{"id":5031,"date":"2026-07-17T09:29:14","date_gmt":"2026-07-17T07:29:14","guid":{"rendered":"https:\/\/www.meltone.com\/can-we-trust-generative-ai-in-finance\/"},"modified":"2026-08-07T09:43:49","modified_gmt":"2026-08-07T07:43:49","slug":"can-we-trust-generative-ai-in-finance","status":"publish","type":"post","link":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/","title":{"rendered":"Can we trust generative AI in Finance?"},"content":{"rendered":"\n\n\n<section    \n        class=\"block-hero block-hero-article theme-dark deco-article align-left\"\n        data-halo\n        data-bg-dark=\"true\"\n        >\n                <div class=\"hero-deco\" aria-hidden=\"true\">\n                                            <svg viewBox=\"0 0 1440 738\" preserveAspectRatio=\"xMidYMid slice\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n  <defs>\n    <radialGradient id=\"articleGlow\" cx=\"0.5\" cy=\"0.5\" r=\"0.5\">\n      <stop stop-color=\"#11163A\"\/>\n      <stop offset=\"1\" stop-color=\"#404672\" stop-opacity=\"0\"\/>\n    <\/radialGradient>\n  <\/defs>\n  <!-- Glow navy centre-droite (Ellipse 57) -->\n  <circle cx=\"1233\" cy=\"615\" r=\"485\" fill=\"url(#articleGlow)\"\/>\n  <!-- Cercle jaune coin bas-droite (Ellipse 101) -->\n  <circle cx=\"1445.5\" cy=\"766.5\" r=\"250.5\" fill=\"#FCFDD5\"\/>\n  <!-- Ruban pastel (Vector 34, pivot\u00e9) -->\n  <g transform=\"translate(1394.08 447.89) rotate(-47.93) translate(-1109.87 -445.02)\">\n    <path d=\"M61.82 28.5275C194.533 316.12 578.292 46.9245 653.141 382.104C654.417 535.861 589.028 689.62 708.744 764.513C856.261 845.08 975.976 736.712 1020.8 616.429C1065.62 496.146 1153.49 283.658 1327.58 313.766C1503.1 344.122 1514.25 624.794 1765.03 775.716C1879.9 844.848 2032.71 839.997 2180.79 736.735\" stroke=\"#CED9FA\" stroke-width=\"136.17\"\/>\n  <\/g>\n<\/svg>\n\n                    <\/div>\n    \n    <span class=\"halo\"><\/span>\n    <div class=\"container\">\n                                <div class=\"breadcrumb\">\n            <ul class=\"breadcrumb-list\">\n                <li class=\"breadcrumb-list_item\">\n                    <a href=\"https:\/\/www.meltone.com\/en\/\">Accueil<\/a>\n                <\/li>\n\n                                                        \n                                                                                                                                                                                                                                                                                                \n                                                                                                    \n                                        <li class=\"breadcrumb-list_item\"><\/li>\n                            <\/ul>\n        <\/div>\n    \n                <div class=\"grid\">\n            <div class=\"grid_content\">\n                                \n                                \n                                        \n                        \n                    \n            <h1\n        class=\"title title-hero\"\n                        >Can we trust generative AI in Finance?<\/h1>\n    \n                                                                \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>Here we will analyze the unnatural but relevant link\u2014if placed under control\u2014between a Finance function anchored in its need for truth and the arrival of generative AI* at its service, whose strength is precisely to invent narratives but also figures. An irreconcilable paradox? <\/p>\n\n        <\/div>\n    \n                                                    <span class=\"block-hero-article_tag\">Opinion<\/span>\n                                                    <div class=\"block-hero-article_meta\">\n                                                    <span class=\"author-by\">\n                                R\u00e9dig\u00e9 par\n                                                                    <a class=\"author\" href=\"https:\/\/www.meltone.com\/en\/author\/l-deleuse\/\">l.deleuse<\/a>\n                                                            <\/span>\n                                                <span class=\"date\">17 July 2026<\/span>                    <\/div>\n                            <\/div>\n        <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone\"\n                >\n                <div class=\"block-content-text-meltone_deco\" aria-hidden=\"true\">\n            <span class=\"deco deco--circle\"><svg viewBox=\"0 0 361 361\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <circle cx=\"180.5\" cy=\"180.5\" r=\"180.5\" fill=\"#404672\"\/>\n<\/svg>\n<\/span>\n            <span class=\"deco deco--arabesque\"><svg viewBox=\"0 0 699 1104.12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <path d=\"M490.629 52.3639C180.108 186.874 -55.8918 558.409 103.573 660.574C243.033 749.921 491.159 636.844 579.833 551.803C691.514 444.698 654.91 330.416 530.048 359.901C387.578 393.544 248.471 577.054 237.206 839.268C226.498 1088.51 463.222 968.525 637.798 774.562C641.376 770.588 648.324 775.163 645.893 779.926C608.135 853.903 537.488 996.6 530.048 1051.76\" stroke=\"#CED9FA\" stroke-width=\"104.701\" stroke-linecap=\"round\"\/>\n<\/svg>\n<\/span>\n        <\/div>\n            <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Finance is also a story of explainability<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>Beyond the figure that allows a decision to be made, the explainability of that figure is also key. When announcing to a BU manager overseeing several billion that they are not meeting their objectives and that, consequently, corrective action will be taken, one must be able to argue, document, and be sure of the sources. This is the notion of explainability.  <\/p>\n<p>And we regularly come across management controllers who spend one hour producing an analysis and another hour checking the calculation method and sources. How many meetings have ended after 5 minutes because no one was aligned on the figures? <\/p>\n<p>It is in the DNA of Finance, and it also allows for direct action without hesitating over the numbers.<\/p>\n<blockquote>\n<figure class=\"wp-block-table\">\n<table>\n<tbody>\n<tr>\n<td><strong>Key takeaway:<\/strong><br \/>\nA figure without a source or documented method is worthless to a BU manager.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n<\/blockquote>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone is-dark\"\n        data-halo        >\n                    <span class=\"halo\" aria-hidden=\"true\"><\/span>\n        <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Generative AI is first and foremost a story of creativity<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>Generative AI is a family of algorithms based on LLMs* that has the fascinating ability to generate natural language text that appears completely realistic. And for good reason: the major LLMs on the market have analyzed the entire global web and are capable, from a fragment of a sentence or a question, of predicting the most credible sequences of words relative to the context. These are not exactly the most relevant; they are the words that are statistically most probable according to its digital vision of the world. Therefore, it is not the most accurate result via logical reasoning that is produced, but the generation of the most probable.   <\/p>\n<p>Another important nuance: to give an even greater illusion of naturalness, the word sequences are not always those with the highest probability but are drawn at random from among the most probable. Why? To give an impression of naturalness. If you ask it the same question twice, it will phrase the answer in two different ways with small nuances. This moves even further away from a deterministic logic which, via documented deduction, leads to a stable result.    <\/p>\n<p>It is on this principle that ChatGPT, Claude, Mistral, or Gemini respond to you so fluidly, regardless of the context.<\/p>\n<p style=\"text-align: center;\"><strong>Key takeaway:<\/strong><br \/>\n<strong>An LLM generates the most probable word, not the truest.<\/strong><\/p>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone\"\n                >\n                <div class=\"block-content-text-meltone_deco\" aria-hidden=\"true\">\n            <span class=\"deco deco--circle\"><svg viewBox=\"0 0 361 361\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <circle cx=\"180.5\" cy=\"180.5\" r=\"180.5\" fill=\"#404672\"\/>\n<\/svg>\n<\/span>\n            <span class=\"deco deco--arabesque\"><svg viewBox=\"0 0 699 1104.12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <path d=\"M490.629 52.3639C180.108 186.874 -55.8918 558.409 103.573 660.574C243.033 749.921 491.159 636.844 579.833 551.803C691.514 444.698 654.91 330.416 530.048 359.901C387.578 393.544 248.471 577.054 237.206 839.268C226.498 1088.51 463.222 968.525 637.798 774.562C641.376 770.588 648.324 775.163 645.893 779.926C608.135 853.903 537.488 996.6 530.048 1051.76\" stroke=\"#CED9FA\" stroke-width=\"104.701\" stroke-linecap=\"round\"\/>\n<\/svg>\n<\/span>\n        <\/div>\n            <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Generative AI knows neither how to count nor how to explain itself: and that is its main risk<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>An LLM alone has no computational logic. If it answers &#8220;2+2=&#8221; correctly, it is not because it calculated it, but because it has seen this answer millions of times in its training data. As soon as we move away from simple or highly documented calculations, an LLM alone can &#8220;hallucinate&#8221;: invent a result that appears real, without ever detecting the error, and sometimes build upon it with complete confidence.  <\/p>\n<p>This limitation is aggravated by a second problem: unexplainability. Even when an LLM is trained to break down its reasoning into more readable steps, these steps remain probabilities, never a logical and reproducible deduction that could be audited step-by-step. For your function, which thrives on traceability and argumentation before a BU manager, this is the number one point of vigilance.  <\/p>\n<p>The good news: this risk is identified, understood, and we already know how to put it under control; that is the whole subject of the rest of this article.<\/p>\n<blockquote>\n<figure class=\"wp-block-table\">\n<table>\n<tbody>\n<tr>\n<td><strong>Key takeaway:<\/strong><br \/>\nAn LLM alone does not calculate and cannot justify its reasoning step-by-step.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n<\/blockquote>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone is-dark\"\n        data-halo        >\n                    <span class=\"halo\" aria-hidden=\"true\"><\/span>\n        <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >How to reconcile AI hallucination and Finance?<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>We can see that on paper, the raw combination of Finance and generative AI seems antinomical and explosive. Yet there is enormous potential if two essential rules are respected: <\/p>\n<ul>\n<li style=\"text-align: left;\">Knowing its limits and strengths, only apply generative AI where it makes sense&#8230;<\/li>\n<li style=\"text-align: left;\">Put it under control and validate it&#8230;<\/li>\n<\/ul>\n<p>Nothing very new, in short. Let&#8217;s detail this approach. <\/p>\n<p>&nbsp;<\/p>\n<p style=\"text-align: center;\"><strong>Key takeaway:<\/strong><br \/>\n<strong>Only apply AI where it makes sense, and always validate it.<\/strong><\/p>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone\"\n                >\n                <div class=\"block-content-text-meltone_deco\" aria-hidden=\"true\">\n            <span class=\"deco deco--circle\"><svg viewBox=\"0 0 361 361\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <circle cx=\"180.5\" cy=\"180.5\" r=\"180.5\" fill=\"#404672\"\/>\n<\/svg>\n<\/span>\n            <span class=\"deco deco--arabesque\"><svg viewBox=\"0 0 699 1104.12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <path d=\"M490.629 52.3639C180.108 186.874 -55.8918 558.409 103.573 660.574C243.033 749.921 491.159 636.844 579.833 551.803C691.514 444.698 654.91 330.416 530.048 359.901C387.578 393.544 248.471 577.054 237.206 839.268C226.498 1088.51 463.222 968.525 637.798 774.562C641.376 770.588 648.324 775.163 645.893 779.926C608.135 853.903 537.488 996.6 530.048 1051.76\" stroke=\"#CED9FA\" stroke-width=\"104.701\" stroke-linecap=\"round\"\/>\n<\/svg>\n<\/span>\n        <\/div>\n            <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Generative AI is a bad calculator but an excellent analyst<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <h2><\/h2>\n<p>In Finance, you don&#8217;t just work with numbers. The decisions you are asked to make are certainly based on figures, but also on contextual information which can be narrative, graphic, video&#8230; Market context, competitor behavior, the socio-economic climate, evolving standards, your own internal rules from one division to another&#8230; these are all formats that this type of algorithm is relevant for linking, analyzing, and synthesizing. Specifically trained for Finance, AI agents* are excellent analysts that can absorb and contextualize a mass of information that you could not process otherwise.  <\/p>\n<blockquote>\n<figure class=\"wp-block-table\">\n<table class=\" aligncenter\">\n<tbody>\n<tr>\n<td><strong>Key takeaway:<\/strong><br \/>\nGenerative AI excels at analyzing context, not at calculating.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n<\/blockquote>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone is-dark\"\n        data-halo        >\n                    <span class=\"halo\" aria-hidden=\"true\"><\/span>\n        <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Generative AI is the voice and the pilot of your calculator<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <h2><\/h2>\n<p>While LLMs do not know how to calculate, they are the new translators and analysts between you and your complex tools.<\/p>\n<p>Let&#8217;s take a concrete example: explaining an 8% variance on a cost line compared to the budget during a closing. Traditionally, a management controller manually crosses several sources\u2014extractions from their EPM tool, comments sent back by subsidiaries, history from previous months\u2014to write a reasoned variance comment. This work easily takes 30 to 45 minutes per significant line, repeated over dozens of lines each closing.  <\/p>\n<p>With a correctly configured AI agent, the mechanics change: the calculation tool (your EPM) produces the exact and verifiable variance: the deterministic data does not change hands. The AI agent then reads the subsidiary comments, cross-references them with the history, identifies recurring causes, and writes an initial structured draft of the comment. The management controller then spends their time no longer searching and writing, but verifying, challenging, and enriching, reducing the work to 10-15 minutes of quality control rather than 30-45 minutes of production.  <\/p>\n<p>The gain is therefore not in producing a different figure: the figure remains the same, produced by the same reliable calculation tool. But it frees up the management controller&#8217;s time from the most time-consuming and least analytical part of the work. Tools perform deterministic* calculations that can be 100% certain, and LLMs handle writing the context and transmitting the correct hypotheses to them. Each solution focuses on its strengths for a combination that multiplies the potential of your teams.   <\/p>\n<p>&nbsp;<\/p>\n<p style=\"text-align: center;\"> <strong>Key takeaway:<\/strong><br \/>\n<strong>The figure remains the same; it is the context writing time that disappears.<\/strong><\/p>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone\"\n                >\n                <div class=\"block-content-text-meltone_deco\" aria-hidden=\"true\">\n            <span class=\"deco deco--circle\"><svg viewBox=\"0 0 361 361\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <circle cx=\"180.5\" cy=\"180.5\" r=\"180.5\" fill=\"#404672\"\/>\n<\/svg>\n<\/span>\n            <span class=\"deco deco--arabesque\"><svg viewBox=\"0 0 699 1104.12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <path d=\"M490.629 52.3639C180.108 186.874 -55.8918 558.409 103.573 660.574C243.033 749.921 491.159 636.844 579.833 551.803C691.514 444.698 654.91 330.416 530.048 359.901C387.578 393.544 248.471 577.054 237.206 839.268C226.498 1088.51 463.222 968.525 637.798 774.562C641.376 770.588 648.324 775.163 645.893 779.926C608.135 853.903 537.488 996.6 530.048 1051.76\" stroke=\"#CED9FA\" stroke-width=\"104.701\" stroke-linecap=\"round\"\/>\n<\/svg>\n<\/span>\n        <\/div>\n            <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Small specialized AI agents are easier to control<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>To reduce hallucinations in your future agents, a simple technique consists of breaking down the tasks assigned to them into sub-tasks, each assigned to a dedicated agent. This technique makes it easier to train these algorithms and reduces their hallucinations, as they are less exposed to contexts they do not know. <\/p>\n<p>Thus, one could have a single agent to whom all questions are asked\u2014documentary research, P&#038;L analysis, consolidation of subsidiary comments, identification of causes, or production of reporting&#8230; But it would be extremely complex to train correctly and very sensitive to any change in scope or data.<\/p>\n<p>We therefore favor a collection of agents that collaborate with each other. This often includes a &#8220;Pilot &#038; Synthesis&#8221; agent that receives questions and redirects them to the right sub-agents (analyst, formatting, web search&#8230;) and synthesizes the results. Each is more specialized and easier to control and evolve.  <\/p>\n<blockquote>\n<figure class=\"wp-block-table\">\n<table class=\" aligncenter\">\n<tbody>\n<tr>\n<td> <strong> Key takeaway:<\/strong><br \/>\nSeveral specialized agents are better than a single generalist agent.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n<\/blockquote>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone is-dark\"\n        data-halo        >\n                    <span class=\"halo\" aria-hidden=\"true\"><\/span>\n        <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >AI agents are very good at self-monitoring<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>In this galaxy of agents that will equip your department, many will be dedicated to monitoring and challenging the results of their peers. We know that LLMs hallucinate, but we can train other LLMs to detect these hallucinations, poor report layout, figures without real references&#8230; This is now a standard. Every LLM process has its integrated LLM safeguard.  <\/p>\n<p>&nbsp;<\/p>\n<p style=\"text-align: center;\"> <strong>Key takeaway:<\/strong><br \/>\n<strong>Every AI agent should have its own AI safeguard.<\/strong><\/p>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone\"\n                >\n                <div class=\"block-content-text-meltone_deco\" aria-hidden=\"true\">\n            <span class=\"deco deco--circle\"><svg viewBox=\"0 0 361 361\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <circle cx=\"180.5\" cy=\"180.5\" r=\"180.5\" fill=\"#404672\"\/>\n<\/svg>\n<\/span>\n            <span class=\"deco deco--arabesque\"><svg viewBox=\"0 0 699 1104.12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <path d=\"M490.629 52.3639C180.108 186.874 -55.8918 558.409 103.573 660.574C243.033 749.921 491.159 636.844 579.833 551.803C691.514 444.698 654.91 330.416 530.048 359.901C387.578 393.544 248.471 577.054 237.206 839.268C226.498 1088.51 463.222 968.525 637.798 774.562C641.376 770.588 648.324 775.163 645.893 779.926C608.135 853.903 537.488 996.6 530.048 1051.76\" stroke=\"#CED9FA\" stroke-width=\"104.701\" stroke-linecap=\"round\"\/>\n<\/svg>\n<\/span>\n        <\/div>\n            <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Good governance and data quality improve the relevance of AI agents<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>This is the part no one wants to read. No, generative AI is not magic. If your data is chaotic, siloed, inconsistent from one source to another, and with business definitions of varying scope, then AI or no AI, you will produce hallucinations = figures you believe are real but have no reality. Adding an autonomous AI agent to these areas greatly increases the potential for hallucination.   <\/p>\n<p>Exploiting the potential of generative AI therefore implies up-to-standard governance and data quality. AI can help improve quality, but it cannot do it without you. <\/p>\n<blockquote><p> <strong>Key takeaway:<\/strong><br \/>\nWithout quality data, AI amplifies hallucinations rather than avoiding them.<\/p><\/blockquote>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone is-dark\"\n        data-halo        >\n                    <span class=\"halo\" aria-hidden=\"true\"><\/span>\n        <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Generative AI also raises the question of where your data goes<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>There is a risk we haven&#8217;t addressed yet that often causes concern faster than hallucination: what happens to data once it is entered into a generative AI tool? There is a fundamental difference between consumer use, where data can be kept and reused by the publisher, and secure enterprise use, where it is contractually protected and not reused for model training. <\/p>\n<p>Pasting a P&#038;L extract or a supplier negotiation into an unvalidated tool is equivalent to letting sensitive data leave your control perimeter\u2014this is what is called &#8220;Shadow AI.&#8221; The rule is simple: the same confidentiality reflexes that apply to an email apply to an AI prompt. And this is not a regulatory vacuum: the European AI Act and your internal controls (SOX or equivalent) apply to the results produced by an AI exactly as they do to any other result.  <\/p>\n<p style=\"text-align: center;\"><strong>Key takeaway:<\/strong><br \/>\n<strong>An AI prompt deserves the same confidentiality reflexes as an email.