[{"data":1,"prerenderedAt":554},["ShallowReactive",2],{"blog-en-internal-ai-workflows":3,"blog-related-en-internal-ai-workflows":138},{"id":4,"title":5,"author":6,"body":7,"coAuthors":122,"date":124,"description":125,"detail":126,"extension":127,"featured":126,"image":128,"imageSize":129,"meta":130,"navigation":131,"path":132,"railTitle":129,"seo":133,"stem":134,"summary":129,"tags":135,"unlisted":126,"updatedAt":129,"video":129,"videoWebm":126,"__hash__":137},"blog\u002Fblog\u002Fen\u002Finternal-ai-workflows.md","Our internal AI workflows","Pierre-Louis",{"type":8,"value":9,"toc":114},"minimark",[10,14,17,20,25,28,31,34,41,45,48,51,54,60,64,67,70,73,76,80,83,92,95,103,106],[11,12,13],"p",{},"Enterprise AI is usually sold as a new colleague: the autonomous employee that reads your codebase, files the tickets, ships the fix and updates the documentation while the founders sleep. It is a tidy story, and almost nobody runs their company on it. The agents that have earned a place in a working engineering team are far less cinematic. They take over a specific, repetitive chore, they do it quickly, and they hand the result back to a person who decides what happens next.",[11,15,16],{},"Last week Anthropic held an event at our Paris office, and our engineering team put three of these on the table. None of them runs the company. Each one removes a particular kind of toil that used to eat an engineer's afternoon, and in every case a human still signs off before anything reaches a customer. That second part is not a caveat we bolt on for comfort. It is the design. An agent that scales the work while a person keeps the judgment is a throughput multiplier you can actually trust. An agent that decides on its own is a liability waiting for its first bad day.",[11,18,19],{},"Here is what that looks like in practice.",[21,22,24],"h2",{"id":23},"the-sre-agent-compressing-the-slow-part-of-an-incident","The SRE agent: compressing the slow part of an incident",[11,26,27],{},"Site reliability work has a familiar shape. Monitoring flags an anomaly, and then the clock starts on the part that hurts: figuring out what broke. The fix is often quick once you know where to look. The looking is what costs you, usually at an inconvenient hour, usually under pressure, usually across a week's worth of changes any of which could be the culprit.",[11,29,30],{},"Our SRE agent goes after exactly that bottleneck. When monitoring flags an anomaly, the agent pulls the latest code changes, checks them against the symptom, and proposes a likely cause within minutes. It does not deploy anything. It does not roll anything back. It narrows the search and tells the on-call engineer where to start, and the engineer decides and fixes.",[11,32,33],{},"The value is not that the agent is clever. It is that diagnosis, the genuinely tedious detective work, no longer depends on one person.",[11,35,36],{},[37,38],"img",{"alt":39,"src":40},"Presenting the SRE agent","\u002Fblog\u002Fdogfood-sre-agent.jpg",[21,42,44],{"id":43},"the-help-center-agent-documentation-that-keeps-pace","The help-center agent: documentation that keeps pace",[11,46,47],{},"We ship every week. Documentation rots accordingly. Help articles are right on the day they are written, but the product moves and the articles do not, so a customer reads last quarter's screenshot of a screen that no longer exists.",[11,49,50],{},"Our help-center agent reads each new release, detects the features that are new or changed, and drafts or revises the matching article. When a screen has moved, it takes fresh screenshots of the relevant parts of the app so the visuals match what the customer is actually looking at. Then the product team reviews the changes before anything is published.",[11,52,53],{},"That review step is the whole point. The agent does the labour of keeping a sprawling help center current, which no human enjoys and few do consistently. A person still owns the words a customer reads. The result is documentation that tracks the product week by week instead of trailing it by a release or three, without a quarterly scramble to catch up.",[11,55,56],{},[37,57],{"alt":58,"src":59},"Presenting the help-center agent","\u002Fblog\u002Fdogfood-helpcenter-agent.jpg",[21,61,63],{"id":62},"the-worktree-switcher-space-for-several-agents-at-once","The worktree switcher: space for several agents at once",[11,65,66],{},"The third example is not an agent at all, and that is exactly why it matters.",[11,68,69],{},"A worktree switcher lets a developer keep several versions of a project open on the same machine at