FAQ
Configure which AI models your workspace uses by enabling or disabling providers like Azure AI Foundry, Vertex AI, and Bedrock in admin settings
Data Protection
What role does awork play in processing customer data?
For data that customers process within their workspace, awork is generally a processor as defined in Article 28 of the GDPR. Customers remain the controllers responsible for any personal data they process in awork.
Where is our data stored?
The main data storage for the awork platform is hosted in Microsoft Azure in Germany.
For specific features - particularly optional AI, support, or communication functions - data may additionally be processed by subprocessors listed in our Data Processing Agreement.
Is data processed outside the EU?
The central storage of workspace data takes place in Germany.
Depending on which features you use, individual subprocessors may process data in other EU/EEA countries or, in limited cases, in third countries. The specific locations and safeguards are documented in our current Data Processing Agreement.
Where can I find the current list of subprocessors?
The complete and legally binding list is part of our current Data Processing Agreement.
AI
Which AI does awork use?
awork doesn't rely on just a single model or AI provider.
Depending on which model you select, processing can take place via Microsoft Azure AI Foundry, Google Vertex AI, Amazon Bedrock, or TensorX, among others.
Are customer data used to train AI models?
No. The AI services used by awork are configured and contractually bound in such a way that customer data is not used to train general AI models.
Where are AI requests processed?
That depends on which model you select.
The AI processing that awork regularly offers takes place in the documented European or EEA processing regions respectively. You can find details under "Sub-processors & Data Flows".
Is AI mandatory in awork?
No.
You can use the awork platform without AI data processing. Admins can disable AI features. The additional awork AI Add-on is purchased separately.
Can we disable specific AI models?
Yes. Workspace admins can disable the AI models available for regular use. Some technical or basic AI features may rely on fixed models that aren't part of the regular model selection.
Can AI access all the data in our workspace?
No.
AI features only receive the context needed for the specific request and that the respective user can access based on their existing awork permissions.
If you connect external systems or MCP servers, the access available there additionally depends on their configuration and permissions. For these systems, the data protection terms and processing locations of the respective third-party provider also apply.
Does awork AI make decisions on its own?
No.
awork AI supports users and can perform actions based on a user instruction or previously configured settings. It does not make independent decisions with legal or similarly significant consequences.
Is awork AI a high-risk AI system under the EU AI Act?
Based on the currently planned scope of use, awork's AI features do not constitute high-risk AI systems under the EU AI Act.
Regardless, we take into account the transparency requirements that apply to generative and interactive AI.