<\/strong><\/p>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-content-text-meltone\"\n                >\n                <div class=\"block-content-text-meltone_deco\" aria-hidden=\"true\">\n            <span class=\"deco deco--circle\"><svg viewBox=\"0 0 361 361\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <circle cx=\"180.5\" cy=\"180.5\" r=\"180.5\" fill=\"#404672\"\/>\n<\/svg>\n<\/span>\n            <span class=\"deco deco--arabesque\"><svg viewBox=\"0 0 699 1104.12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\" aria-hidden=\"true\" focusable=\"false\">\n    <path d=\"M490.629 52.3639C180.108 186.874 -55.8918 558.409 103.573 660.574C243.033 749.921 491.159 636.844 579.833 551.803C691.514 444.698 654.91 330.416 530.048 359.901C387.578 393.544 248.471 577.054 237.206 839.268C226.498 1088.51 463.222 968.525 637.798 774.562C641.376 770.588 648.324 775.163 645.893 779.926C608.135 853.903 537.488 996.6 530.048 1051.76\" stroke=\"#CED9FA\" stroke-width=\"104.701\" stroke-linecap=\"round\"\/>\n<\/svg>\n<\/span>\n        <\/div>\n            <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Always, always keep the human in the loop<\/h2>\n    \n                                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>Even with these precautions, risks of hallucination always remain. Yes, our analytical power will be multiplied, yes our time to action will be reduced, but let us never neglect the time for monitoring and validating results. It is in our nature, in our duties, and it should be natural. We remain responsible for the figures and analyses produced. Part of the evolution of our role will be to understand when and why they hallucinate and to enrich the safeguards and controls we apply to them.    <\/p>\n<blockquote>\n<figure class=\"wp-block-table\">\n<table>\n<tbody>\n<tr>\n<td><strong>Key takeaway:<\/strong><br \/>\nAI multiplies your analytical power, but you remain responsible for the result.<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<\/figure>\n<\/blockquote>\n\n        <\/div>\n    \n            <\/div>\n                    <\/div>\n    <\/div>\n<\/section>\n\n\n<section    \n        class=\"block-content block-content-5-meltone\"\n        >\n    <div class=\"block-content-5-meltone_deco\" aria-hidden=\"true\">\n        <div class=\"block-content-5-meltone_glow\">\n            <svg preserveAspectRatio=\"none\" width=\"100%\" height=\"100%\" overflow=\"visible\" viewBox=\"0 0 970 970\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n<circle cx=\"485\" cy=\"485\" r=\"485\" fill=\"url(#content5Glow)\"\/>\n<defs>\n<radialGradient id=\"content5Glow\" cx=\"0\" cy=\"0\" r=\"1\" gradientUnits=\"userSpaceOnUse\" gradientTransform=\"translate(485 485) rotate(90) scale(485)\">\n<stop stop-color=\"#CED9FA\"\/>\n<stop offset=\"1\" stop-color=\"#CED9FA\" stop-opacity=\"0\"\/>\n<\/radialGradient>\n<\/defs>\n<\/svg>\n\n        <\/div>\n        <div class=\"block-content-5-meltone_yellow\">\n            <svg preserveAspectRatio=\"none\" width=\"100%\" height=\"100%\" overflow=\"visible\" viewBox=\"0 0 542 542\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n<circle cx=\"271\" cy=\"271\" r=\"271\" fill=\"#FCFDD5\"\/>\n<\/svg>\n\n        <\/div>\n    <\/div>\n    <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_title\">\n                                \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >In summary: There is therefore no paradox in mixing Finance and generative AI.<\/h2>\n    \n            <\/div>\n            <div class=\"grid_content\">\n                                            \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>It is an extremely powerful new duo that does not just perform existing tasks faster, but opens up new analyses and projections that were previously humanly impossible. Its integration must be progressive, applying it where it makes sense, while controlling its potential deviations and, above all, never abdicating our role as those responsible for the result. <\/p>\n\n        <\/div>\n    \n                            <\/div>\n        <\/div>\n    <\/div>\n<\/section>\n\n\n        \n<section    \n        class=\"block-content block-content-4-meltone\"\n        >\n    <div class=\"block-content-4-meltone_deco\" aria-hidden=\"true\">\n        <div class=\"block-content-4-meltone_glow block-content-4-meltone_glow--left\">\n            <svg preserveAspectRatio=\"none\" width=\"100%\" height=\"100%\" overflow=\"visible\" viewBox=\"0 0 970 970\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n<circle cx=\"485\" cy=\"485\" r=\"485\" fill=\"url(#content4GlowLeft)\"\/>\n<defs>\n<radialGradient id=\"content4GlowLeft\" cx=\"0\" cy=\"0\" r=\"1\" gradientUnits=\"userSpaceOnUse\" gradientTransform=\"translate(485 485) rotate(90) scale(485)\">\n<stop stop-color=\"#9DD5D5\"\/>\n<stop offset=\"1\" stop-color=\"#9DD5D5\" stop-opacity=\"0\"\/>\n<\/radialGradient>\n<\/defs>\n<\/svg>\n\n        <\/div>\n        <div class=\"block-content-4-meltone_glow block-content-4-meltone_glow--right\">\n            <svg preserveAspectRatio=\"none\" width=\"100%\" height=\"100%\" overflow=\"visible\" viewBox=\"0 0 970 970\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n<circle cx=\"485\" cy=\"485\" r=\"485\" fill=\"url(#content4GlowRight)\"\/>\n<defs>\n<radialGradient id=\"content4GlowRight\" cx=\"0\" cy=\"0\" r=\"1\" gradientUnits=\"userSpaceOnUse\" gradientTransform=\"translate(485 485) rotate(90) scale(485)\">\n<stop stop-color=\"#CED9FA\"\/>\n<stop offset=\"1\" stop-color=\"#CED9FA\" stop-opacity=\"0\"\/>\n<\/radialGradient>\n<\/defs>\n<\/svg>\n\n        <\/div>\n    <\/div>\n    <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_title\">\n                                \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Also discover<\/h2>\n    \n            <\/div>\n                            <div class=\"grid_content\">\n                                                \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>By <a href=\"https:\/\/www.linkedin.com\/in\/stephane-portier\/\">St\u00e9phane Portier<\/a>, Innovation Manager at MeltOne:<\/p>\n<p><a href=\"https:\/\/www.meltone.com\/news\/limpact-ia-en-finance-comment-vos-architectures-evoluent\/\">The impact of AI in Finance: how are your architectures evolving?<\/a><\/p>\n\n        <\/div>\n    \n                <\/div>\n                                        <div class=\"grid_media grid_media-video\">\n                                <div class=\"plyr-player\">\n            <iframe loading=\"lazy\" class=\"plyr\" allow=\"autoplay\" title=\"Quelle place pour l\u2019IA G\u00e9n\u00e9rative dans la finance ?\" width=\"640\" height=\"360\" src=\"https:\/\/www.youtube.com\/embed\/XrDfVGlLvXM?