Amazon are excited to introduce to you today Amazon Q, a new generative AI-powered personal assistant that can be customized for your company and is meant for work. By connecting to the code, data, enterprise systems, and information repositories of your organization, Amazon Q enables you to engage in dialogues, solve issues, produce content, acquire insights, and take action. Amazon Q gives staff members quick access to pertinent information and guidance to help them complete activities more quickly, make decisions and solve problems more quickly, and encourage creativity and innovation at work.
With Amazon Q’s user-based plans, you may customize the product’s features, cost, and options to suit your needs. Based on your company’s current identities, positions, and permissions, Amazon Q can customize its interactions for every single user. Amazon Q user content is never used by AWS to train the underlying models. Put another way, your company’s information is kept secret and safe.
Your go-to source for business advice is Amazon Q
Let’s examine a few instances of how business customers can employ straightforward natural language suggestions from Amazon Q to accomplish tasks. As a marketing manager, you could ask Amazon Q to summarize a press release, use the release as the basis for an email draft. For example, Amazon Q looks through internal style requirements while searching your company’s content to deliver a response that adheres to your brand standards.
Subsequently, you may request that Amazon Q create customized social media prompts to share your tale on all of your accounts. You can later ask Amazon Q to evaluate your campaign’s outcomes and compile a summary for leadership evaluations.
You can restrict irrelevant topics, set the context for responses, and decide whether to supplement responses with information from the underlying model or solely use reliable company information when responding as an administrator. Hallucinations, a prevalent problem where the underlying model provides responses that sound convincing but are based on misread or nonexistent data, are mitigated by limiting responses to trusted company information.
How to customize Amazon Q for your company
Go to Amazon Q in the console, pick Applications from the left menu, and then click Create application to customize Amazon Q for your company.
The workflow that follows is now underway.
Step 1: Make an application. Give your application name, and select or create an AWS Identity and Access Management (IAM) service role that already exists that Amazon Q is permitted to take on. Select Create after that.
Step 2: Choose the retriever. During a chat, a retriever retrieves data in real time from the index. You have two options: use a current Amazon Kendra retriever or use the native Amazon Q retriever. The data sources that are supported by Amazon Q can be accessed by the native retriever. You can link the related data sources to your Amazon Q application by selecting the current Amazon Kendra retriever if you currently use Amazon Kendra. you choose to use the native retriever. Select Next after that.
Step 3: Assemble data sources. Connectors for widely used data sources and enterprise systems are pre-installed on Amazon Q.
Once the retriever displays the correct number of documents and the data source sync has finished successfully, you can test the online experience and initiate a chat. Keep in mind that depending on the volume and size of data to index, the data source sync may take a few minutes to several hours.
Plugins that control access to enterprise systems, such as Salesforce, Jira, Zendesk, and ServiceNow, can also be connected. Using plugins, Amazon Q can carry out user-requested operations like generating support tickets and estimating sales.