the same time, rather than being pinned to one. On its own that sounds like a minor convenience. It becomes important the moment you start using AI coding agents in earnest. You do not want to run one and wait; you want to run several in parallel, each in its own isolated workspace so they cannot overwrite each other's work, and switch between them without friction.",[11,71,72],{},"This is the unglamorous plumbing that makes parallel AI coding practical instead of chaotic. The constraint is no longer how many agents you can point at a problem, but how many a developer can supervise at once. The supervising, again, stays human.",[11,74,75],{},"A small theme runs through all three. The agent absorbs the part of the job that scales badly with human attention. The human keeps the part that should not be automated at all.",[21,77,79],{"id":78},"the-same-goes-for-the-agents-we-deploy-for-our-customers-in-the-data-room","The same goes for the agents we deploy for our customers in the data room",[11,81,82],{},"This is the principle we built our product on, and shipped to customers first.",[11,84,85,86,91],{},"Entropia was the first data room to launch an MCP server, in 2025, letting customers connect their own AI platform directly to their data room. Most of them use it with Claude. (More on ",[87,88,90],"a",{"href":89},"\u002Fblog\u002Fmcp-server-launch","the launch of our MCP server",".)",[11,93,94],{},"For the uninitiated, MCP, the Model Context Protocol, is an open standard that lets an AI assistant connect to an external system and act inside it through a defined set of permitted operations. The power of this MCP connection then depends on many factors, such as the sophistication of the AI platform connecting to it, but also the tools the data room makes available to it.",[11,96,97,98,102],{},"We draw the same line for customers: the AI can read, search, answer and scale across thousands of documents, and a person still validates the outcome and owns the trail of what was done. Our ",[87,99,101],{"href":100},"\u002Fblog\u002Fautonomous-irl-tracker","autonomous IRL tracker"," is a good example, checking every line of a request list against the data room and returning the file marked complete, partial or missing, ready for review. The discipline is identical on both sides of the wall.",[11,104,105],{},"Which leaves the question that every team adopting these tools eventually reaches. It was never really whether the agents can do more; they can, and they will. The interesting question is where you choose to keep the judgment. We have made our answer fairly explicit, internally and in the product. The agents handle the toil. The decisions stay with us.",[11,107,108,109,113],{},"See what the same discipline does on your deal. ",[87,110,112],{"href":111},"\u002Fcontact","Contact us",".",{"title":115,"searchDepth":116,"depth":116,"links":117},"",2,[118,119,120,121],{"id":23,"depth":116,"text":24},{"id":43,"depth":116,"text":44},{"id":62,"depth":116,"text":63},{"id":78,"depth":116,"text":79},[123],"Pierre Dulac","2026-06-15","How Entropia runs AI agents for reliability, documentation and code, with a human validating every output. Shown at Anthropic's Paris event.",false,"md","\u002Fblog\u002Fdogfood-hero-entropia-anthropic.png",null,{},true,"\u002Fblog\u002Fen\u002Finternal-ai-workflows",{"title":5,"description":125},"blog\u002Fen\u002Finternal-ai-workflows",[136],"insights","b5oWMyC0GxiF8rv4BVjB8zz4kGi15f5ZDF6X94_vgJo",[139,314,471],{"id":140,"title":141,"author":6,"body":142,"coAuthors":302,"date":305,"description":306,"detail":126,"extension":127,"featured":126,"image":307,"imageSize":129,"meta":308,"navigation":131,"path":309,"railTitle":129,"seo":310,"stem":311,"summary":129,"tags":312,"unlisted":126,"updatedAt":129,"video":129,"videoWebm":126,"__hash__":313},"blog\u002Fblog\u002Fen\u002Fma-tech-circle-paris.md","M&A Tech Circle Paris",{"type":8,"value":143,"toc":290},[144,147,150,154,160,163,166,185,188,192,198,201,204,209,212,216,219,223,226,230,233,237,240,243,247,253,256,282],[11,145,146],{},"On 3 June, at la maison.ai, we held the first edition of the M&A Tech Circle with Arx and Aesus Advisory. On the agenda: an ecosystem announcement, a concrete demonstration of deliverable automation, and cocktails to keep the conversation going.",[11,148,149],{},"The idea behind the M&A Tech Circle is to bring together M&A professionals who take AI seriously, to strengthen the ecosystem and explore how technology is transforming the profession through real cases and first-hand accounts. This first edition delivered on that