Error

Additional AI Terms - Third-Party AI

Download

Table of Contents


Additional AI Terms - Third-Party AI

When the Third-Party AI listed below is used in the Content Innovation Cloud, Additional AI Terms apply. The applicable AI Service Card identifies the Third-Party AI included in each Content Innovation Cloud product.

Amazon Bedrock

Some AI Models are provided via Amazon Bedrock. All such AI Models are subject to the applicable sections of the AWS Service Terms (the “AWS Model Terms”), which include the following terms (among others):

(i) These models may be used in connection with supporting healthcare services, provided that they are not medical devices under applicable law and regulations and are not intended to be used by themselves for any clinical diagnosis, decision-making, treatment, or other clinical use. Customer is responsible for all harm or liability that may arise in connection with any such uses;

(ii) Customer will not, and will not allow any third party to, modify, tamper with, remove, obscure, or otherwise alter Provenance Data. “Provenance Data” means information included as part of Output, such as metadata, digital signatures, or watermarks to identify it is generated using a generative AI model;

(iii) Customer’s use of these models must comply with applicable laws and regulations and AWS’s Acceptable Use Policy, and AWS’s Responsible AI Policy.

(iv) Amazon Bedrock may use automated abuse detection mechanisms designed to detect harmful content. If these mechanisms detect apparent child sexual abuse material, Customer agrees and instructs Amazon report the incident to the National Center for Missing and Exploited Children or other authority.

(v) To improve performance, such services may use cross-region inference, using the optimal AWS Region to process your content when running model inference.

1. Amazon Titan Text Embedding V2. Governed by the AWS Model Terms.

2. Amazon Rerank 1.0. Governed by the AWS Model Terms.

3. Amazon Nova Micro. Governed by the AWS Model Terms.

4. Amazon Transcribe - Batch. Governed by the AWS Model Terms.

5. Llama 3 8B: Governed the AWS Model Terms, the Meta Llama 3 Community License Agreement and the Meta Llama 3 Acceptable Use Policy. The Llama 3 model terms include the following restrictions (among other terms):

(i) Customer will not, and will not allow any third party to, use a model from the available AI Models, or Output therefrom, to improve any large language model (excluding any Meta Llama model).

(ii) If Customer uses the Output of a Llama Applicable Model to improve any AI model (excluding a Meta Llama model) that is distributed or made available, then Customer must (1) prominently display “Built with Llama” on a related website, user interface, blogpost, about page, or product documentation, and (2) include “Llama” at the beginning of any such AI model name.

(iii) If Customer distributes, licenses or sells the Output of a Llama Applicable Model or any AI model trained on such Output, then Customer must provide applicable copies of the Llama licenses.

(iv) Customer must not institute litigation or other proceedings against Meta alleging that the Output of a Llama Applicable Model or any portion thereof constitutes infringement of intellectual property or other rights owned or licensable by Customer.

6. Llama 3.2 11B Instruct. Governed by the AWS Model Terms, the Llama 3.2 Community License Agreement and the Llama 3.2 Acceptable Use Policy. The Llama 3.2 model terms include the restrictions listed in section 5(i) - 5(iv) above (among other terms).

7. Llama 4.0 Models. Governed By the AWS Model Terms, the Llama 4 Community License Agreement and the Llama 4 Acceptable Use Policy. The Llama 4 model terms include the restrictions listed in section 5(i) - 5(iv) above (among other terms), as well as the following: With respect to any multimodal models included in Llama 4, the rights granted under Section 1(a) of the Llama 4 Community License Agreement are not being granted to you if you are an individual domiciled in, or a company with a principal place of business in, the European Union. This restriction does not apply to end users of a product or service that incorporates any such multimodal models.

8. Anthropic Models. Governed By the AWS Model Terms and the Anthropic on Bedrock Terms, which includes the following restriction (among others terms): Customer may not and must not attempt to (i) access the services to build a competing product or service, including to train competing AI models except as expressly approved by Anthropic; (ii) reverse engineer or duplicate the services; or (iii) support any third party’s attempt at any of the conduct restricted in this sentence. Customer and its Users may only use the services in the countries and regions Anthropic currently supports.


The most current version of this document shall be such in effect as of 12:00am EST (Eastern Standard Time) of the date stamped on such online version.