{"version":"1.0","type":"agent_native_article","locale":"en","slug":"ai-budget-investment-wrong-place-enterprise-value-mppv0483","title":"The AI Budget That Hurts Most Isn't the One You Lose, It's the One That Never Reaches Where It Matters","primary_category":"innovation","author":{"name":"Lucía Navarro","slug":"lucia-navarro"},"published_at":"2026-05-28T18:02:50.126Z","total_votes":76,"comment_count":0,"has_map":true,"urls":{"human":"https://sustainabl.net/en/articulo/ai-budget-investment-wrong-place-enterprise-value-mppv0483","agent":"https://sustainabl.net/agent-native/en/articulo/ai-budget-investment-wrong-place-enterprise-value-mppv0483"},"summary":{"one_line":"Enterprise AI spending is concentrated at the model layer while the foundational infrastructure that generates real business value remains chronically underfunded, creating a gap between AI activity and AI results.","core_question":"Why do companies that invest heavily in AI consistently fail to show measurable business value, and what does a budget architecture that actually captures that value look like?","main_thesis":"The AI value gap is not a technology problem but a budgetary architecture problem: investment accumulates at the visible model layer while the four foundations that determine whether models produce results — processes, technical architecture, skills, and data — receive a fraction of the funding they require."},"content_markdown":"## The AI budget that hurts most is not the one that gets lost, but the one that never reaches where it matters\n\nMore than $1.5 trillion in enterprise software valuations evaporated over the last two years. Not for lack of investment in artificial intelligence, but because that investment landed in the wrong place. This is the paradox that defines the current moment: companies have never spent so much on AI and, at the same time, it has never been so difficult to show where the value actually is.\n\nRohit Kedia, CEO of Xoriant, articulated this with precision in a recent analysis: the vast majority of corporate AI budget accumulates at the model layer — that is, in platform licenses, computing infrastructure, vendor partnerships, and proof-of-concept development. That layer captures attention, generates announcements, and produces demonstrations. What it does not produce, consistently, are results that show up on the income statement.\n\nGartner estimated in February 2025 that, through 2026, 60% of AI projects will be abandoned due to a lack of data that is ready to be processed. This is not a technological failure. It is a failure of budgetary architecture: the model was funded, not the infrastructure that sustains it.\n\nThe question that technology leaders should be asking themselves is not whether to invest in AI. That decision has already been made. The question is whether the money is building operational capacity or financing the appearance of modernity.\n\n## The expensive theatre no one wants to name\n\nThere is a pattern that repeats with striking regularity in boardrooms. Every week brings a new announcement of a partnership with a model provider, a demonstration of autonomous agents, a promise of transformed workflows. The noise is loud. The theatre, convincing.\n\nWhen one looks beyond the press releases and asks what actually changed in any concrete way in how the company creates value for its customers, the honest answer is usually disappointing. McKinsey reported in its State of AI 2025 that only one third of companies have begun to scale their AI programs at the enterprise level, even though 88% report active investment. The rest is spending. Just not where it counts.\n\nThis phenomenon has an identifiable structure. AI budgets in 2023 and 2024 lived primarily in innovation or R&D lines, with lax return requirements and light financial oversight. That created the perfect conditions for the proliferation of overlapping tools, departmental pilots disconnected from core processes, and infrastructure contracts that no one measured against a specific business outcome.\n\nThe problem is not that companies experiment. Experimentation has value. The problem is when the experiment becomes the product, when the demonstration replaces deployment, and when the innovation budget functions as a way of appearing modern without committing to anything concrete.\n\nDeloitte found that around 66% of organisations that adopted enterprise AI report improvements in productivity and efficiency as the main benefit obtained. That is a reasonable number. But it must be read carefully: productivity and efficiency are process metrics, not necessarily indicators of structural economic impact. Reducing the time it takes an analyst to prepare a report is not the same as reconfiguring the chain of decisions that makes that report matter.\n\nThe distinction between \"using AI\" and \"applying intelligence\" is, at its core, a budgetary distinction. Using AI means placing tools on top of existing workflows: a copilot here, a chatbot there, an analytics layer over a dashboard that already existed. Applying intelligence means something categorically different: embedding automated decision-making capacity within the way the company produces and delivers value, with traceability back to the outcome that decision affects.\n\nThat second option requires funding things that do not generate headlines: data cleansing, process redesign, modernisation of legacy architectures, team formation. Goldman Sachs noted in its March 2026 report that AI is not eating the software market; it is expanding it, because it reduces the cost of writing code while raising the ceiling of what that code can do. That implies the space of applicable value has grown. But capturing it requires having built the foundations that support it.\n\n## The four foundations that the budget ignores\n\nThere are four areas where investment systematically arrives late or falls short, and all four determine whether any spending on models produces value or simply produces activity.