promise, with two highlights.",[21,151,153],{"id":152},"entropia-arx-integration","Entropia × Arx integration",[11,155,156],{},[37,157],{"alt":158,"src":159},"Julien Belon and Pierre-Louis on stage for the integration announcement","\u002Fblog\u002Fmatc-arx-stage.jpg",[11,161,162],{},"Arx's CRM module, connected to the M&A ecosystem, lets you manage deal tracking from sourcing through closing. But visibility stops the moment due diligence opens in the data room. And that is exactly where the seriousness of acquirers shows: the depth of analysis, by firm and by individual. The technical integration between Arx and Entropia closes that gap.",[11,164,165],{},"The integration, announced by Julien Belon (Arx) and Pierre-Louis (Entropia), will let Arx clients:",[167,168,169,178],"ul",{},[170,171,172,173,177],"li",{},"launch a data room, or a ",[174,175,176],"strong",{},"complimentary pre-data room",", in one click from Arx, keeping their Arx identity and passing users and groups across, with no double entry;",[170,179,180,181,184],{},"feed ",[174,182,183],{},"counterparty analytics"," tracked in the data room back into the CRM.",[11,186,187],{},"This integration is brought to you by the only two sovereign solutions, each in its own specialty (European companies, hosted on European cloud — Scaleway \u002F OVH), not only to give you better visibility on your deals, but also to strengthen the strategic independence of European M&A.",[21,189,191],{"id":190},"aesus-advisory-automating-ma-deliverables-for-real","Aesus Advisory: automating M&A deliverables, for real",[11,193,194],{},[37,195],{"alt":196,"src":197},"The demo projected on screen","\u002Fblog\u002Fmatc-aesus-stage.jpg",[11,199,200],{},"Gabriel d'Agay, founding partner of Aesus Advisory, walked through a real, anonymised case: a buy-side deal, the consolidation-led build-up of an accounting firm. The stack comes down to two pieces: the Entropia data room (notably its built-in agentic workflows and its MCP server enabling collaboration with Claude), connected to a Claude project equipped with instructions, in-house skills (graphic charter, acquisition terms, DD templates drawn from past deals) and files.",[11,202,203],{},"Five deliverables were produced, from pre-LOI analysis to Due Diligence preparation.",[205,206,208],"h3",{"id":207},"pre-loi-report","Pre-LOI report",[11,210,211],{},"From the data room (call notes, three years of accounts, budget, legal org chart, payroll data): a complete report with history, org chart, revenue breakdown, an EBITDA bridge from reported to adjusted, points of attention, and a valuation range aligned with the firm's house guidelines. Plus a ready-to-use Q&A.",[205,213,215],{"id":214},"financial-model","Financial model",[11,217,218],{},"P&L, balance sheet and cash flow over three years, an adjusted-EBITDA tab, a bridge, and a full LBO model: free cash flows, debt coverage, leverage.",[205,220,222],{"id":221},"letter-of-intent-loi","Letter of intent (LOI)",[11,224,225],{},"A complete document: scope, valuation, governance, purchase undertaking, liquidity, financing, warranties, exclusivity, signatories, down to the personal wealth dimension. The first version was corrected by hand (formatting, a request for a proper bridge), a useful reminder that the human stays in the loop to arbitrate and validate.",[205,227,229],{"id":228},"irl-checklist","IRL checklist",[11,231,232],{},"More advanced: an Entropia agentic workflow checks the data room's actual contents against the information requests sent, and flags what is missing. A check that used to be done by hand.",[205,234,236],{"id":235},"due-diligence-preparation-report","Due Diligence preparation report",[11,238,239],{},"Finally, a red flag report and Q&A preparation, again from the data room.",[11,241,242],{},"Across these demos, Gabriel walked through the setup in detail: how he assembled each of the skills used, the time it took to produce each deliverable, the importance of review and verification, and finally the impact on Aesus's productivity and its ability to raise the level of service it offers, especially to its small-cap clients.",[21,244,246],{"id":245},"cocktail","Cocktail",[11,248,249],{},[37,250],{"alt":251,"src":252},"Conversations over cocktails","\u002Fblog\u002Fmatc-ambient.jpg",[11,254,255],{},"The conversation carried on over cocktails with the banks and advisory firms present, all the richer for a room mixing MDs, AI taskforce leads and associates. Make or buy, productivity gains, sovereignty, opportunities for juniors: there was no shortage of