Deepgram policy evolution
Before/after stance changes across captured policy versions, with exact citations. If no before/after delta is available yet, AIRIN shows the latest citation-backed stance events instead.
No before/after stance delta is available for this filter yet. Latest citation-backed stance events are shown below.
Aug 7, 2026traininglow
content ownership
Latest stance: user retains rights
“This opt-out applies to the individual request in which it is set, to give you more flexibility over how Your Content is used. For Saga users, you can access opt-out choices in your settings. Once you opt out, any new content you submit won't be used to train our models. 3.4 Ownership of Your Content. We do not claim ownership of Your Content. As between you and us, you retain any right, title, and interest you possess in Your Content. Due to the nature of artificial intelligence and our Services, an Output may not be unique to you, other users may create similar outputs using our Services, and we make no representations that you own or have rights in any similar output created by the Services for other customers. For clarity, Output does not include any portion of any machine learning models, algorithms, neural networks, and associated weights, parameters, and configurations (“ Models ”) that may be used to provide the Services. 3.5 Responsibility for Your Content. Our Services are designed to help you build and create, and we have some rules to make sure that our Services are used responsibly. You are solely responsible for all use of the Output, the Actions, and your Application, and for evaluating the accuracy and appropriateness of Output for your use case. You represent and warrant that: You have all rights, licenses and permissions necessary to grant us the license to Your Content in Section 3.2.”
Open citationAug 7, 2026traininghigh
content license
Latest stance: sublicensable or transferable
“3.1 Intro. You may provide content, data, and information, including audio recordings or a transcript, as an input to our Services (“ Input ”). The Services may generate and return audio, text, or other output based on the Input (“ Output ”). Input and Output are collectively, “ Your Content . ” You may also enable the Services, including our Voice Agent API, to take actions on your behalf (“ Actions ”), such as data processing and system interactions based on the Input. 3.2 License. We use Your Content to maintain and provide our Services to you. We may also use Your Content to improve our Services, to develop other products and services, and for other business purposes, including training and testing our Models. You hereby grant us an irrevocable, perpetual, transferable, sublicensable (through multiple tiers), fully paid, royalty-free, nonexclusive and worldwide right and license to use, reproduce, store, and modify Your Content for these purposes, including following any expiration or termination of these Terms. 3.3 Training Opt-Out. With more data, our models get sharper at understanding how people actually talk — accents, regional dialects, slang, natural pauses, and rhythm. If you'd rather we not use Your Content to improve our Models, you can do the following: For our APIs, you may opt out of model training on a per-request basis by setting the applicable parameter in your request as described here .”
Open citationAug 7, 2026trainingmedium
content license
Latest stance: broad license
“3.1 Intro. You may provide content, data, and information, including audio recordings or a transcript, as an input to our Services (“ Input ”). The Services may generate and return audio, text, or other output based on the Input (“ Output ”). Input and Output are collectively, “ Your Content . ” You may also enable the Services, including our Voice Agent API, to take actions on your behalf (“ Actions ”), such as data processing and system interactions based on the Input. 3.2 License. We use Your Content to maintain and provide our Services to you. We may also use Your Content to improve our Services, to develop other products and services, and for other business purposes, including training and testing our Models. You hereby grant us an irrevocable, perpetual, transferable, sublicensable (through multiple tiers), fully paid, royalty-free, nonexclusive and worldwide right and license to use, reproduce, store, and modify Your Content for these purposes, including following any expiration or termination of these Terms. 3.3 Training Opt-Out. With more data, our models get sharper at understanding how people actually talk — accents, regional dialects, slang, natural pauses, and rhythm. If you'd rather we not use Your Content to improve our Models, you can do the following: For our APIs, you may opt out of model training on a per-request basis by setting the applicable parameter in your request as described here .”
Open citationAug 7, 2026traininglow
model training
Latest stance: no training claim
“3.1 Intro. You may provide content, data, and information, including audio recordings or a transcript, as an input to our Services (“ Input ”). The Services may generate and return audio, text, or other output based on the Input (“ Output ”). Input and Output are collectively, “ Your Content . ” You may also enable the Services, including our Voice Agent API, to take actions on your behalf (“ Actions ”), such as data processing and system interactions based on the Input. 3.2 License. We use Your Content to maintain and provide our Services to you. We may also use Your Content to improve our Services, to develop other products and services, and for other business purposes, including training and testing our Models. You hereby grant us an irrevocable, perpetual, transferable, sublicensable (through multiple tiers), fully paid, royalty-free, nonexclusive and worldwide right and license to use, reproduce, store, and modify Your Content for these purposes, including following any expiration or termination of these Terms. 3.3 Training Opt-Out. With more data, our models get sharper at understanding how people actually talk — accents, regional dialects, slang, natural pauses, and rhythm. If you'd rather we not use Your Content to improve our Models, you can do the following: For our APIs, you may opt out of model training on a per-request basis by setting the applicable parameter in your request as described here .”
Open citationAug 7, 2026trainingmedium
model training
Latest stance: training with opt out
“3.1 Intro. You may provide content, data, and information, including audio recordings or a transcript, as an input to our Services (“ Input ”). The Services may generate and return audio, text, or other output based on the Input (“ Output ”). Input and Output are collectively, “ Your Content . ” You may also enable the Services, including our Voice Agent API, to take actions on your behalf (“ Actions ”), such as data processing and system interactions based on the Input. 3.2 License. We use Your Content to maintain and provide our Services to you. We may also use Your Content to improve our Services, to develop other products and services, and for other business purposes, including training and testing our Models. You hereby grant us an irrevocable, perpetual, transferable, sublicensable (through multiple tiers), fully paid, royalty-free, nonexclusive and worldwide right and license to use, reproduce, store, and modify Your Content for these purposes, including following any expiration or termination of these Terms. 3.3 Training Opt-Out. With more data, our models get sharper at understanding how people actually talk — accents, regional dialects, slang, natural pauses, and rhythm. If you'd rather we not use Your Content to improve our Models, you can do the following: For our APIs, you may opt out of model training on a per-request basis by setting the applicable parameter in your request as described here .”
Open citationAug 7, 2026trainingmedium
model training
Latest stance: training with opt out
“This opt-out applies to the individual request in which it is set, to give you more flexibility over how Your Content is used. For Saga users, you can access opt-out choices in your settings. Once you opt out, any new content you submit won't be used to train our models. 3.4 Ownership of Your Content. We do not claim ownership of Your Content. As between you and us, you retain any right, title, and interest you possess in Your Content. Due to the nature of artificial intelligence and our Services, an Output may not be unique to you, other users may create similar outputs using our Services, and we make no representations that you own or have rights in any similar output created by the Services for other customers. For clarity, Output does not include any portion of any machine learning models, algorithms, neural networks, and associated weights, parameters, and configurations (“ Models ”) that may be used to provide the Services. 3.5 Responsibility for Your Content. Our Services are designed to help you build and create, and we have some rules to make sure that our Services are used responsibly. You are solely responsible for all use of the Output, the Actions, and your Application, and for evaluating the accuracy and appropriateness of Output for your use case. You represent and warrant that: You have all rights, licenses and permissions necessary to grant us the license to Your Content in Section 3.2.”
Open citationGenerated from live stance events. Informational only, not legal advice.