{"id":71082,"date":"2026-09-23T14:03:00","date_gmt":"2026-09-23T07:03:00","guid":{"rendered":"https:\/\/thaipropertynews.com\/feeds\/?p=71082"},"modified":"2026-09-23T14:03:00","modified_gmt":"2026-09-23T07:03:00","slug":"wenge-ai-subsidiary-d2-intelligence-launches-documind-an-enterprise-document-intelligence-and-decision-platform","status":"publish","type":"post","link":"https:\/\/thaipropertynews.com\/feeds\/?p=71082","title":{"rendered":"Wenge AI Subsidiary D2 Intelligence Launches DocuMind, an Enterprise Document Intelligence and Decision Platform"},"content":{"rendered":"

HONG KONG<\/span>, Sept. 23, 2026<\/span> \/PRNewswire\/ —\u00a0D2 Intelligence, a subsidiary of Wenge AI (1956.HK), has launched DocuMind, an enterprise document intelligence and decision platform. Designed for financial services, insurance and public services \u2014 where accuracy, compliance and traceability are critical \u2014 DocuMind brings document recognition, information extraction, cross-document verification and risk review into a single workflow. The platform enables institutions to make verifiable and auditable business decisions supported by a comprehensive evidence trail.<\/p>\n

Enterprises handle large volumes of unstructured documents, including bills of lading, invoices, contracts and identity documents. Critical information is often scattered across formats, languages and business systems. Traditional OCR extracts text but misses relationships and business logic. Conventional compliance systems rely on manually coded rules, costly to maintain, while LLMs may produce inconsistent outputs with unclear sources. Consistent, audit\u2011ready document review remains a key barrier to enterprise AI adoption.<\/p>\n

\n

\"Documind<\/a>
Documind Review Workspace<\/span><\/p>\n<\/div>\n

Built on Wenge AI’s DOMA architecture \u2014 Data, Ontology, Models and Agents \u2014 DocuMind integrates data governance, ontology modelling, model adaptation and agent orchestration. The platform uses OCR and vision\u2011language models to convert multilingual, multimodal documents into structured data. It applies domain ontologies, knowledge graphs and business rules to interpret entity relationships, constraints and transaction logic across documents. LLMs and multi\u2011agent systems use knowledge bases and external data to assess document completeness, credibility, transaction logic and related risks.<\/p>\n

Every DocuMind conclusion is traceable for compliance and accuracy. Extracted fields link to their source locations, and each conclusion keeps its rationale, evidence and audit trail. Institutions can configure fields, rules, thresholds and workflows to fit internal policies. Low\u2011risk cases run automatically, medium\u2011risk cases go to reviewers, and high\u2011risk cases are escalated. A side\u2011by\u2011side interface lets reviewers compare documents with structured results, locate evidence and complete approvals efficiently.<\/p>\n

DocuMind supports trade finance, credit documentation, KYC\/KYB and compliance review. In trade finance, it extracts parties, amounts, ports, routes and goods descriptions from invoices and bills of lading. It cross\u2011checks data against sanctions lists, dual\u2011use databases and institutional rules to flag inconsistencies and compliance risks. In the HKMA GenA.I. Sandbox, the system processed a trade finance document of up to 10 pages in under 30 seconds, compared with 10\u201320 minutes manually. Extraction accuracy for core fields exceeded 95%, and the review automation rate reached approximately 85%. DocuMind also supports data masking, isolated environments and on\u2011premises deployment for institutional privacy and security.<\/p>\n

DocuMind supports insurance underwriting and claims by classifying and linking identity documents, medical records, reports, invoices and prescriptions. It detects blurred, duplicated, recaptured or tampered images and reconstructs timelines for fragmented submissions, helping reviewers understand cases, identify anomalies and complete reviews.<\/p>\n

DocuMind supports public\u2011sector application reviews, approvals, onboarding, due diligence and compliance evidence management. The platform organises fragmented documents into unified, verifiable review records, reducing duplicate entry and manual comparison and providing a clear basis for approvals, risk reviews and outcomes.<\/p>\n

A Wenge AI spokesperson said enterprise AI must move beyond content generation. Real value comes from working within business processes using rules, traceable evidence and human oversight. DocuMind applies Wenge AI’s full\u2011stack capabilities to deliver flexibility, controls and traceability for secure, responsible AI adoption.<\/p>\n

About D2 Intelligence<\/b><\/p>\n

D2 Intelligence is a data and decision intelligence company powered by AI LLMs. We provide products and solutions for digital transformation, cross\u2011modal analytics, AI\u2011assisted decision optimization, AIGC content creation, financial risk management and investment research. We serve global brands, financial institutions and media companies including Haier, BYD, Kantar, China Construction Bank, China Merchants Bank, Guotai Junan Securities and CMG Asia-Pacific. Our headquarters is in Hong Kong, with R&D teams across major cities in Mainland China.<\/p>\n

For more information, visit https:\/\/www.d2-ai.com\/<\/a>.<\/p>\n

Website: https:\/\/www.wenge.com<\/a><\/span><\/p>\n

\u00a0<\/p>\n

<\/div>","protected":false},"excerpt":{"rendered":"

