Construct public-facing generative AI functions utilizing Amazon Q Enterprise for nameless customers

Amazon Q Enterprise is a generative AI-powered assistant that solutions query, supplies summaries, generates content material, and securely completes duties primarily based on enterprise information and knowledge. It connects to firm information sources, functions, and inside techniques to supply related, contextual solutions whereas sustaining organizational safety and compliance requirements.

Immediately, we’re excited to announce that Amazon Q Enterprise now helps nameless person entry. With this new characteristic, now you can create Amazon Q Enterprise functions with nameless person mode, the place person authentication will not be required and content material is publicly accessible. These nameless person functions can be utilized in use instances similar to public web site Q&A, documentation portals, and buyer self-service experiences.

This functionality permits visitor customers to make use of Amazon Q Enterprise generative AI capabilities to shortly discover product info, get technical solutions, navigate documentation, and troubleshoot points. Your public-facing web sites, documentation, and help portals can now ship the identical highly effective AI-driven help that authenticated customers obtain, creating an expertise that enriches the visitor person journey throughout your digital environments.

With this launch, you possibly can seamlessly combine an nameless Amazon Q Enterprise software into your web sites and internet functions by two pathways: both by embedding the ready-to-use internet expertise into your web sites utilizing an iframe for fast deployment, or by utilizing our Chat, ChatSync, and PutFeedback APIs to construct fully custom-made interfaces inside your personal functions. For nameless Amazon Q Enterprise functions, we’ve carried out a easy consumption-based pricing mannequin the place you’re charged primarily based on the variety of Chat or ChatSync API operations your nameless Amazon Q Enterprise functions make.

On this publish, we reveal the way to construct a public-facing generative AI software utilizing Amazon Q Enterprise for nameless customers.

Resolution overview

On this resolution, we stroll you thru creating an nameless Amazon Q Enterprise software utilizing each the AWS Administration Console and AWS Command Line Interface (AWS CLI). Our instance demonstrates a sensible situation: serving to web site guests discover info on public-facing documentation web sites.

We reveal the way to check the implementation with pattern queries by the built-in internet expertise URL. The ensuing software may be custom-made and embedded immediately into your web sites (utilizing the API or the iframe methodology), offering quick worth in your customers.

Stipulations

To comply with together with this publish, you will have the next:

• An AWS account.
• A minimum of one Amazon Q Enterprise Professional person that has admin permissions to arrange and configure Amazon Q Enterprise. For pricing info, see Amazon Q Enterprise pricing.
• AWS Identification and Entry Administration (IAM) permissions to create and handle IAM roles and insurance policies.
• Public content material to index (paperwork, FAQs, data base articles) that may be shared with unauthenticated customers.
• A supported information supply to attach, similar to an Amazon Easy Storage Service (Amazon S3) bucket containing your public paperwork.
• The AWS CLI configured with acceptable permissions (if following the AWS CLI methodology).

Create an nameless Amazon Q Enterprise software utilizing the console

On this part, we stroll by the steps to implement the answer utilizing the console.

Create an IAM function for the online expertise

Earlier than creating your Amazon Q Enterprise software, you will have to arrange an IAM function with the suitable permissions:

1. On the IAM console, select Roles within the navigation pane and select Create function.
2. Select AWS service because the trusted entity
3. Choose Amazon Q Enterprise from the service listing.
4. Select Subsequent: Permissions.
5. Create a customized coverage or connect the mandatory read-only insurance policies, and add permissions for nameless entry.

We strongly advocate that you just use a restricted coverage for the function, just like the one proven within the following screenshot, which can be used to create the online expertise for nameless entry software environments.

An instance of a restricted function coverage for calling the Chat API for nameless entry software environments could be arn:aws:qbusiness:::software/.

6. Create an IAM function with a belief coverage that permits the Amazon Q Enterprise service principal to imagine the function utilizing AWS Safety Token Service (AWS STS), particularly scoped to your software’s Amazon Useful resource Identify (ARN) within the designated AWS Area.
   

Create an Amazon Q Enterprise software

Now you’re able to create your Amazon Q Enterprise software:

1. On the Amazon Q Enterprise console, select Create software.
2. For Software identify, enter a reputation (for instance, SupportDocs-Assistant).
3. For Person entry, choose Nameless entry for this software surroundings.
4. Choose Net expertise to create a managed internet expertise to entry the Amazon Q Enterprise software.
   

