
Portfolio intelligence is therefore not a purely statistical exercise. It requires an understanding of how losses may develop, which dependencies may become consequential under stress and what the institution is seeking to achieve through its risk appetite. Applied intelligence can help ensure that portfolio information is not confined to periodic reporting, but contributes to decisions about selection, pricing, referrals, authority and exposure management.
The objective is a more coherent relationship between the individual risk and the portfolio of which it forms part. The underwriter must understand both the merits of the submission and its contribution to the collective position. Applied intelligence helps connect these perspectives without treating an algorithmic indication of concentration as a substitute for professional judgement.
Exceptions, claims and the intelligence of experience
Insurance underwriting is inevitably shaped by exceptions. An unusual submission, an unfamiliar contractual arrangement, a claim that develops differently from expectation or a combination of circumstances outside the standard classification may reveal something important about the risk. Yet conventional systems can be designed around normal cases and may treat deviations principally as problems to be resolved. Applied intelligence approaches exceptions differently: as events that may require investigation because they challenge the current understanding.
An exception may represent an error, an unusual but acceptable risk, an emerging opportunity or evidence that an existing model is insufficient. The appropriate response depends upon the facts and the consequences. Artificial intelligence can help identify exceptions at scale, compare them with previous cases and establish which differences deserve further consideration. Professional judgement then determines whether the deviation is material and what action is justified.
Claims provide another important source of learning. They reveal how insured events occur, how losses develop, which assumptions remain sound and where the underwriting process may have misunderstood an exposure. These lessons can inform future risk selection, pricing, terms, referral criteria and portfolio strategy. Yet claims experience must be interpreted carefully. A single loss does not necessarily demonstrate that the original decision was unreasonable, just as the absence of a loss does not establish that the risk was adequately understood.
Applied intelligence seeks to connect the original underwriting decision with subsequent evidence. It should be possible, where records permit, to examine what was known at inception, which assumptions informed the decision, what later occurred and what can reasonably be learned. This creates a more disciplined feedback process between underwriting and claims, while reducing the danger of evaluating earlier decisions solely with the benefit of hindsight.
The objective is not to turn every exception into a new rule or every claim into a verdict on underwriting quality. It is to improve the institution's capacity to identify meaningful patterns, reconsider methods and learn from experience. Intelligence becomes applied when those lessons influence future decisions in a manner supported by evidence and proportionate to the circumstances.
The professional and machine relationship
The introduction of artificial intelligence into underwriting changes the scale at which information can be examined, but it does not remove the need for professional responsibility. Machines may classify, compare, summarise, monitor and identify patterns. Underwriters must establish whether those outputs are reliable, relevant and sufficient to influence the decision. The distinction becomes more important as the volume and apparent sophistication of automated analysis increase.
A model may produce a confident recommendation that depends upon incomplete data, inappropriate comparators or assumptions that do not hold for the risk under consideration. An automated system may identify a relationship without establishing its cause. A workflow may appear efficient while making it difficult to determine who authorised the decision or how the output was evaluated. These are not reasons to reject artificial intelligence; they are reasons to design its application carefully.
Applied intelligence requires a deliberate allocation of responsibilities. Routine decisions supported by reliable evidence may be suitable for a high degree of automation within defined boundaries. Unusual, ambiguous or consequential cases may require referral and deeper human investigation. The system should help identify where professional attention is most valuable, rather than attempting to eliminate human involvement regardless of context.
The underwriter's role becomes more intellectually demanding as information processing becomes more capable. The professional must define the question, set the evidential standard, understand the method, examine the output, challenge the assumptions and decide what action is justified. This requires familiarity with the limits of artificial intelligence, including data quality, model drift, bias, uncertainty and the distinction between prediction and explanation.
The strongest relationship is therefore neither one of uncritical reliance upon machines nor one of refusing technological assistance. It is a disciplined partnership in which computation expands the field of enquiry and professional judgement determines the significance and use of the resulting intelligence. The purpose is to make the underwriter more capable, while keeping authority and accountability clear.
APPLIED INTELLIGENCE® at GENERAL INTELLIGENCE PLC
APPLIED INTELLIGENCE® is an artificial intelligence consultancy framework provided by X from GENERAL INTELLIGENCE PLC for chartered insurance underwriting agents. It addresses the strategic, technical and institutional dimensions of applying intelligence at the point where insurance risk is selected, priced, structured and managed. Its purpose is to examine how artificial intelligence can improve the relationship between evidence and professional action within underwriting organisations, while preserving the judgement, accountability and discipline upon which effective underwriting depends.