topics.",[167,257,258,264,270,276],{},[170,259,260,263],{},[174,261,262],{},"Auditability."," Where it still breaks, and how to guarantee no error or omission.",[170,265,266,269],{},[174,267,268],{},"Productivity and quality."," How far quality can rise at constant fees.",[170,271,272,275],{},[174,273,274],{},"Maintenance and replicability."," The cost of maintaining the system and transposing it to a larger structure.",[170,277,278,281],{},[174,279,280],{},"Staffing."," What becomes of junior roles and how they learn the trade.",[283,284],"matc-credits",{"cta-href":285,"cta-label":286,"cta-prefix":287,"entropia-href":288,"label":289},"\u002Fbook-a-demo","book a demo","To see Entropia on your case,","\u002F","Organized by",{"title":115,"searchDepth":116,"depth":116,"links":291},[292,293,301],{"id":152,"depth":116,"text":153},{"id":190,"depth":116,"text":191,"children":294},[295,297,298,299,300],{"id":207,"depth":296,"text":208},3,{"id":214,"depth":296,"text":215},{"id":221,"depth":296,"text":222},{"id":228,"depth":296,"text":229},{"id":235,"depth":296,"text":236},{"id":245,"depth":116,"text":246},[303,304],"Gabriel d'Agay","Julien Belon","2026-06-03","Entropia × Arx announcement, and Aesus Advisory demoing the production of a full chain of deliverables with the Entropia × Claude integration on a real deal.","\u002Fblog\u002Fmatc-hero.png",{},"\u002Fblog\u002Fen\u002Fma-tech-circle-paris",{"title":141,"description":306},"blog\u002Fen\u002Fma-tech-circle-paris",[136],"5baQ7SX8_H3HuaS4Cke6BF8UydZfCU-G9OzDorXv3h8",{"id":315,"title":316,"author":6,"body":317,"coAuthors":129,"date":461,"description":462,"detail":126,"extension":127,"featured":126,"image":463,"imageSize":129,"meta":464,"navigation":131,"path":465,"railTitle":129,"seo":466,"stem":467,"summary":129,"tags":468,"unlisted":126,"updatedAt":129,"video":129,"videoWebm":126,"__hash__":470},"blog\u002Fblog\u002Fen\u002Fcontext-adaptive-file-redaction.md","Context-adaptive document redaction",{"type":8,"value":318,"toc":459},[319,322,325,328,331,336,339,350,353,356,360,363,366,369,372,375,378,382,385,399,403,406,409,412,415,418,421,424,427,430,434,437,440,443,446,449,452],[11,320,321],{},"A sell-side analyst uploads 847 employment contracts to a data room. Each contains social security numbers, home addresses, salary details. The VDR promises to redact them all automatically. One click, the marketing materials said. Forty-five seconds later, a confirmation appears: \"1,247 redactions complete.\"",[11,323,324],{},"The analyst opens a random file. The CEO's social security number sits there, unmasked. Three more spot checks reveal similar gaps. The automated system missed variations in formatting, struggled with scanned documents, failed to recognize abbreviated names.",[11,326,327],{},"Eight hours of manual review follow. Better than risking a GDPR violation or tanking a deal because confidential client names leaked to a competitor.",[11,329,330],{},"\"I'd rather have my intern spend eight hours on redaction than five minutes with an AI solution and take the risk,\" says Victor, an M&A associate at a major European investment bank. The bank's reputation sits on the line with every document shared. No algorithm gets to make that call unsupervised.",[332,333,335],"h4",{"id":334},"what-redaction-actually-means-in-ma","What redaction actually means in M&A",[11,337,338],{},"Document redaction in mergers and acquisitions isn't about hiding embarrassing details. It serves three precise functions that carry legal and financial consequences when done poorly.",[167,340,341,344,347],{},[170,342,343],{},"First, regulatory compliance. European data protection rules require removing personally identifiable information before sharing employee records with third parties. Social security numbers, home addresses, phone numbers, bank details. Missing a single instance can trigger regulatory investigations. Financial penalties for GDPR violations reach into millions of euros, calculated as percentages of global revenue.",[170,345,346],{},"Second, competitive protection. A target company's client list, pricing structures, supplier contracts, and strategic partnerships constitute valuable intelligence. Buyers conducting due diligence need to verify commercial relationships exist without learning specific terms that could advantage them in negotiations or, worse, leak to competitors if a deal collapses.",[170,348,349],{},"Third, staged disclosure. M&A transactions progress through phases. Early-stage buyers see summary financials and anonymized contracts. Later, as negotiations advance, additional details