You will notice a discover about consumption-based billing for nameless Amazon Q Enterprise functions. For extra particulars on pricing, consult with Amazon Q Enterprise pricing.

5. Depart the default service function possibility until you might have particular necessities.
6. For Encryption, use the default AWS managed key until you want customized encryption.
7. For Net expertise settings, you should utilize an present IAM function out of your account or authorize Amazon Q Enterprise to generate a brand new function with acceptable permissions. For this publish, we choose Use an present service function and select the IAM function created earlier (QBusinessAnonymousWebRole).
8. Optionally, customise the online expertise title and welcome message.
   
9. Evaluate all of your configuration choices and select Create to create the appliance.

You must see a affirmation that your nameless entry software has been created efficiently.

You will see the mandatory parameters and particulars of your Amazon Q Enterprise software on the touchdown web page displayed after profitable creation like the next screenshot, which supplies complete details about your newly created Amazon Q Enterprise software.

Add information sources

After you create your software, that you must add an index and information sources. To be taught extra, consult with Index. You will notice a pop-up like the next indicating that nameless entry is enabled.

Full the next steps:

1. Out of your software dashboard, select Add index.
2. Identify your index (for instance, Supportdocs-Exterior) and maintain the default settings.
3. Select Add an index.
   
4. After you create the index, you possibly can add information sources to it.

For our instance, we use the Amazon Q Enterprise public documentation as our information supply by including the URL https://docs.aws.amazon.com/amazonq/newest/qbusiness-ug/what-is.html. The Net Crawler will robotically index the content material from this documentation web page, making it searchable by your nameless Amazon Q Enterprise software.

For extra details about Net Crawler configuration choices and finest practices, consult with Connecting Net Crawler to Amazon Q Enterprise.

5. Out of your index dashboard, select Add information supply.
6. Enter a reputation in your information supply and elective description.
7. For Supply, choose Supply URLs and enter the URLs of the general public web sites you need to index.
   
8. For Authentication, choose No authentication.
   
9. Configure the sync run schedule and subject mappings.
10. Select Add information supply.
    

Alternatively, you possibly can add Amazon S3 as the info supply:

1. Out of your index dashboard, select Add information supply.
2. Choose Amazon S3 because the supply.
3. Configure your S3 bucket settings (make sure that the bucket has public entry).
4. Full the info supply creation course of.

It’s essential to solely ingest publicly obtainable information sources with out entry management lists (ACLs).

Generate an nameless internet expertise URL

After your information sources are arrange, full the next steps:

1. Out of your software dashboard, select your software.
2. Within the Net expertise settings part, select Share one-time URL.
   

The nameless internet expertise URL may be shared as a single-use hyperlink that should be redeemed and accessed inside 5 minutes. After it’s activated, the Amazon Q Enterprise session stays lively with a configurable timeout starting from 15–60 minutes. This lets you expertise the online interface and check its performance earlier than deploying or providing the nameless software to visitor customers.

Check your nameless Amazon Q Enterprise software

To check the appliance, select Preview internet expertise.

The next screenshot reveals the welcome web page in your nameless Amazon Q Enterprise software’s internet interface. Let’s start asking Amazon Q Enterprise some questions in regards to the Amazon Q index.

Within the first question, we ask “What’s Q index? How is it helpful for ISV’s?” The next screenshot reveals the response.

Within the following question, we ask “How can Q index enrich generative AI experiences for ISVs?”

In our subsequent question, we ask “How is Q index priced?”

Having efficiently examined our nameless Amazon Q Enterprise software by the console, we’ll now discover the way to create an equal software utilizing the AWS CLI.

Create your nameless software utilizing the AWS CLI

Guarantee that your AWS CLI is configured with permissions to create Amazon Q Enterprise assets and IAM roles.

