I wanted to ask more information about the security within Vapi. Whilst, I do understand the security measures I can take like Hippa, deployment on our own cloud infrastructure and much more. My concern is more rooted in the API in which it works aside. The LLM, TTS, VTT, ETC. Do these API's store information for their training? If so, how could I avoid this?

@Brian At VAPI, we take security and privacy very seriously, and we understand your concerns regarding the handling of data by the APIs we integrate with, such as Language Models (LLM), Text-to-Speech (TTS), Voice-to-Text (VTT), etc. Our APIs are designed with privacy in mind. We do not store any information for training purposes unless explicitly permitted by the user for the purpose of improving the service. Moreover, we provide options for users to opt-out of data storage for training, ensuring that their data remains private and is not used for any other purposes. If you wish to avoid your data being used for training, you can configure your API usage settings to disable data retention. This way, you can ensure that the data processed by our APIs is not stored or used for training models. Additionally, we comply with industry standards and regulations, such as HIPAA, to ensure that sensitive information is handled appropriately. If you need specific instructions on how to configure these settings or have any further questions about our privacy practices, please let us know, and we'll be happy to assist you further.

@Vapi, I am not sure if my question was answered. I will explain in greater detail. If I use Chat GPT API or Anthropic API for the LLM, whilst VAPI may be secure we still are sending a prompt, and are recieving a response from that LLM API. I know Anthropic specifically in their Privacy Policy are allowed to collect data for training purposes and I don't believe there is a way to turn this off. That's the security part i'm concerned about, and if the other llms, and, and VTT have this as well? Also does Vapi have it's own LLM?

@Brian I understand your concern better now. When using third-party APIs like Chat GPT or Anthropic for LLM, the data handling and privacy practices of those services come into play. Indeed, some providers may collect data for training and improvement purposes, and their ability to opt out of such collection can vary. For each third-party API, it's essential to review their privacy policies and data handling practices. If a provider does not offer an opt-out for data collection, and this is a concern for your use case, it may be necessary to consider alternative providers that align more closely with your privacy requirements. Regarding other services like TTS and VTT, the same principle applies. Each service provider will have its own policies, and it's crucial to ensure that they meet your privacy standards before integration. As for VAPI having its own LLM, we focus on providing a secure and privacy-conscious platform for voice developers, which includes integrating with various third-party APIs. We do not have our own proprietary LLM at this time. Our priority is to ensure that when you use these integrated services through VAPI, your data is handled securely and in accordance with the highest privacy standards. If you have specific requirements or need assistance in configuring your services to maximize privacy, please let us know, and we can provide more detailed support or guidance.

@Vapi Can I get in contact with a real person?

@Brian Please ask support for more info on this.

@Vapi I want to talk to a real person, and not a bot.

@Brian Please ask support for more info on this.

Hey @Brian we don't store any of your data to train any models, for more you can check about it from here: https://security.vapi.ai