\n\n**Processes** are the first point of failure. AI applied on top of a broken workflow produces broken results faster. Every dollar invested in the model without redesigning the process surrounding it is a dollar that generates speed, not direction. The most frequent mistake in enterprise AI programmes is to assume that the intelligence of the model will compensate for the dysfunction of the process. It does not. It amplifies it.\n\n**Technological architecture** is the second problem. Fragmented legacy systems cannot support intelligence embedded at the point of decision. Purchasing a more powerful model does not resolve an integration problem. What appears to be an AI capability problem is, frequently, an unresolved technical debt problem that the AI budget never touched, because technical debt does not generate attractive demonstrations.\n\n**Skills** occupy the third position and are perhaps the most costly deficit due to their invisibility. There is a difference between a workforce that knows what AI is and one that knows how to work with it. The first can answer an adoption survey. The second can redefine how a team in operations, risk, or customer service makes decisions. The transformation of capabilities remains one of the most consistently underestimated budget lines in enterprise AI programmes, treated as change management at the end of a project rather than as a delivery condition from the outset.\n\n**Data** closes the picture. No model, however sophisticated, produces reliable intelligence from unreliable data. And yet data preparation receives a fraction of the investment directed toward model selection and platform acquisition. Gartner's finding is not merely statistical: it is a diagnosis of priorities. Companies invest where there is visibility and recognition. Clean, well-governed data that is connected to the right processes does not generate spectacular demonstrations. It generates results. That difference explains why 60% of projects do not survive.\n\nA cost study published in 2026 estimates that production-ready AI systems — with genuine regulatory compliance and real scalability — cost between $250,000 and more than one million dollars per system, once engineering, data work, governance, and integration are fully accounted for. That figure includes the recurring costs of model maintenance, retraining, and monitoring. Almost no pilot was designed to sustain that structure. Which explains why pilots do not scale.\n\n## The budgetary architecture that separates those who capture value from those who merely observe it\n\nThe difference between the companies that are capturing sustainable value with AI and those that are accumulating implementation debt is not found in which models they chose. It lies in how they built the investment architecture around those models.\n\nCompanies that are generating measurable returns share a three-layer pattern. The first is foundational investment: data preparation work, process redesign, modernisation of legacy systems, and training programmes with adoption metrics. It is the unglamorous work that determines whether everything else functions. The second layer is intelligence deployment: AI integrated natively into real workflows, not as a parallel tool but as a capability within the platform, the product, or the customer journey, with direct traceability to a business outcome. The third is orchestration — both human and agentic — but it only holds value when the two previous layers have already been built.\n\nDeloitte's projection indicates that the number of companies with more than 40% of their AI projects in production will double in the next planning cycle. That number matters less as an indicator of adoption than as a signal of what type of company will be able to compete on a structurally different cost basis.\n\nChief financial officers are beginning to move AI budgets from innovation lines with lax oversight into operational technology budgets, subject to the same demands applied to an investment in ERP systems or customer relationship management platforms. That has two immediate consequences. The first is that projects unable to demonstrate a measurable operational return will lose funding. The second is that the vendors and systems integrators that survive will be those capable of connecting their offering to concrete process metrics, not to promises of abstract transformation.\n\nThe core argument of Kedia's analysis — and the one that most unsettles technology teams — is that the AI investment that will matter most in the coming year is the one that today appears least attractive. That is not rhetorical paradox. It is a precise description of how value is distributed in markets with high information asymmetry: those who invest in what cannot be demonstrated in a presentation capture the value that others merely describe in their annual reports.\n\nThe budgetary architecture that builds applied intelligence is, by definition, less visible than the one that funds experiments with advanced models. But it is the only one that produces results capable of withstanding a genuine value audit. And in an environment where boards of directors are beginning to demand exactly that, the visibility of spending has ceased to be its principal advantage.","article_map":{"title":"The AI Budget That Hurts Most Isn't the One You Lose, It's the One That Never Reaches Where It Matters","entities":[{"name":"Rohit Kedia","type":"person","role_in_article":"CEO of Xoriant; primary analytical source whose framework structures the article's central argument about budget misallocation"},{"name":"Xoriant","type":"company","role_in_article":"Technology services firm whose CEO provided the core diagnosis of AI budget concentration at the model layer"},{"name":"Gartner","type":"institution","role_in_article":"Source of the statistic that 60% of AI projects will be abandoned through 2026 due to data readiness failures"},{"name":"McKinsey","type":"institution","role_in_article":"Source of State of AI 2025 data showing the gap between reported AI investment and actual enterprise-scale deployment"},{"name":"Deloitte","type":"institution","role_in_article":"Source