All Activity

IntroductionThe European Union Artificial Intelligence Act (AI Act), Regulation (EU) 2024/1689, establishes a legal framework for the development, deployment, and use of artificial intelligence systems within the European Union. Its objective is to ensure that AI systems are developed and used in a manner that respects fundamental rights, safety, transparency, and accountability while supporting innovation. ClicShopping AI has been designed as an AI-assisted e-commerce platform that integrates external or locally hosted Large Language Models (LLMs) to assist merchants with content generation and administrative tasks. More as ClicShopping AI, this content is for all tools use the AI, below some recommandations to follow. More information : AI Eurepean Act AI-Assisted FeaturesClicShopping AI provides AI-assisted tools that support, but do not replace, human decision-making. Typical use cases include: Product description generation SEO content generation FAQ generation Product content enhancement Translation assistance Product creation assistance Administrative report generation Marketing content generation These features are initiated voluntarily by the user and operate as productivity tools. Human OversightOne of the principles promoted by the AI Act is appropriate human oversight. ClicShopping AI follows this principle by generating content intended for human review before publication or operational use. Users remain responsible for validating generated content, correcting inaccuracies, and deciding whether the generated output should be published or modified. ClicShopping AI does not make autonomous commercial or legal decisions on behalf of the user. TransparencyAI-assisted functionality is clearly identified within the application. Users intentionally invoke AI features through dedicated actions or interfaces. Generated content is presented as AI-generated assistance and should not be considered authoritative without verification. AI ProvidersClicShopping AI does not develop or train artificial intelligence models. Instead, it connects to AI providers selected by the system administrator. Supported providers include, without limitation: OpenAI Anthropic Google Gemini Mistral AI Ollama (local deployment) The administrator remains responsible for selecting the provider that best satisfies organizational, legal, privacy, and security requirements. Data ProcessingTo perform a requested AI operation, ClicShopping AI transmits only the information required by that operation. Depending on the selected feature, transmitted information may include: Product titles Product descriptions Product specifications Keywords Category names User prompts Context required for content generation The exact information transmitted depends on the selected feature and configured AI provider. Sensitive information should only be processed where appropriate safeguards have been implemented. High-Risk AI SystemsThe AI Act defines a category of High-Risk AI Systems for applications such as: Critical infrastructure Employment and recruitment Education Creditworthiness assessment Law enforcement Migration management Judicial decision support Access to essential public and private services The intended purpose of ClicShopping AI does not fall within these categories. ClicShopping AI is designed as an AI-assisted content generation platform intended to improve productivity within an e-commerce environment. If the software is integrated into workflows that qualify as high-risk under applicable legislation, additional legal obligations may apply to the deployer or integrator. Accuracy and LimitationsLike all generative AI systems, responses may contain: factual inaccuracies; incomplete information; outdated content; unsupported statements; hallucinations. Generated content should always be reviewed before publication or business use. ClicShopping AI does not guarantee the accuracy, completeness, originality, or legal compliance of AI-generated content. Privacy and Data ProtectionWhen external AI providers are used, submitted information is processed according to the policies of the selected provider. Administrators should ensure that the chosen provider complies with applicable data protection requirements, including the General Data Protection Regulation (GDPR), where applicable. Local AI deployments, such as Ollama, may be used when organizations require that data remain within their own infrastructure. Open Source ArchitectureClicShopping AI is an open source software project. The platform provides a modular architecture allowing administrators to choose their preferred AI provider without modifying the application's core architecture. The software acts as an integration layer between business workflows and AI services and does not itself constitute a foundation model or a general-purpose AI model. DocumentationThe following documents provide additional information inside the github repository: AI_USAGE.md — AI features and supported providers. AI_PROVIDERS.md — AI provider configuration. AI_LIMITATIONS.md — Known limitations of AI-assisted features. AI_ACT_COMPLIANCE.md — Compliance statement regarding the AI Act. DisclaimerThis article provides a general overview of how ClicShopping AI has been designed with regard to the principles established by the European AI Act. It is intended for informational purposes only and does not constitute legal advice. Organizations deploying ClicShopping AI remain responsible for ensuring that their implementation complies with the laws and regulations applicable in their jurisdiction.