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe>\n        <\/div>\n    \n                <\/div>\n                                <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-faq block-faq-2-meltone accordion-root\"\n        >\n    <div class=\"block-faq-2-meltone_glow\" aria-hidden=\"true\">\n        <svg preserveAspectRatio=\"none\" width=\"100%\" height=\"100%\" overflow=\"visible\" viewBox=\"0 0 970 970\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n<g id=\"faq2-glow\">\n<circle cx=\"485\" cy=\"485\" r=\"485\" fill=\"url(#faq2MeltoneGlow)\"\/>\n<circle cx=\"335.5\" cy=\"407.5\" r=\"271.5\" fill=\"#ffffff\"\/>\n<\/g>\n<defs>\n<radialGradient id=\"faq2MeltoneGlow\" cx=\"0\" cy=\"0\" r=\"1\" gradientUnits=\"userSpaceOnUse\" gradientTransform=\"translate(485 485) rotate(90) scale(485)\">\n<stop stop-color=\"#CED9FA\"\/>\n<stop offset=\"1\" stop-color=\"#CED9FA\" stop-opacity=\"0\"\/>\n<\/radialGradient>\n<\/defs>\n<\/svg>\n\n    <\/div>\n    <div class=\"block-faq-2-meltone_content\">\n        <div class=\"container\">\n            <div class=\"grid\">\n                <div class=\"grid_title\">\n                                    \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >FAQ<\/h2>\n    \n                <\/div>\n                                    <div class=\"grid_content\">\n                                                    <div class=\"accordion-item\">\n                                <div class=\"accordion-item_header\">\n                                    <h3 class=\"question\">\n                                        <button class=\"question_btn\"\n                                                type=\"button\"\n                                                aria-expanded=\"false\"\n                                                aria-controls=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-1\"\n                                                id=\"faq-q-block_a305a59cdfccc449692f641b2406a745-1\">\n                                            Can generative AI be trusted to produce Finance figures?\n                                        <\/button>\n                                    <\/h3>\n                                    <div class=\"icon\" aria-hidden=\"true\">\n                                        <svg width=\"8\" height=\"12\" viewBox=\"0 0 8 12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n                                            <path d=\"M3.34113 11.3549C3.53712 11.5494 3.8537 11.5482 4.04823 11.3522L7.21832 8.15841C7.41285 7.96242 7.41167 7.64584 7.21569 7.45131C7.0197 7.25677 6.70312 7.25796 6.50858 7.45394L3.69072 10.2929L0.851769 7.47504C0.65578 7.2805 0.3392 7.28168 0.144667 7.47767C-0.0498658 7.67366 -0.0486854 7.99024 0.147304 8.18477L3.34113 11.3549ZM3.65234 0L3.15235 0.00186431L3.19336 11.0019L3.69336 11L4.19336 10.9981L4.15234 -0.00186431L3.65234 0Z\" fill=\"white\"\/>\n                                        <\/svg>\n                                    <\/div>\n                                <\/div>\n                                <div class=\"accordion-item_content\"\n                                     id=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-1\"\n                                     role=\"region\"\n                                     aria-labelledby=\"faq-q-block_a305a59cdfccc449692f641b2406a745-1\">\n                                                                \n                                    \n                \n        <div class=\"editor answer\"\n                                                        >\n        <p>No, not alone. An LLM can hallucinate on complex calculations. Figures must always come from a deterministic calculation tool (EPM, ERP&#8230;), with generative AI handling context and synthesis.  <\/p>\n<p>&nbsp;<\/p>\n\n        <\/div>\n    \n                                <\/div>\n                            <\/div>\n                                                    <div class=\"accordion-item\">\n                                <div class=\"accordion-item_header\">\n                                    <h3 class=\"question\">\n                                        <button class=\"question_btn\"\n                                                type=\"button\"\n                                                aria-expanded=\"false\"\n                                                aria-controls=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-2\"\n                                                id=\"faq-q-block_a305a59cdfccc449692f641b2406a745-2\">\n                                            What is the main risk of generative AI in Finance?\n                                        <\/button>\n                                    <\/h3>\n                                    <div class=\"icon\" aria-hidden=\"true\">\n                                        <svg width=\"8\" height=\"12\" viewBox=\"0 0 8 12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n                                            <path d=\"M3.34113 11.3549C3.53712 11.5494 3.8537 11.5482 4.04823 11.3522L7.21832 8.15841C7.41285 7.96242 7.41167 7.64584 7.21569 7.45131C7.0197 7.25677 6.70312 7.25796 6.50858 7.45394L3.69072 10.2929L0.851769 7.47504C0.65578 7.2805 0.3392 7.28168 0.144667 7.47767C-0.0498658 7.67366 -0.0486854 7.99024 0.147304 8.18477L3.34113 11.3549ZM3.65234 0L3.15235 0.00186431L3.19336 11.0019L3.69336 11L4.19336 10.9981L4.15234 -0.00186431L3.65234 0Z\" fill=\"white\"\/>\n                                        <\/svg>\n                                    <\/div>\n                                <\/div>\n                                <div class=\"accordion-item_content\"\n                                     id=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-2\"\n                                     role=\"region\"\n                                     aria-labelledby=\"faq-q-block_a305a59cdfccc449692f641b2406a745-2\">\n                                                                \n                                    \n                \n        <div class=\"editor answer\"\n                                                        >\n        <p>Two main risks: hallucination (an invented result that appears real) and data security (Shadow AI, use of tools not validated by the organization).<\/p>\n\n        <\/div>\n    \n                                <\/div>\n                            <\/div>\n                                                    <div class=\"accordion-item\">\n                                <div class=\"accordion-item_header\">\n                                    <h3 class=\"question\">\n                                        <button class=\"question_btn\"\n                                                type=\"button\"\n                                                aria-expanded=\"false\"\n                                                aria-controls=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-3\"\n                                                id=\"faq-q-block_a305a59cdfccc449692f641b2406a745-3\">\n                                            How can the risks of AI agent hallucination be reduced?\n                                        <\/button>\n                                    <\/h3>\n                                    <div class=\"icon\" aria-hidden=\"true\">\n                                        <svg width=\"8\" height=\"12\" viewBox=\"0 0 8 12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n                                            <path d=\"M3.34113 11.3549C3.53712 11.5494 3.8537 11.5482 4.04823 11.3522L7.21832 8.15841C7.41285 7.96242 7.41167 7.64584 7.21569 7.45131C7.0197 7.25677 6.70312 7.25796 6.50858 7.45394L3.69072 10.2929L0.851769 7.47504C0.65578 7.2805 0.3392 7.28168 0.144667 7.47767C-0.0498658 7.67366 -0.0486854 7.99024 0.147304 8.18477L3.34113 11.3549ZM3.65234 0L3.15235 0.00186431L3.19336 11.0019L3.69336 11L4.19336 10.9981L4.15234 -0.00186431L3.65234 0Z\" fill=\"white\"\/>\n                                        <\/svg>\n                                    <\/div>\n                                <\/div>\n                                <div class=\"accordion-item_content\"\n                                     id=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-3\"\n                                     role=\"region\"\n                                     aria-labelledby=\"faq-q-block_a305a59cdfccc449692f641b2406a745-3\">\n                                                                \n                                    \n                \n        <div class=\"editor answer\"\n                                                        >\n        <p>By breaking down tasks between specialized agents that are easier to train and control, adding dedicated quality control agents, and systematically maintaining human validation.