Policy on the use of AI in the Lucca solutions

This document is intended for clients and future clients. The purpose of this document is to transparently describe the terms and conditions for the use of AI in the Lucca solutions, so that clients can deactivate any features they do not want to use.

1. Terms and conditions for the use of AI in the Lucca solutions

To help you better understand this document, you'll find a glossary at the end of the document.

Lucca undertakes to comply with the AI Act and the GDPR.

The solutions proposed directly by Lucca use Azure Open IA artificial intelligence services, localized in the following regions

  • Azure Switzerland region for Lucca clients hosted in Switzerland
  • Azure Europe region for all other Lucca clients

Azure Open AI artificial intelligence services are hosted by Microsoft ISO 27001 certified data centers that provide strong guarantees regarding the non-reuse of data shared.

The Lucca solutions use market AI models. Lucca does not generate models (including fine-tuning) with its clients' data.

The uses of AI within applications take the form of prompts (this is the name given to an instruction given to an AI), for certain features, embedding data from the client instance. 

Except for common uses of AI (defined as "minimal risk" in the AI ​​Act), clients can deactivate disable AI-driven features in the Lucca solutions.

Example of a deactivation interface: 

PENG_Settings_AI_analysis_partial_EN.pngThe AI summary feature is deactivated in the settings for the Engagement solution

2. "Minimal risk" AI features that cannot be deactivated in the Lucca solutions

Application Feature
Engagement Translating labels
Compensation Automatically suggest column header and remuneration type matches for remuneration imports
Recruitment Reading CVs (without sorting or categorizing)

3. AI-driven features that can be deactivated in the Lucca solutions

Application Feature
Engagement Summarizing comments
Expenses and Invoices Automatically reading photos and documents. (via our partner Mindee)

4. Planned AI-driven features 

Application Feature
Engagement Helping to create a survey
Engagement Suggesting action plans
Performance Detecting psychosocial risks
Performance Summarizing a review
Performance Summarizing a peer review
Performance Generating a skills framework
Performance Analyzing a difference between employee and manager
Core HR Providing an initial AI-assisted response on HR Ticketing

5. Glossary

LLM (Large Language Model): An artificial intelligence that can understand and generate text in natural language, like an assistant that answers questions or helps draft documents.

Prompt: An instruction or question given to AI to obtain a specific answer. The clearer the prompt, the more accurate the response will be.

AI Act: A European regulation regulating the use of artificial intelligence, particularly to guarantee transparency and security within companies while promoting innovation.

Page content

Was this article helpful?
0 out of 0 found this helpful