Vérifiez votre IA fournisseur, modèle, conformité, risques, sous-traitants, transparencefournisseurmodèleconformitérisquessous-traitantstransparencefournisseur

ProcBay is an AI-powered procure-to-pay (P2P) and strategic sourcing platform for enterprise procurement teams. It unifies intake and demand capture, strategic sourcing (RFx), supplier lifecycle management, supplier collaboration, spend insights and budget control, smart catalogue management, eAuctions, contract intelligence and an AI procurement orchestrator. Intended users include procurement, finance/AP, operations, legal/compliance and supplier stakeholders. Evidenced use cases: structured intake and policy-aware routing, AI-ranked supplier shortlisting and scoring, RFx questionnaire generation, bid analysis and comparison, supplier onboarding & KYC, clause recommendation, contract building, risk scoring, and predictive spend/approval-bottleneck analytics. The platform reports outcomes such as ~60% less manual work, 25-40% faster sourcing and 30-50% faster approvals, and integrates with ERP (SAP, Oracle, MS Dynamics 365, NetSuite), accounting and SSO systems.

Mithra is spend intelligence software that cleans, classifies and enriches procurement transaction data and surfaces savings opportunities such as tail spend, maverick buying, price variance and contract leakage. It is used by procurement and finance teams; Mithra shows SPAR, Compass, AGRA and Bertelsmann and case studies with Lamb Weston and Franke. Mithra says Data Foundation Agents classify and normalise transactions with human review, Opportunity Agents find savings, and the Atlas copilot answers spend questions, using customer-specific AI trained on the customer's data rather than shared models, with confidence scores and audit trails.

Coupa is a spend management platform for procurement, sourcing, supplier management, invoicing, payments and supply chain design. It is used by procurement and finance teams at large companies. Coupa says its Navi AI agents create sourcing events and compare bids, onboard suppliers with duplicate and sanctions checks, summarise supplier risk assessments, and (in Spend Pulse) scan for supplier risk, delayed shipments and price variance using Coupa community data, with humans approving actions; it has also absorbed Scoutbee's AI supplier discovery.

SAP Ariba is a source-to-pay suite covering procurement intake, guided buying, sourcing, contracts, supplier management and invoicing; the next-gen version runs on SAP Business Technology Platform. It is used by procurement teams at large enterprises. SAP says its Joule assistant is integrated into procurement workflows: a Bid Analysis Agent evaluates supplier bid scenarios including total cost, Joule summarises and extracts information from contracts and answers buyer questions, and an intake agent routes procurement requests.

EcoVadis IQ Plus maps and monitors sustainability and human-rights risk across a company's supplier base, helping buyers prioritise suppliers for assessment and meet due-diligence and reporting duties such as CSRD, CSDDD and the German Supply Chain Act. It is used by procurement and sustainability teams at buying companies, alongside EcoVadis Ratings. EcoVadis says IQ Plus uses AI sustainability data mining (DocScan, which scans about 30 million documents a month with transformer models) and live news monitoring of over 400,000 sources in eight languages with AI summaries and sentiment analysis, with insights validated by its analysts.

Satelligence monitors deforestation and land-use change in agricultural commodity supply chains (palm oil, soy, cocoa, coffee, rubber, cattle, wood) to support EUDR due diligence, deforestation- and conversion-free sourcing and Scope 3 reporting. It is used by commodity traders, food manufacturers and banks such as Cargill, Bunge, Mondelez, Unilever, Tony's Chocolonely and Rabobank. Satelligence says it runs specialised AI/machine-learning software in a high-performance cloud environment over optical and radar satellite data, LiDAR and over 200 million ground-truth points, and uses deforestation and planted-crop detection algorithms and machine-learning models for carbon estimates.