<\/p>\n\n        <\/div>\n    \n                                <\/div>\n                            <\/div>\n                                                    <div class=\"accordion-item\">\n                                <div class=\"accordion-item_header\">\n                                    <h3 class=\"question\">\n                                        <button class=\"question_btn\"\n                                                type=\"button\"\n                                                aria-expanded=\"false\"\n                                                aria-controls=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-4\"\n                                                id=\"faq-q-block_a305a59cdfccc449692f641b2406a745-4\">\n                                            Does generative AI replace the management controller?\n                                        <\/button>\n                                    <\/h3>\n                                    <div class=\"icon\" aria-hidden=\"true\">\n                                        <svg width=\"8\" height=\"12\" viewBox=\"0 0 8 12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n                                            <path d=\"M3.34113 11.3549C3.53712 11.5494 3.8537 11.5482 4.04823 11.3522L7.21832 8.15841C7.41285 7.96242 7.41167 7.64584 7.21569 7.45131C7.0197 7.25677 6.70312 7.25796 6.50858 7.45394L3.69072 10.2929L0.851769 7.47504C0.65578 7.2805 0.3392 7.28168 0.144667 7.47767C-0.0498658 7.67366 -0.0486854 7.99024 0.147304 8.18477L3.34113 11.3549ZM3.65234 0L3.15235 0.00186431L3.19336 11.0019L3.69336 11L4.19336 10.9981L4.15234 -0.00186431L3.65234 0Z\" fill=\"white\"\/>\n                                        <\/svg>\n                                    <\/div>\n                                <\/div>\n                                <div class=\"accordion-item_content\"\n                                     id=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-4\"\n                                     role=\"region\"\n                                     aria-labelledby=\"faq-q-block_a305a59cdfccc449692f641b2406a745-4\">\n                                                                \n                                    \n                \n        <div class=\"editor answer\"\n                                                        >\n        <p>No. It automates research, cross-referencing, and first-level writing tasks, but the responsibility for the result and its validation remains human. <\/p>\n\n        <\/div>\n    \n                                <\/div>\n                            <\/div>\n                                                    <div class=\"accordion-item\">\n                                <div class=\"accordion-item_header\">\n                                    <h3 class=\"question\">\n                                        <button class=\"question_btn\"\n                                                type=\"button\"\n                                                aria-expanded=\"false\"\n                                                aria-controls=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-5\"\n                                                id=\"faq-q-block_a305a59cdfccc449692f641b2406a745-5\">\n                                            Glossary\n                                        <\/button>\n                                    <\/h3>\n                                    <div class=\"icon\" aria-hidden=\"true\">\n                                        <svg width=\"8\" height=\"12\" viewBox=\"0 0 8 12\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\">\n                                            <path d=\"M3.34113 11.3549C3.53712 11.5494 3.8537 11.5482 4.04823 11.3522L7.21832 8.15841C7.41285 7.96242 7.41167 7.64584 7.21569 7.45131C7.0197 7.25677 6.70312 7.25796 6.50858 7.45394L3.69072 10.2929L0.851769 7.47504C0.65578 7.2805 0.3392 7.28168 0.144667 7.47767C-0.0498658 7.67366 -0.0486854 7.99024 0.147304 8.18477L3.34113 11.3549ZM3.65234 0L3.15235 0.00186431L3.19336 11.0019L3.69336 11L4.19336 10.9981L4.15234 -0.00186431L3.65234 0Z\" fill=\"white\"\/>\n                                        <\/svg>\n                                    <\/div>\n                                <\/div>\n                                <div class=\"accordion-item_content\"\n                                     id=\"faq-ans-block_a305a59cdfccc449692f641b2406a745-5\"\n                                     role=\"region\"\n                                     aria-labelledby=\"faq-q-block_a305a59cdfccc449692f641b2406a745-5\">\n                                                                \n                                    \n                \n        <div class=\"editor answer\"\n                                                        >\n        <p><strong>Generative AI<\/strong>: a family of algorithms based on LLMs, capable of generating text, images, or other content in natural language from an instruction.<br \/>\n<strong>LLM (Large Language Model)<\/strong>: a language model trained on immense volumes of text, which predicts the most probable sequence of words in a given context.<br \/>\n<strong>AI Agent<\/strong>: a program built around one or more LLMs, specialized in a business task (analysis, research, synthesis, etc.) and capable of interacting with other tools.<br \/>\n<strong>Deterministic calculation<\/strong>: a calculation whose result is always identical for the same input data, as opposed to a probabilistic generation of the LLM type<\/p>\n\n        <\/div>\n    \n                                <\/div>\n                            <\/div>\n                                            <\/div>\n                            <\/div>\n        <\/div>\n    <\/div>\n<\/section>\n\n\n\n<section    \n        class=\"block-content block-cta-contact cta--variation-1\"\n        >\n    <div class=\"container\">\n        <div class=\"grid\">\n            <div class=\"grid_content\">\n                <div class=\"card\">\n                    <div class=\"card_content\">\n                        <div class=\"card_deco\" aria-hidden=\"true\">\n                            <svg viewBox=\"0 0 520 412\" preserveAspectRatio=\"xMidYMid slice\" fill=\"none\" xmlns=\"http:\/\/www.w3.org\/2000\/svg\"><svg fill=\"none\" x=\"-108\" y=\"315\" width=\"306\" height=\"306\" viewBox=\"0 0 306.0 306.0\" overflow=\"visible\">\n<circle id=\"Ellipse 59\" cx=\"153\" cy=\"153\" r=\"153\" fill=\"#9DD5D5\"\/>\n<\/svg><svg fill=\"none\" x=\"-43.0\" y=\"165.0\" width=\"597.0\" height=\"597.0\" viewBox=\"0 0 597.0 597.0\" overflow=\"visible\">\n<circle id=\"Ellipse 58\" cx=\"298.5\" cy=\"298.5\" r=\"298.5\" fill=\"url(#g0)\"\/>\n<defs>\n<radialGradient id=\"g0\" cx=\"0\" cy=\"0\" r=\"1\" gradientUnits=\"userSpaceOnUse\" gradientTransform=\"translate(298.5 