unlock. The redaction system needs to support this gradual revelation, with precise control over what each party sees and when.",[11,351,352],{},"The consequences of failure are concrete. A leaked client list can trigger contract renegotiations. Exposed pricing data undermines competitive positioning. Personal information breaches generate regulatory scrutiny that delays or kills transactions entirely.",[11,354,355],{},"And then there's the technical requirement: true redaction must destroy the underlying data bytes, not merely overlay black boxes. Some transaction advisory firms openly discuss tools that can \"un-redact\" documents where the original text remains embedded in the file structure. If buyers or their advisors possess such tools, cosmetic redaction becomes disclosure.",[332,357,359],{"id":358},"the-ai-redaction-theater","The AI-redaction theater",[11,361,362],{},"Legacy VDR providers spotted an opportunity. Artificial intelligence was the solution to every enterprise software problem, apparently including document redaction. Marketing materials promised dramatic results: \"Redact 1,000 files in one click.\" \"80% time savings.\" \"AI-powered pattern recognition.\"",[11,364,365],{},"Datasite, one of the established players, promotes its automated redaction feature prominently. But the demo showcased on their website fails spectacularly. When it functions, the system might identify obvious patterns like social security numbers in structured documents, but complexity defeats it quickly. PDFs with unusual layouts confuse the parser. Scanned images remain completely opaque. Tables scramble its logic.",[11,367,368],{},"The system might correctly identify confidential information in some employment agreements while missing others where formatting differs slightly. No uncertainty scores appear. No confidence intervals. Just green checkmarks and buried liabilities.",[11,370,371],{},"What legacy vendors actually sell is autonomous AI. The system makes decisions about what to redact based on pattern matching and training data. It executes those decisions without meaningful human oversight. The human role reduces to clicking \"approve\" on bulk operations affecting hundreds or thousands of documents simultaneously.",[11,373,374],{},"This autonomy becomes the vulnerability. M&A teams don't need software that makes decisions. They need tools that preserve their ability to make decisions while eliminating mechanical friction.",[11,376,377],{},"The pricing model reveals the vendors' actual priorities. Legacy vendors charges up to an additional €2,500 per data room for AI-assisted redaction. And some count redacted pages as separate files, effectively charging twice for the same document. Teams upload a fifty-page contract, redact ten pages, and discover they're being billed for sixty pages of storage. This has side effects: users redact outside the VDR to avoid inflated charges, undermining the entire purpose of a centralized, auditable system.",[332,379,381],{"id":380},"what-users-actually-need","What users actually need",[11,383,384],{},"The conversation about AI-powered redaction typically focuses on speed and automation. That misses what M&A professionals actually want from their tools.",[167,386,387,390,393,396],{},[170,388,389],{},"Control, first. Deal teams need visibility into every redaction decision. Not trust, verification. The system can suggest, flag, and accelerate, but humans must validate before anything executes. This isn't inefficiency, it's risk management in an environment where a single mistake can cost millions.",[170,391,392],{},"Completeness, second. The fear isn't just making errors, it's missing items entirely. An analyst redacting Pierre-Louis Corteel's information needs the system to identify every variation: P.L. Corteel, Corteel PL, Pierre L. Corteel. Across all documents in the data room, not just the one currently open. Manual review might catch variations in the same file. It won't catch them across 847 employment contracts spread through different folders.",[170,394,395],{},"Security comes third, though in practice it underlies everything. True redaction must destroy data at the byte level. Visual overlays aren't sufficient. The underlying text must disappear from the file structure entirely, preventing any attempt at recovery.",[170,397,398],{},"And pricing transparency matters more than vendors acknowledge. Teams need predictable costs that don't penalize thoroughness. If redacting documents becomes expensive, the economic incentive shifts toward doing less redaction, which directly