In this article, you will find the answers to the most common questions about AI usage in Tactiq.
What do you use for the AI generation of summaries, action items, meeting kits, and other features?
We use a GPT family or models provided by Microsoft Azure OpenAI Services (https://azure.microsoft.com, https://openai.com).
Does Tactiq use ChatGPT?
No. We are using Microsoft Azure OpenAI's enterprise API offering, which is separate from OpenAI and different from ChatGPT.
Is my data used for LLM training?
No. Azure OpenAI will not use data submitted via their API to train or improve their models. This differs from the consumer-oriented offering ChatGPT provided by OpenAI.com, which is not used by Tactiq.
Does your GPT provider store my data?
No. We have a service modification with Microsoft that disables content logging, so the data is not retained even for abuse and misuse monitoring purposes.
Is there an Azure OpenAI policy that I can read?
Yes - you can find it at https://learn.microsoft.com/en-us/legal/cognitive-services/openai/data-privacy.
Can I disable AI features completely?
Yes, you can disable Large Language Models (AI features) on the Settings page at https://app.tactiq.io/#/settings/general.
Can I disable AI features for all users in my team?
Yes, you can disable AI features for all users in your team. If you would like to disable AI features for all users in the organisation, this is possible in the scope of enterprise agreements. Please email [email protected] to configure this setting for your enterprise plan.
How is my data protected in transit to the LLM?
Our connection is secured by TLS (Transport Layer Security) and is not visible to other parties.
Can I delete the source and generated data?
Yes. You can archive a meeting in Tactiq and either wait for it to be deleted automatically in 30 days or purge it from the archive manually.
Tactiq's browser extension has "GPT" in its name. Does it mean that the extension is using GPT directly?
No. The only place GPT is used is at https://app.tactiq.io when you request AI generation for individual transcripts. It is also controlled by the setting mentioned above.
My security team or system administrators have more questions - what should they do?
Please ask them to reach out to [email protected]. We routinely go through security reviews and audits and are happy to help.