Built with DocStarAI Transparency
How viaSocket Uses AI
We believe you deserve to know exactly how artificial intelligence is used in our platform — which providers we work with, what data they can access, and what we never do with your information.
AI Providers We Use
viaSocket integrates with the following third-party AI service providers to power AI-assisted features within the platform:
OpenAI (ChatGPT) Powers natural language understanding for workflow creation, AI-assisted automation suggestions, and smart descriptions. | Anthropic (Claude) Used to assist users with workflow logic, step configuration, and generating automation instructions from plain language. |
viaSocket does not train, host, or operate its own AI model.
Our Data Commitments
We take your data privacy seriously. Here is exactly what we do and do not do with your data in relation to AI:
We do not share your data
Your personal data, workflow configurations, and connected app credentials are never shared with OpenAI, Anthropic, or any other AI provider beyond what is strictly required to process your specific request in that moment.
Your data is not used to train AI models
Your data is never used to train AI models — neither by viaSocket nor by any third-party AI provider we work with. We explicitly opt out of training data sharing wherever this option is available under the provider's API terms.
Data is used only to serve you
Any data sent to an AI provider is used solely to generate a response to your request in that session. It is not retained, repurposed, or reused by the AI provider beyond that interaction.
What Data Is Sent to AI Providers
When you use AI-powered features in viaSocket, only the minimum context necessary to fulfil your request is transmitted to an AI provider. This typically includes:
Natural language descriptions of the workflow you want to create or modify
Step names or labels where needed to assist with logic generation
Anonymous structural context about the automation being built
What is never included in AI prompts: - Your authentication credentials or API keys - Personal records or sensitive data passing through your workflows - Webhook payloads or raw integration data - Account details or billing information |
Frequently Asked Questions
Does viaSocket use its own AI model?
viaSocket uses its own AI-powered engine to help users build and manage automation workflows. To enhance this capability, we also integrate with third-party AI APIs — specifically OpenAI and Anthropic — which assist with natural language understanding and workflow suggestions.
Can I use viaSocket without AI features?
Yes. AI-powered assistance is an optional enhancement. All core automation and workflow features work independently without any AI involvement. You are never required to use AI features.
Is my data subject to OpenAI's or Anthropic's privacy policies?
When AI features are used, a minimal portion of interaction data passes through the respective AI provider's API. Both OpenAI and Anthropic operate under their own privacy policies and data processing agreements, which prohibit using API data for model training by default.
Will viaSocket add more AI providers in the future?
We may integrate additional AI services as our platform evolves. This page will always reflect the current state of AI usage, and we will notify users of any significant changes to how AI is used.
How can I opt out of AI features?
AI features can be avoided by simply not using AI-assisted tools within the platform. If you have specific concerns or require a data processing agreement, please contact our support team.
Questions about AI usage?
Reach us at https://viasocket.com/support — we are happy to clarify anything on this page or provide a formal data processing agreement if required by your organization.