of data on productivity gains from enterprise AI adoption and projections on production deployment rates"},{"name":"Goldman Sachs","type":"institution","role_in_article":"Source of March 2026 analysis arguing AI is expanding the software market rather than cannibalizing it"},{"name":"Enterprise AI","type":"technology","role_in_article":"Central subject of analysis — the deployment of artificial intelligence at organizational scale and the budget dynamics surrounding it"},{"name":"AI model layer","type":"technology","role_in_article":"The layer of the AI stack — platform licenses, compute, vendor partnerships, PoC development — where corporate budgets disproportionately concentrate"}],"tradeoffs":["Visible AI spending (model licenses, PoC demos, vendor partnerships) vs. invisible foundational investment (data, process redesign, skills) — the former generates recognition, the latter generates results","Speed of AI deployment vs. depth of infrastructure readiness — deploying fast on broken processes amplifies dysfunction rather than correcting it","Innovation budget flexibility with lax oversight vs. operational budget discipline with return requirements — the first enables experimentation, the second enables scaling","Productivity and efficiency gains (measurable but shallow) vs. structural economic impact (harder to measure but strategically significant)","Pilot economics vs. production economics — pilots are designed for demonstration, production systems cost 10x to 100x more and require different governance structures"],"key_claims":[{"claim":"More than $1.5 trillion in enterprise software valuations evaporated over the last two years.","confidence":"high","support_type":"reported_fact"},{"claim":"Gartner estimated in February 2025 that 60% of AI projects will be abandoned through 2026 due to lack of data ready to be processed.","confidence":"high","support_type":"reported_fact"},{"claim":"McKinsey State of AI 2025 reports that only one third of companies have begun to scale AI programs at enterprise level despite 88% reporting active investment.","confidence":"high","support_type":"reported_fact"},{"claim":"Deloitte found that around 66% of organisations that adopted enterprise AI report improvements in productivity and efficiency as the main benefit.","confidence":"high","support_type":"reported_fact"},{"claim":"Goldman Sachs noted in its March 2026 report that AI is expanding the software market, not eating it, by reducing the cost of writing code while raising the ceiling of what that code can do.","confidence":"high","support_type":"reported_fact"},{"claim":"Production-ready AI systems cost between $250,000 and more than $1 million per system once all engineering, data, governance, and integration costs are included.","confidence":"medium","support_type":"reported_fact"},{"claim":"Deloitte projects that the number of companies with more than 40% of AI projects in production will double in the next planning cycle.","confidence":"medium","support_type":"reported_fact"},{"claim":"The AI investment that will matter most in the coming year is the one that today appears least attractive — foundational, unglamorous, and invisible in presentations.","confidence":"high","support_type":"editorial_judgment"}],"main_thesis":"The AI value gap is not a technology problem but a budgetary architecture problem: investment accumulates at the visible model layer while the four foundations that determine whether models produce results — processes, technical architecture, skills, and data — receive a fraction of the funding they require.","core_question":"Why do companies that invest heavily in AI consistently fail to show measurable business value, and what does a budget architecture that actually captures that value look like?","core_tensions":["AI investment is growing while demonstrable business value from that investment is declining — more spend, less return","The most important AI investments are the least visible and least rewarded in organizational recognition systems","Boards demand AI progress while CFOs demand financial discipline — the two pressures point toward different budget architectures","Enterprise AI is simultaneously expanding in scope (Goldman Sachs) and failing to scale in practice (McKinsey, Gartner) — the market opportunity is growing while execution capability lags"],"open_questions":["What governance mechanisms allow organizations to fund unglamorous foundational AI work when organizational incentives reward visible demonstrations?","At what point does the CFO shift from innovation to operational budget classification actually occur in practice, and what triggers it?","How should companies measure the return on foundational AI investments — data preparation, process redesign, skills — when those investments are preconditions rather than direct value generators?","Is the $250,000–$1M+ per production-ready system estimate stable across industries, or does it vary significantly by sector and regulatory environment?","Will the doubling of companies with 40%+ of AI projects in production actually materialize, or will the CFO governance shift reduce total project counts before scaling rates improve?"],"training_value":{"recommended_for":["Chief Financial Officers evaluating AI budget governance structures","Chief Technology Officers diagnosing AI program performance gaps","Strategy and transformation leaders designing enterprise AI roadmaps","Investors assessing enterprise software valuations in an AI-saturated market","Vendors and systems integrators repositioning their value proposition around measurable process outcomes"],"when_this_article_is_useful":["When evaluating an AI investment proposal or budget allocation request","When auditing an existing AI program to diagnose why it is not generating expected returns","When advising a board or CFO on how to restructure AI budget governance","When assessing vendor or systems integrator claims about AI transformation value","When designing a scaling pathway from pilot to production for an enterprise AI initiative","When building a business case for