contradicts security objectives.",[332,400,402],{"id":401},"context-aware-redaction-an-innovative-approach","Context-aware redaction: an innovative approach",[11,404,405],{},"Entropia takes a different approach to redaction. Rather than attempting to make AI autonomous, the platform makes it context-aware.",[11,407,408],{},"The system observes what users do, learns patterns from specific environments, and scales individual decisions across bounded contexts. Critically, every suggestion requires human validation before execution.",[11,410,411],{},"The workflow begins with manual redaction, but amplifies it. An analyst redacts a social security number in an employment contract. That single action creates a bounded context the system replicates. The platform searches every document in the data room for that exact pattern and flags each instance. The analyst reviews the results, confirms matches to apply redaction.",[11,413,414],{},"One judgment call, executed fifty times. The human decided what to redact. The software handled pattern matching and execution.",[11,416,417],{},"This extends to more complex scenarios. A team uploads financial statements monthly throughout a deal. They redact specific salary bands and contract values in the first batch. When new statements arrive weeks later, the system identifies analogous fields based on position, formatting, and previous patterns. It flags them for review. The team confirms or adjusts, then applies.",[11,419,420],{},"The system also handles format variations. Redact \"Pierre-Louis Corteel\" once, and the platform flags \"P.L. Corteel,\" \"Corteel PL,\" and similar variations across all documents. Not automatically redacted, automatically flagged. The distinction matters. Users maintain control while the system prevents oversights that would occur during pure manual review.",[11,422,423],{},"Scanned documents or pdf are usually excluded from automations, but in Entropia, they are fully searchable, with the same redaction workflows applying regardless of how documents originated. This removes a major blind spot where legacy systems fail.",[11,425,426],{},"The search functionality integrates directly with redaction. Search for an item in a document, find ten instances, redact them all in one click. Then the system alerts you to other occurrences in different files. The workflow becomes: search, review results in current document, redact confirmed instances, then address flagged items elsewhere in the data room.",[11,428,429],{},"Pricing follows a different logic entirely. Assisted redaction isn't an add-on feature. Redacted documents don't lead to a per-page upcharge. It's included in the platform. No artificial incentives to redact outside the system. No penalties for thorough document protection.",[332,431,433],{"id":432},"the-pattern-recognition-problem","The pattern recognition problem",[11,435,436],{},"Perhaps the more interesting question is whether M&A teams will demand better tools, or simply adapt to elevated risks. Early evidence suggests they're rejecting automation theater in favor of control.",[11,438,439],{},"The vendors marketing \"one-click redaction for 1,000 files\" aren't addressing what M&A professionals actually need. They're selling the idea of productivity gains to budget holders who don't sit in the trenches reviewing documents. The people doing the actual work understand the difference between speed and thoroughness.",[11,441,442],{},"A context-aware system like Entropia's offer a middle path. The AI doesn't make decisions, it scales decisions humans already made. The human expertise remains central. The software eliminates repetitive mechanical tasks while preserving oversight at critical junctures. This matters because document volumes in M&A continue growing. A mid-sized transaction easily involves thousands of files. Pure manual review becomes a nightmare at scale. But autonomous systems that make unsupervised decisions about sensitive data aren't the answer either.",[11,444,445],{},"The solution involves AI, but not the AI that vendors have been marketing. Not systems that replace human judgment with algorithmic confidence. Instead, tools that extend human judgment across larger datasets while maintaining visibility and control at every step. Whether the VDR industry moves in this direction depends on whether buying decisions come from deal teams who understand the requirements, or from procurement departments optimizing for feature checklists and cost reduction.",[11,447,448],{},"The evidence from conversations with M&A professionals suggests