On AI assistant It belongs to the category of assistance systems: it is a software system that assists people in certain situations or when performing certain actions. The person activates the assistance system and can over-control it.
Functionality
An assistance system analyzes the current situation and, based on it, makes predictions if necessary. Machine learning methods make the system context-sensitive. This allows it to interact better with humans step by step. Advanced voice assistants record instructions using speech recognition, answer questions, and conduct dialogues. Such systems are often called intelligent personal assistants (IPAs).
Connection to data and tools
An AI model can be embedded in a software architecture that integrates external tools (tool usage). The model then uses, for example, real-time information, executes program code, or incorporates dynamic data environments. Retrieval Augmented Generation (RAG) follows a related approach: The language model retrieves information from a database or knowledge base without being trained anew. Therefore, for companies, the ability to integrate is crucial: How well can an AI model be integrated into existing IT landscapes, processes, and workflows?
AI assistants and SAP Business One
AI assistants like Claude or Copilot write emails, summarize documents, and explain technical terms. Without integration, they do not know the SAP Business One database of a company or the practical knowledge from its ERP daily operations. Versino offers two ways to achieve this: the SAP B1 MCP for accessing one’s own data and the Versino Skills for SAP B1 technical knowledge.
Demarcation
A chatbot is a virtual dialogue system that communicates via text input and output in natural language, for example in customer service. Agentic AI goes further: AI agents independently pursue predefined goals, plan in multiple stages, and act proactively. Moreover, assistance systems differ from automated and autonomous systems. The higher the level of automation, the less human supervision is required.
Related entries: Chatbot; Artificial Intelligence (AI); Large Language Model (LLM); AI agent for SAP B1; RAG (Retrieval-Augmented Generation) in ERP context; Skill (AI); SAP-B1-MCP (Versino); Versino Skills.
AI assistant meets SAP Business One: Two paths out of the Versino AI shop
Sprint and Swirl: How an AI agent divides work and learns from trials
LATS: Considering several paths simultaneously with AI — and the availability check in SAP B1
ReAct and RLEF: How Artificial Intelligence Learns from Real Mistakes — and What Automated Bank Reconciliation in SAP Business One Has to Do with It
SAP B1 10.0 FP2608: Service Layer AI as a transactional layer