298.5) rotate(90) scale(298.5)\">\n<stop stop-color=\"#292F5C\"\/>\n<stop offset=\"1\" stop-color=\"#292F5C\" stop-opacity=\"0\"\/>\n<\/radialGradient>\n<\/defs>\n<\/svg><g transform=\"translate(608.05 285.95) rotate(90) scale(1 -1) translate(-384.994 -544.06)\"><svg fill=\"none\" width=\"769.988\" height=\"1088.12\" viewBox=\"0 0 769.988 1088.12\" overflow=\"visible\">\n<path id=\"Vector 17\" d=\"M719.976 50.0125C493.848 111.657 366.458 246.835 366.458 414.156C366.458 623.307 491.644 739.514 578.481 688.914C677.66 631.123 654.738 473.158 491.644 390.379C328.55 307.599 95.8099 378.05 53.4937 627.27C29.7479 767.12 130.355 923.829 345.529 1038.11\" stroke=\"#292F5C\" stroke-width=\"100\" stroke-linecap=\"round\"\/>\n<\/svg><\/g><\/svg>\n\n                        <\/div>\n                        <div class=\"card_text\">\n                                            \n                                \n                                        \n                        \n                    \n            <h2\n        class=\"title title-section\"\n                        >Let\u2019s discuss your challenges and our solutions<\/h2>\n    \n                                                        \n                                    \n                \n        <div class=\"editor\"\n        >\n        <p>Let\u2019s talk about your project and discover how MeltOne can turn your challenges into concrete, high-performing solutions.<\/p>\n\n        <\/div>\n    \n                                                            <div class=\"action\">\n                                                                                \n    \n    \n                \n    <a href=\"https:\/\/www.meltone.com\/en\/contact\/\"\n       class=\"btn btn-light has-icon has-icon--right\"\n                               aria-label=\"Contact us\"                            >\n        <span class=\"btn_title\">Contact us<\/span>\n                    <span class=\"btn_icon\"><svg xmlns=\"http:\/\/www.w3.org\/2000\/svg\" viewBox=\"0 0 12 8\" fill=\"none\"><path d=\"M11.3549 4.02215C11.5494 3.82617 11.5482 3.50959 11.3522 3.31505L8.15841 0.144959C7.96242 -0.0495739 7.64584 -0.0483934 7.45131 0.147595C7.25677 0.343584 7.25795 0.660165 7.45394 0.854697L10.2929 3.67256L7.47504 6.51151C7.2805 6.7075 7.28168 7.02408 7.47767 7.21861C7.67366 7.41315 7.99024 7.41197 8.18477 7.21598L11.3549 4.02215ZM0 3.71094L0.00186433 4.21093L11.0019 4.16992L11 3.66992L10.9981 3.16993L-0.00186433 3.21094L0 3.71094Z\" fill=\"currentColor\"\/><\/svg><\/span>\n            <\/a>\n\n                                                                    <\/div>\n                                                    <\/div>\n                    <\/div>\n                                            <div class=\"card_image\">\n                                        \n                                                                                                                                            \n                <picture >\n\n                                        <source media=\"(min-width: 1024px)\" srcset=\"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/img-meltone-9-1200x0-c-default.jpg 1x, https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/img-meltone-9-2400x0-c-default.jpg 2x\">\n\n                                        <source media=\"(min-width: 768px)\" srcset=\"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/img-meltone-9-800x0-c-default.jpg 1x, https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/img-meltone-9-1600x0-c-default.jpg 2x\">\n\n                                        <img src=\"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/img-meltone-9-400x0-c-default.jpg\"\n                         srcset=\"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/img-meltone-9-400x0-c-default.jpg 1x, https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/img-meltone-9-800x0-c-default.jpg 2x\"                         alt=\"Img Meltone 9\"\n                          width=\"2400\"                           height=\"1599\"                                                                                loading=\"lazy\" decoding=\"async\"\n                                             >\n                <\/picture>\n                        \n                        <\/div>\n                                    <\/div>\n            <\/div>\n        <\/div>\n    <\/div>\n<\/section>\n","protected":false},"excerpt":{"rendered":"<p>The impact of AI in Finance on your architecture: AI is integrated through successive layers in each brick, without revolution, through progressive enrichment. The autonomous agents layer is the one that carries the true productive value and is enriching the fastest. MCP improves communication between applications without specific development. Data quality and governance [\u2026]  <\/p>\n","protected":false},"author":4,"featured_media":5032,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"content-type":"","footnotes":""},"categories":[154],"tags":[169,160,161,166],"cas-client-offre":[],"class_list":["post-5031","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-opinion","tag-ai","tag-erp","tag-performance-management","tag-smart-erp"],"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.4 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>The Impact of AI in Finance: How Are Your Architectures Evolving?<\/title>\n<meta name=\"description\" content=\"Discover the impact of AI in Finance on ERP, EPM, and Data Platforms: AI agents, MCP, data governance, and the new Finance architecture.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"The Impact of AI in Finance: How Are Your Architectures Evolving?\" \/>\n<meta property=\"og:description\" content=\"Discover the impact of AI in Finance on ERP, EPM, and Data Platforms: AI agents, MCP, data governance, and the new Finance architecture.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/\" \/>\n<meta property=\"og:site_name\" content=\"MeltOne\" \/>\n<meta property=\"article:published_time\" content=\"2026-07-17T07:29:14+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-08-07T07:43:49+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1200\" \/>\n\t<meta property=\"og:image:height\" content=\"630\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"l.deleuse\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"l.deleuse\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/\"},\"author\":{\"name\":\"l.deleuse\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#\\\/schema\\\/person\\\/71719c78de779a1e8cd0a511ccc47963\"},\"headline\":\"Can we trust generative AI in Finance?\",\"datePublished\":\"2026-07-17T07:29:14+00:00\",\"dateModified\":\"2026-08-07T07:43:49+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/\"},\"wordCount\":7,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.meltone.com\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp\",\"keywords\":[\"AI\",\"ERP\",\"Performance management\",\"Smart ERP\"],\"articleSection\":[\"Opinion\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/\",\"url\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/\",\"name\":\"The Impact of AI in Finance: How Are Your Architectures Evolving?\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/www.meltone.com\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp\",\"datePublished\":\"2026-07-17T07:29:14+00:00\",\"dateModified\":\"2026-08-07T07:43:49+00:00\",\"description\":\"Discover the impact of AI in Finance on ERP, EPM, and Data Platforms: AI agents, MCP, data governance, and the new Finance architecture.