Responsible use of nextAI
nextAI is ISPnext's AI platform. It supports users in processing and analysing information and generating proposals, insights and summaries, among other things.
This article explains what you can expect from nextAI, which general limitations apply and which checks are required to use the functionality responsibly.
Support from AI
The output of nextAI is intended to support users. Despite the controls and validation applied, AI-generated results may be incomplete, inaccurate or incorrect.
The quality of the output may be affected by, among other things:
- the quality and completeness of the information provided;
- deviating documents, structures or languages;
- missing context;
- unclear or contradictory source information;
- technical or functional limitations of the AI technology used.
nextAI therefore does not replace the user's substantive assessment and responsibility.
Human oversight
Always check AI output before it is processed further or used for decision-making.
AI output must not be used without appropriate human oversight as the sole basis for a final financial, contractual, legal or other binding decision.
Users must have sufficient knowledge, information and authority to assess the output. Where necessary, the user must be able to correct, reject or disregard the output.
Responsibilities
ISPnext is responsible for the design, integration, security, validation, monitoring and support of nextAI.
As a customer, you are responsible for:
- providing and using documents and data lawfully;
- correctly configuring and managing user rights;
- using nextAI within the agreed and supported purpose;
- checking, correcting or rejecting AI output;
- assessing whether the output is suitable for use within your own business process;
- preventing AI output from leading to a binding decision without appropriate human oversight;
- reporting errors, unexpected results or possible incidents.
The user is responsible for using nextAI within the authorised powers and for assessing the output before it is processed further.
Data protection and security
nextAI is integrated into the ISPnext platform and uses Microsoft Azure AI services within the Azure environment managed by ISPnext.
Customer data is processed in accordance with the applicable contractual, privacy and security agreements. Customer data is not used to train publicly available or external AI models.
ISPnext's existing security measures, including access management, logging, monitoring and incident management, also apply to nextAI.
Availability and changes
The available nextAI functionality may vary by ISPnext module, configuration and contractual agreement.
ISPnext may modify the functionality to improve its quality, security, reliability or user-friendliness. Material changes that affect customers' use or responsibilities will be announced through the designated communication channels.
Reporting errors, questions or concerns
Report errors or unexpected results to ISPnext Support when:
- information is processed incorrectly or incompletely;
- AI output does not correspond to the available source information;
- unexpected, inappropriate or potentially unwanted output is generated;
- you suspect that data has been processed incorrectly;
- you have questions or concerns about the use of nextAI.
Where possible, include the following in your report:
- the ISPnext module used;
- the functionality concerned;
- when the problem occurred;
- a description of the expected and actual result;
- relevant screenshots or other information that can be used to investigate the problem.
Do not share additional sensitive data in a Support report unless this is necessary.
ISPnext assesses reports in accordance with the regular Support and incident management process. Relevant reports are used for investigation, improvement, monitoring and, where necessary, reassessment of the risks and measures relating to nextAI.