the former is winning. When an associate says \"I'd rather have my intern spend eight hours on redaction than take the risk with AI,\" that's not technophobia. That's someone who understands exactly where liability sits when documents leak or regulators investigate.",[11,450,451],{},"Control, it turns out, means knowing when to automate and when to verify. The best redaction system isn't the one that promises to do everything automatically. It's the one that helps humans do their job better without removing them from the process.",[11,453,454,455,113],{},"-> Read more about our context-aware data room in ",[87,456,458],{"href":457},"\u002Fblog\u002Fdont-let-your-vdrs-ai-tank-your-deal","this article",{"title":115,"searchDepth":116,"depth":116,"links":460},[],"2025-12-29","Legacy VDRs charge extra for AI-redaction that doesn't work. Entropia has a better idea.","\u002Fblog\u002Fscreenshot-2025-12-29-at-18.50.06.png",{},"\u002Fblog\u002Fen\u002Fcontext-adaptive-file-redaction",{"title":316,"description":462},"blog\u002Fen\u002Fcontext-adaptive-file-redaction",[469,136],"solution","aMrZy3g2VQUHblD9pAcwgEOVOA7xMBQN9ovQP4RqIUM",{"id":472,"title":473,"author":6,"body":474,"coAuthors":129,"date":545,"description":546,"detail":126,"extension":127,"featured":126,"image":547,"imageSize":129,"meta":548,"navigation":131,"path":549,"railTitle":129,"seo":550,"stem":551,"summary":129,"tags":552,"unlisted":126,"updatedAt":129,"video":129,"videoWebm":126,"__hash__":553},"blog\u002Fblog\u002Fen\u002Fdata-sovereignty-what-european-customers-need-to-know.md","European sovereignty’s true test: ownership",{"type":8,"value":475,"toc":543},[476,479,483,486,489,493,496,499,502,506,509,512,516,519,522,525,528,531,534,540],[11,477,478],{},"A Munich law firm uploads due diligence files to a Frankfurt data center. A Paris investment bank stores M&A documents on servers in Amsterdam. Both assume their data remains European. Yet both remain exposed to American surveillance laws.",[332,480,482],{"id":481},"the-geography-trap","The geography trap",[11,484,485],{},"The assumption that hosting data within EU borders shields it from foreign access has become one of corporate Europe's most dangerous misconceptions. Under the US CLOUD Act of 2018 and FISA Section 702, American authorities can compel US-owned companies to surrender data regardless of its physical location. The laws follow ownership, not geography. Microsoft acknowledged this reality in testimony before the French Senate earlier this year, unable to guarantee that customer data stored in European data centers would never be transferred to US authorities.",[11,487,488],{},"This jurisdictional tension has turned critical for the European M&A market, where document security determines deal outcomes. The continent processed 9,800 M&A transactions in 2023, a 12 percent increase over the prior year according to LSEG Data & Analytics. Each involved thousands of confidential documents cycling through virtual data rooms. A single compliance breach carries penalties reaching 20 million euros or 4 percent of global revenue under GDPR. More damaging still is the reputational cost when sensitive transaction details surface through foreign surveillance channels.",[332,490,492],{"id":491},"sovereignty-washing","Sovereignty washing",[11,494,495],{},"The technical response from American hyperscalers has been predictable. Microsoft, Amazon, and Google now market \"sovereign cloud\" solutions featuring European data centers and local partnerships. Critics call this \"sovereignty washing.\" As Cristina Caffarra, a Brussels-based competition economist, explained to The Register: \"A company subject to the extraterritorial laws of the United States cannot be considered sovereign for Europe. That simply doesn't work.\" The parent company's American domicile ensures continued CLOUD Act jurisdiction, regardless of subsidiary structures or contractual promises.",[11,497,498],{},"Genuine sovereignty requires European ownership, not merely European hosting. French cloud provider Scaleway operates under this principle, with data centers in Paris, Amsterdam, and Warsaw controlled entirely by French parent company Iliad Group. No American parent company means no CLOUD Act exposure. The distinction matters operationally.",[11,500,501],{},"When Entropia, a virtual data room provider built by former Google engineers, evaluated infrastructure partners for its M&A platform, ownership structure determined the shortlist. The company's MCP server integration allows clients to access AI platforms like Claude and ChatGPT while