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/can-we-trust-generative-ai-in-finance\\\/#primaryimage\",\"url\":\"https:\\\/\\\/www.meltone.com\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp\",\"contentUrl\":\"https:\\\/\\\/www.meltone.com\\\/wp-content\\\/uploads\\\/2026\\\/07\\\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp\",\"width\":1200,\"height\":630,\"caption\":\"Can Generative AI Be Trusted in Finance\"},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#website\",\"url\":\"https:\\\/\\\/www.meltone.com\\\/\",\"name\":\"Meltone\",\"description\":\"\",\"publisher\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/www.meltone.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#organization\",\"name\":\"Meltone\",\"url\":\"https:\\\/\\\/www.meltone.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/www.meltone.com\\\/wp-content\\\/uploads\\\/2026\\\/06\\\/logo-meltone-black.svg\",\"contentUrl\":\"https:\\\/\\\/www.meltone.com\\\/wp-content\\\/uploads\\\/2026\\\/06\\\/logo-meltone-black.svg\",\"width\":1,\"height\":1,\"caption\":\"Meltone\"},\"image\":{\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/www.linkedin.com\\\/company\\\/meltone-advisory?originalSubdomain=fr\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/www.meltone.com\\\/#\\\/schema\\\/person\\\/71719c78de779a1e8cd0a511ccc47963\",\"name\":\"l.deleuse\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/d11858f47a387b31eae0c35d5866a35ed223aafb204dc4577a456a44be028911?s=96&d=mm&r=g\",\"url\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/d11858f47a387b31eae0c35d5866a35ed223aafb204dc4577a456a44be028911?s=96&d=mm&r=g\",\"contentUrl\":\"https:\\\/\\\/secure.gravatar.com\\\/avatar\\\/d11858f47a387b31eae0c35d5866a35ed223aafb204dc4577a456a44be028911?s=96&d=mm&r=g\",\"caption\":\"l.deleuse\"},\"url\":\"https:\\\/\\\/www.meltone.com\\\/en\\\/author\\\/l-deleuse\\\/\"}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"The Impact of AI in Finance: How Are Your Architectures Evolving?","description":"Discover the impact of AI in Finance on ERP, EPM, and Data Platforms: AI agents, MCP, data governance, and the new Finance architecture.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/","og_locale":"en_US","og_type":"article","og_title":"The Impact of AI in Finance: How Are Your Architectures Evolving?","og_description":"Discover the impact of AI in Finance on ERP, EPM, and Data Platforms: AI agents, MCP, data governance, and the new Finance architecture.","og_url":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/","og_site_name":"MeltOne","article_published_time":"2026-07-17T07:29:14+00:00","article_modified_time":"2026-08-07T07:43:49+00:00","og_image":[{"width":1200,"height":630,"url":"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp","type":"image\/webp"}],"author":"l.deleuse","twitter_card":"summary_large_image","twitter_misc":{"Written by":"l.deleuse"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/#article","isPartOf":{"@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/"},"author":{"name":"l.deleuse","@id":"https:\/\/www.meltone.com\/#\/schema\/person\/71719c78de779a1e8cd0a511ccc47963"},"headline":"Can we trust generative AI in Finance?","datePublished":"2026-07-17T07:29:14+00:00","dateModified":"2026-08-07T07:43:49+00:00","mainEntityOfPage":{"@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/"},"wordCount":7,"commentCount":0,"publisher":{"@id":"https:\/\/www.meltone.com\/#organization"},"image":{"@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/#primaryimage"},"thumbnailUrl":"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp","keywords":["AI","ERP","Performance management","Smart ERP"],"articleSection":["Opinion"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/#respond"]}]},{"@type":"WebPage","@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/","url":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/","name":"The Impact of AI in Finance: How Are Your Architectures Evolving?","isPartOf":{"@id":"https:\/\/www.meltone.com\/#website"},"primaryImageOfPage":{"@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/#primaryimage"},"image":{"@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/#primaryimage"},"thumbnailUrl":"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp","datePublished":"2026-07-17T07:29:14+00:00","dateModified":"2026-08-07T07:43:49+00:00","description":"Discover the impact of AI in Finance on ERP, EPM, and Data Platforms: AI agents, MCP, data governance, and the new Finance architecture.","breadcrumb":{"@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.meltone.com\/en\/can-we-trust-generative-ai-in-finance\/#primaryimage","url":"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp","contentUrl":"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/07\/Peut-on-faire-confiance-a-lIA-generative-en-Finance-.webp","width":1200,"height":630,"caption":"Can Generative AI Be Trusted in Finance"},{"@type":"WebSite","@id":"https:\/\/www.meltone.com\/#website","url":"https:\/\/www.meltone.com\/","name":"Meltone","description":"","publisher":{"@id":"https:\/\/www.meltone.com\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/www.meltone.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/www.meltone.com\/#organization","name":"Meltone","url":"https:\/\/www.meltone.com\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/www.meltone.com\/#\/schema\/logo\/image\/","url":"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/06\/logo-meltone-black.svg","contentUrl":"https:\/\/www.meltone.com\/wp-content\/uploads\/2026\/06\/logo-meltone-black.svg","width":1,"height":1,"caption":"Meltone"},"image":{"@id":"https:\/\/www.meltone.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/www.linkedin.com\/company\/meltone-advisory?originalSubdomain=fr"]},{"@type":"Person","@id":"https:\/\/www.meltone.com\/#\/schema\/person\/71719c78de779a1e8cd0a511ccc47963","name":"l.deleuse","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/secure.gravatar.com\/avatar\/d11858f47a387b31eae0c35d5866a35ed223aafb204dc4577a456a44be028911?s=96&d=mm&r=g","url":"https:\/\/secure.gravatar.com\/avatar\/d11858f47a387b31eae0c35d5866a35ed223aafb204dc4577a456a44be028911?s=96&d=mm&r=g","contentUrl":"https:\/\/secure.gravatar.com\/avatar\/d11858f47a387b31eae0c35d5866a35ed223aafb204dc4577a456a44be028911?s=96&d=mm&r=g","caption":"l.deleuse"},"url":"https:\/\/www.meltone.com\/en\/author\/l-deleuse\/"}]}},"_links":{"self":[{"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/posts\/5031","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/comments?post=5031"}],"version-history":[{"count":5,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/posts\/5031\/revisions"}],"predecessor-version":[{"id":7148,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/posts\/5031\/revisions\/7148"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/media\/5032"}],"wp:attachment":[{"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/media?parent=5031"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/categories?post=5031"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/tags?post=5031"},{"taxonomy":"cas-client-offre","embeddable":true,"href":"https:\/\/www.meltone.com\/en\/wp-json\/wp\/v2\/cas-client-offre?post=5031"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}