The European AI Act (Regulation (EU) 2024/1689) introduces a legal framework for the development and deployment of Artificial Intelligence systems within the European Union. Although ClicShopping AI integrates AI capabilities, its primary role is to assist merchants with content creation and administrative tasks. It does not make autonomous decisions affecting individuals or perform activities classified as High-Risk AI Systems under the AI Act. To support the principles of the regulation, ClicShopping AI has been designed around several key concepts: Human oversight – AI generates suggestions that remain under the user's control and responsibility. Transparency – AI features are explicitly invoked by the user through dedicated interfaces. Provider independence – Administrators can choose their preferred AI provider, including OpenAI, Anthropic, Google Gemini, Mistral AI, Ollama, and other compatible services. Privacy awareness – Only the information required for a requested AI operation is transmitted to the configured provider. Open architecture – ClicShopping AI does not develop or train AI models; it provides a flexible integration layer between the application and AI providers. To improve transparency and documentation, the project now includes dedicated documentation covering: AI usage and supported providers AI limitations AI Act compliance These documents describe how AI features operate, what information may be exchanged with AI providers, known limitations of generative AI, and the measures implemented to align with the principles of the European AI Act. As with any AI-assisted system, generated content should always be reviewed before publication or business use. Note all these points will be integrated, in the next realease. More information : AI Eurepean Act

ClicShopping strengthens its AI ecosystem through MCP compatibility with LmStudio ClicShopping announces its compatibility with LmStudio through MCP (Model Context Protocol). This integration allows users to select a chat directly within LmStudio and initiate a conversation with ClicShopping without accessing the ClicShopping interface. The integration simplifies the use of local AI models while maintaining centralized control. New features can be created easily, with all governance, business rules, and permissions fully managed by ClicShopping. LmStudio functions solely as a conversational access point, without duplicating application logic. An example image below illustrates a response generated by LmStudio in this integrated context. What is MCP? MCP (Model Context Protocol) is a standardized protocol that enables external applications to interact with structured services or models while delegating business logic, contextual data, and governance to a dedicated system. It enforces a clear separation between conversational interfaces and application backends. Why use MCP with ClicShopping? Direct access to ClicShopping capabilities from local AI environments such as LmStudio Centralized governance, security, and business logic managed by ClicShopping Rapid feature development without modifying AI client applications Reduced integration complexity and improved maintainability Support for advanced use cases combining e-commerce and conversational AI This compatibility example demonstrates ClicShopping’s strategy to build an open, interoperable AI ecosystem while retaining full control over its platform and governance model.

We are proud now to announce you ClicShoppingAI is a native agentic e-commerce platform built on an open, multi-agent architecture orchestrated by a central Orchestrator Agent. Designed for extensibility, the platform enables the dynamic addition of new agents and functional domains as business needs evolve. Multi-Agent Architecture At the core of the system, the Orchestrator Agent analyzes user intent and routes requests to the appropriate domain agents. Specialized agents operate across key functions: Semantic Agent for semantic search and content understanding using vector embeddings and retrieval-augmented generation (RAG). Analytics Agent for internal data analysis, automated SQL generation, and business intelligence. WebSearch Agent for retrieving external information from the public web and competitive sources. Hybrid Agent combining semantic, analytics, and web search capabilities to handle complex queries. Conversational Interface ClicShoppingAI features a conversational chat interface that supports natural language queries and delivers contextualized responses through intelligent intent analysis, contextual awareness, response validation, and multilingual support. Monitoring and Performance A centralized dashboard provides real-time visibility into platform activity with key performance indicators, including agent performance metrics, system latency, cache usage, alerts, trends, and token consumption statistics. Extensible by Design The platform relies on standardized interfaces and a clear three-layer architecture (Domains, Apps, Agents), ensuring clean separation between query processing, business logic, and autonomous agent behavior. This design allows rapid integration of new agents and domains without disrupting existing functionality. An example below : An request example across the chat :