maintaining document access controls, but the underlying infrastructure must remain immune to non-EU legal frameworks. Partnering with Scaleway solved the jurisdictional problem without compromising technical capabilities.",[332,503,505],{"id":504},"the-market-responds","The market responds",[11,507,508],{},"The shift is measurable. Germany's Schleswig-Holstein completed migration of 24,000 civil servants from Microsoft products to open-source alternatives in 2024. The International Criminal Court switched to European collaboration tools after chief prosecutor Karim Khan was temporarily locked from his Outlook account. France TV, GENCI, and the French National Center for Scientific Research signed partnerships with Scaleway rather than hyperscalers for infrastructure requiring full sovereign control. Corporate legal departments report similar patterns, with 67 percent experiencing deal delays due to cross-border compliance verification, according to the European Confederation of Directors' Associations.",[11,510,511],{},"The FISA Section 702 renewal in April 2024 sharpened these concerns. The law now covers \"any business with internet-linked infrastructure,\" expanding surveillance reach beyond traditional communications providers to encompass cloud services and data centers. The expansion came despite European Data Protection Board warnings that existing US surveillance laws already fell short of GDPR adequacy standards. Privacy advocates expect the EU-US Data Privacy Framework, which replaced the invalidated Privacy Shield agreement, to face judicial challenge within two years.",[332,513,515],{"id":514},"practical-implications-for-ma","Practical implications for M&A",[11,517,518],{},"For M&A practitioners, the implications are straightforward. Due diligence materials, financial projections, and transaction structures constitute precisely the sensitive commercial information that foreign intelligence agencies target for economic advantage.",[11,520,521],{},"The theoretical risk of CLOUD Act access during a transaction may be small, but the consequences of exposure are catastrophic. European alternatives eliminate the possibility entirely by removing the jurisdictional vulnerability at its source.",[11,523,524],{},"This does not mean European companies must abandon all American cloud services. Hybrid strategies work for many organizations, using global platforms for general computing while routing sensitive transactions through sovereign infrastructure. The key is understanding which workloads demand jurisdictional control. M&A documents, intellectual property repositories, and regulated financial data belong in infrastructure that answers exclusively to European law. Generic collaboration tools and development environments can tolerate broader exposure.",[11,526,527],{},"The virtual data room market reflects this calculus. European VDR revenue reached 860 million dollars in 2024 and is projected to quadruple by 2033, driven by regulatory compliance and cross-border transaction complexity. Providers emphasizing European ownership and ISO 27001 certification capture premium pricing from clients prioritizing sovereignty over convenience. The market dynamic suggests that compliance-conscious organizations increasingly view ownership as a security feature rather than a procurement detail.",[11,529,530],{},"Perhaps inevitably, the American hyperscalers will continue marketing sovereignty solutions. Their scale, innovation velocity, and ecosystem integrations remain formidable competitive advantages. Yet as long as CLOUD Act jurisdiction follows corporate parentage, these offerings cannot deliver true legal independence. The technical capabilities may be identical, but the legal architecture fundamentally differs. For European organizations handling sensitive M&A transactions, that distinction determines where the data lives.",[532,533],"hr",{},[535,536,537],"blockquote",{},[11,538,539],{},"CLOUD, in CLOUD Act, is capitalized because it means: \"Clarifying Lawful Overseas Use of Data\" Act.",[11,541,542],{},"Image credit: Scaleway",{"title":115,"searchDepth":116,"depth":116,"links":544},[],"2025-12-26","Extraterritorial laws override server location. True sovereignty requires European ownership, not only hosting.","\u002Fblog\u002Fscreenshot-2025-12-26-at-15.46.48.png",{},"\u002Fblog\u002Fen\u002Fdata-sovereignty-what-european-customers-need-to-know",{"title":473,"description":546},"blog\u002Fen\u002Fdata-sovereignty-what-european-customers-need-to-know",[136],"rZUpNTPK-fiNzReQZGCyprBZLSPWRLylUUHwxxtnJfE",1782805427864]