The framework begins with the decision rather than with the technology. It considers what the underwriting agent needs to know, which evidence is material, where present processes are constrained and how a more comprehensive intelligence capability might improve the outcome. It examines the application of artificial intelligence to submission analysis, risk selection, pricing support, referral, delegated-authority oversight, portfolio intelligence, claims feedback and institutional learning. These are not independent technological projects. Their value depends upon their contribution to a coherent underwriting process.
APPLIED INTELLIGENCE® treats intelligence as a capability that must reach the point at which it can change a decision. It is concerned with ensuring that relevant information is available in context, that analytical findings can be interpreted by the professional, that material exceptions receive appropriate attention and that the consequences of underwriting decisions can inform future practice. The framework therefore extends beyond efficiency to consider the quality of judgement, the consistency of decision-making and the institution's ability to understand the portfolio it is constructing.
Its application must remain specific to the underwriting agent and the business undertaken. The requirements of a specialist managing general agent may differ materially from those of another underwriting organisation, depending upon the delegated authority, classes of business, distribution arrangements, portfolio objectives, data available and governance obligations. The framework does not assume that every process should be automated or that every use of artificial intelligence will create value. It provides a basis for evaluating where technology can make a demonstrable contribution and where other forms of analytical or organisational improvement may be more appropriate.
The designation APPLIED INTELLIGENCE® expresses this particular consultancy purpose. The phrase has a broader linguistic and professional history and its first use cannot be attributed confidently to a single originator on the evidence available. GENERAL INTELLIGENCE PLC uses it to identify its defined artificial intelligence consultancy framework for chartered insurance underwriting agents. Its significance lies in its application to the practical responsibilities of underwriting: the conversion of relevant evidence into defensible decisions, the management of authority and exposure and the development of an institution capable of learning from the consequences of its own actions.
The applied intelligence test
The ultimate test of APPLIED INTELLIGENCE® is whether intelligence has become more consequential within the underwriting process.
- Does it improve risk selection? Does it strengthen the evidential basis of pricing and terms?
- Does it identify material information before the decision is made?
- Does it distinguish routine submissions from those requiring specialist attention?
- Does it support the effective oversight of delegated authority?
- Does it reveal portfolio concentrations that are difficult to recognise from individual risks?
- Does it treat exceptions as potential sources of knowledge?
- Does it connect claims experience with future underwriting decisions?
The institutional test is equally important.
- Can the underwriting agent preserve professional judgement while making fuller use of machine capability?
- Can it understand where a model may be unreliable?
- Can it distinguish a statistical association from a material risk relationship?
- Can it demonstrate that authority is exercised within appropriate boundaries and that consequential decisions remain accountable?
- Can the organisation learn from its own history without assuming that historical experience will always provide an adequate guide?
These questions define a more demanding standard than the volume of data processed or the number of decisions automated. They concern whether the institution has improved its capacity to act intelligently upon what it knows. A system that generates extensive analysis but does not change the quality of the decision has added activity, not necessarily value. A system that exposes a material weakness, supports a better-calibrated price or directs professional attention towards an otherwise overlooked risk may make a genuine contribution.
Applied intelligence must therefore remain purposeful and proportionate. It should strengthen the evidence available to the underwriter, improve the connection between analysis and action and support learning from the outcomes of decisions. It should not confer unwarranted authority upon a model or treat the presence of technology as evidence of superior underwriting.
The enduring proposition
The intellectual history of applied intelligence rests upon a distinction between the possession of knowledge and its effective use. The underlying words have long histories, while the combined phrase has been employed across different fields without acquiring one universally accepted technical definition or a reliably established single point of origin. Its application to insurance underwriting gives it a particular professional significance: the disciplined conversion of evidence into the acceptance, pricing, structuring and management of risk.
APPLIED INTELLIGENCE® brings this proposition into the domain of consultancy for chartered insurance underwriting agents. It considers how artificial intelligence can help make relevant information more accessible, establish relationships across different sources, identify material exceptions, support delegated-authority oversight and connect the consequences of underwriting decisions with future practice. Its ambition is not to replace professional judgement, but to make that judgement better informed, more consistent and more capable of learning.
The future of underwriting will not be determined simply by the volume of information available or the sophistication of the machines used to analyse it. It will depend upon whether underwriting institutions can identify what matters, evaluate evidence rigorously, exercise authority responsibly and translate their understanding into decisions that remain defensible in changing circumstances. Technology may enlarge the field of analysis, but institutional design determines whether the resulting intelligence is applied effectively.
For GENERAL INTELLIGENCE PLC, this is the central proposition of APPLIED INTELLIGENCE®: intelligence must not remain remote from the decisions it is intended to inform. It must reach the underwriter in a form that can be understood, challenged and used; it must support action proportionate to the evidence; and it must help the institution learn from the consequences of its decisions.