foundational AI investments that lack visible demonstration value"],"what_a_business_agent_can_learn":["How to distinguish between AI spending that generates activity and AI spending that generates measurable business value","The four foundational investment areas — process, architecture, skills, data — that determine whether model investment produces returns","Why pilots systematically fail to scale and what cost structure a production-ready AI system actually requires","How to read productivity and efficiency metrics as process indicators rather than structural economic impact indicators","The three-layer investment architecture shared by companies that generate measurable AI returns","How CFO governance shifts signal market maturity inflection points in enterprise technology cycles"]},"argument_outline":[{"label":"The paradox of high spend and low return","point":"More than $1.5 trillion in enterprise software valuations evaporated over two years not because companies stopped investing in AI but because investment landed at the wrong layer of the stack.","why_it_matters":"Establishes that the problem is structural, not motivational — more budget directed at the same layer will not fix it."},{"label":"The model layer captures attention, not income","point":"Corporate AI budgets concentrate on platform licenses, computing infrastructure, vendor partnerships, and proof-of-concept development — all of which generate announcements but not income statement results.","why_it_matters":"Identifies the specific misallocation pattern so leaders can audit their own budget distribution against it."},{"label":"The expensive theatre pattern","point":"88% of companies report active AI investment but only one third have begun to scale programs at enterprise level, according to McKinsey State of AI 2025. The gap is filled by overlapping tools, disconnected pilots, and infrastructure contracts unmeasured against business outcomes.","why_it_matters":"Quantifies the scale of performative spending and names the organizational conditions that enable it."},{"label":"Using AI vs. applying intelligence","point":"Placing AI tools on top of existing workflows is categorically different from embedding automated decision-making within the way a company produces and delivers value, with traceability back to the affected outcome.","why_it_matters":"Provides a practical distinction that budget owners and technology leaders can use to classify and redirect spending."},{"label":"The four underfunded foundations","point":"Process redesign, legacy architecture modernization, skills transformation, and data preparation are the four areas where investment systematically arrives late or falls short — and all four determine whether model spending produces value or activity.","why_it_matters":"Gives a concrete checklist of where budget reallocation should go before or alongside model investment."},{"label":"The real cost of production-ready AI","point":"Production-ready AI systems with genuine regulatory compliance and real scalability cost between $250,000 and more than $1 million per system once engineering, data work, governance, and integration are fully accounted for. Almost no pilot was designed to sustain that structure.","why_it_matters":"Explains structurally why pilots do not scale and sets realistic cost expectations for enterprise AI deployment."}],"one_line_summary":"Enterprise AI spending is concentrated at the model layer while the foundational infrastructure that generates real business value remains chronically underfunded, creating a gap between AI activity and AI results.","related_articles":[{"reason":"Directly complements the article's argument about foundational AI investment by examining the role of human oversight as a structural requirement for enterprise AI, not a slowdown — both pieces argue against the superficial deployment model","article_id":13161},{"reason":"Challenges the dominant narrative that AI reduces human work, which connects to the article's argument that AI applied without process redesign generates activity rather than value — both examine the gap between AI promise and operational reality","article_id":13049},{"reason":"Examines data governance failures in private markets as AI and analytics scale, directly relevant to the article's claim that data preparation is the most underfunded foundation in enterprise AI programs","article_id":12975}],"business_patterns":["Budget concentration at the most visible layer of a technology stack rather than at the layer that determines outcomes — a pattern that repeats across enterprise technology cycles","Pilot proliferation without scaling pathway — experimentation becomes the product when there is no commitment to production-grade infrastructure","Innovation budget as reputational insurance — spending that signals modernity without committing to measurable operational change","CFO governance shift as a lagging indicator of technology maturity — operational budget discipline arrives after the market has already separated winners from laggards","Information asymmetry advantage — companies that invest in unglamorous foundational work capture value that competitors can only describe"],"business_decisions":["Audit current AI budget distribution across model layer vs. foundational layers before committing to new model investments","Redesign processes before applying AI to them, not after, to avoid amplifying existing dysfunction","Shift AI budget classification from innovation or R&D lines to operational technology budgets with measurable return requirements","Allocate explicit budget lines for data preparation, legacy architecture modernization, and skills transformation as preconditions for model deployment","Evaluate vendors and systems integrators on their ability to connect offerings to concrete process metrics, not transformation narratives","Design pilots with the full cost structure of production-ready systems in mind to avoid non-scalable proof-of-concepts","Establish traceability from AI deployment decisions back to specific business outcomes before approving spend"]}}