<\/p>\n

HONG KONG<\/span>, Sept. 23, 2026<\/span> \/PRNewswire\/ —\u00a0D2 Intelligence, a subsidiary of Wenge AI (1956.HK), has launched DocuMind, an enterprise document intelligence and decision platform. Designed for financial services, insurance and public services \u2014 where accuracy, compliance and traceability are critical \u2014 DocuMind brings document recognition, information extraction, cross-document verification and risk review into a single workflow. The platform enables institutions to make verifiable and auditable business decisions supported by a comprehensive evidence trail.<\/p>\n

Enterprises handle large volumes of unstructured documents, including bills of lading, invoices, contracts and identity documents. Critical information is often scattered across formats, languages and business systems. Traditional OCR extracts text but misses relationships and business logic. Conventional compliance systems rely on manually coded rules, costly to maintain, while LLMs may produce inconsistent outputs with unclear sources. Consistent, audit\u2011ready document review remains a key barrier to enterprise AI adoption.<\/p>\n

\n

\"Documind<\/a>
Documind Review Workspace<\/span><\/p>\n<\/div>\n

Built on Wenge AI’s DOMA architecture \u2014 Data, Ontology, Models and Agents \u2014 DocuMind integrates data governance, ontology modelling, model adaptation and agent orchestration. The platform uses OCR and vision\u2011language models to convert multilingual, multimodal documents into structured data. It applies domain ontologies, knowledge graphs and business rules to interpret entity relationships, constraints and transaction logic across documents. LLMs and multi\u2011agent systems use knowledge bases and external data to assess document completeness, credibility, transaction logic and related risks.<\/p>\n

Every DocuMind conclusion is traceable for compliance and accuracy. Extracted fields link to their source locations, and each conclusion keeps its rationale, evidence and audit trail. Institutions can configure fields, rules, thresholds and workflows to fit internal policies. Low\u2011risk cases run automatically, medium\u2011risk cases go to reviewers, and high\u2011risk cases are escalated. A side\u2011by\u2011side interface lets reviewers compare documents with structured results, locate evidence and complete approvals efficiently.<\/p>\n

DocuMind supports trade finance, credit documentation, KYC\/KYB and compliance review. In trade finance, it extracts parties, amounts, ports, routes and goods descriptions from invoices and bills of lading. It cross\u2011checks data against sanctions lists, dual\u2011use databases and institutional rules to flag inconsistencies and compliance risks. In the HKMA GenA.I. Sandbox, the system processed a trade finance document of up to 10 pages in under 30 seconds, compared with 10\u201320 minutes manually. Extraction accuracy for core fields exceeded 95%, and the review automation rate reached approximately 85%. DocuMind also supports data masking, isolated environments and on\u2011premises deployment for institutional privacy and security.<\/p>\n

DocuMind supports insurance underwriting and claims by classifying and linking identity documents, medical records, reports, invoices and prescriptions. It detects blurred, duplicated, recaptured or tampered images and reconstructs timelines for fragmented submissions, helping reviewers understand cases, identify anomalies and complete reviews.<\/p>\n

DocuMind supports public\u2011sector application reviews, approvals, onboarding, due diligence and compliance evidence management. The platform organises fragmented documents into unified, verifiable review records, reducing duplicate entry and manual comparison and providing a clear basis for approvals, risk reviews and outcomes.<\/p>\n

A Wenge AI spokesperson said enterprise AI must move beyond content generation. Real value comes from working within business processes using rules, traceable evidence and human oversight. DocuMind applies Wenge AI’s full\u2011stack capabilities to deliver flexibility, controls and traceability for secure, responsible AI adoption.<\/p>\n

About D2 Intelligence<\/b><\/p>\n

D2 Intelligence is a data and decision intelligence company powered by AI LLMs. We provide products and solutions for digital transformation, cross\u2011modal analytics, AI\u2011assisted decision optimization, AIGC content creation, financial risk management and investment research. We serve global brands, financial institutions and media companies including Haier, BYD, Kantar, China Construction Bank, China Merchants Bank, Guotai Junan Securities and CMG Asia-Pacific. Our headquarters is in Hong Kong, with R&D teams across major cities in Mainland China.<\/p>\n

For more information, visit https:\/\/www.d2-ai.com\/<\/a>.<\/p>\n

Website: https:\/\/www.wenge.com<\/a><\/span><\/p>\n

\u00a0<\/p>\n

<\/div>\n

<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"rop_custom_images_group":[],"rop_custom_messages_group":[],"rop_publish_now":"initial","rop_publish_now_accounts":[],"rop_publish_now_history":[],"rop_publish_now_status":"pending","footnotes":""},"categories":[5,7],"tags":[],"class_list":["post-71082","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-cision-pr-newswire","category-cision-pr-newswire-en"],"_links":{"self":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/posts\/71082","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=71082"}],"version-history":[{"count":0,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=\/wp\/v2\/posts\/71082\/revisions"}],"wp:attachment":[{"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=71082"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=71082"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/thaipropertynews.com\/feeds\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=71082"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}