Create an IAM function for Amazon Q Enterprise

First, create an IAM function that Amazon Q Enterprise can assume to entry essential assets:

# Create belief coverage doc
cat > trust-policy.json << 'EOF'
{
  "Model": "2012-10-17",
  "Assertion": [
    {
      "Effect": "Allow",
      "Principal": {
        "Service": "qbusiness.amazonaws.com"
      },
      "Action": "sts:AssumeRole"
    }
  ]
}
EOF

# Create IAM function
aws iam create-role 
  --role-name QBusinessAnonymousAppRole 
  --assume-role-policy-document file://trust-policy.json

# Connect essential permissions
aws iam attach-role-policy 
  --role-name QBusinessAnonymousAppRole

Create an nameless Amazon Q Enterprise software

Use the next code to create your software:

#bash
aws qbusiness create-application 
--display-name "PublicKnowledgeBase" 
--identity-type ANONYMOUS 
--role-arn "arn:aws:iam:: :function/QBusinessAnonymousAppRole" 
--description "That is the QBiz software for nameless use-case"

Save the applicationId from the response:

#json

{
  "applicationId": "your-application-id",
  "applicationArn": "arn:aws:qbusiness:area:account-id:software/your-application-id"
}

Create a restrictive coverage for nameless entry

We strongly advocate utilizing the next restricted coverage for the function that can be used to name the chat APIs for nameless entry software environments. This coverage limits actions to solely the mandatory APIs and restricts entry to solely your particular software.

Create the IAM function with the next coverage:

# Create restrictive coverage doc
cat > anonymous-access-policy.json << 'EOF'
{
  "Model": "2012-10-17",
  "Assertion": [
    {
      "Sid": "QBusinessConversationPermission",
      "Effect": "Allow",
      "Action": [
        "qbusiness:Chat",
        "qbusiness:ChatSync",
        "qbusiness:PutFeedback"
      ],
      "Useful resource": "arn:aws:qbusiness:::software/"
    }
  ]
}
EOF

# Connect the coverage to the function
aws iam put-role-policy 
  --role-name QBusinessAnonymousAppRole 
  --policy-name QBusinessAnonymousAccessPolicy 
  --policy-document file://anonymous-access-policy.json

Create an index

Create an index in your content material, then add paperwork utilizing the BatchPutDocument API. For step-by-step steering, see Choose Retriever.

Check your nameless Amazon Q Enterprise software

To reveal the chat performance utilizing the AWS CLI, we uploaded Amazon Q Enterprise documentation in PDF format to our index and examined the appliance utilizing the next pattern queries.

The next is an instance chat interplay utilizing the IAM function credentials. We first ask “What’s Amazon Q index?”

#1)
#bash
aws qbusiness chat-sync 
  --application-id  
  --user-message "What's Amazon Q index?"

The next screenshot reveals a part of the output from the chat-sync API when executed with our nameless Amazon Q Enterprise software ID, as proven within the earlier command.

Subsequent, we ask “How can Q index enrich generative AI experiences for ISV’s?”

2)
#bash
aws qbusiness chat-sync 
  --application-id  
  --user-message "How can Q index enrich generative AI experiences for ISV's?"

The next screenshot reveals a part of the output from the chat-sync API when executed with our nameless Amazon Q Enterprise software ID.

Create an online expertise for the nameless internet software

Use the next code to create the online expertise:

#bash
aws qbusiness create-web-experience 
  --application-id  
  --display-name "PublicKnowledgeBaseExperience" 
  --role-arn "arn:aws:iam:::function/QBusinessAnonymousAppRole" 
  --description "Net interface for my nameless Q Enterprise software"

To generate an nameless URL, use the next code:

#bash
aws qbusiness create-anonymous-web-experience-url 
  --application-id  
  --web-experience-id 

You should utilize the online expertise URL generated by the previous command and embed it into your internet functions utilizing an iframe.

Issues

Contemplate the next when utilizing nameless entry in Amazon Q Enterprise:

• The next are the one chat APIs that help nameless entry software environments:
  • Chat
  • ChatSync
  • PutFeedback
• You must solely ingest publicly obtainable information sources with out ACLs. Examples of public information sources embody:
  • Knowledge from the Amazon Q Enterprise Net Crawler
  • Amazon S3 information with out ACLs
• Amazon Q Enterprise functions with nameless entry are billed on a consumption-based pricing mannequin.
• Chat historical past will not be obtainable for nameless software environments.
• Nameless customers and authenticated customers should not supported on the identical software environments.
• Plugins should not supported for nameless software environments.
• Amazon QuickSight integration will not be supported for nameless software

Environments.

• Amazon Q Apps should not supported for nameless software environments.
• Attachments should not supported for nameless software environments.
• Admin controls and guardrails are read-only for nameless software environments, apart from blocked phrases.
• Matter guidelines utilizing customers and teams should not supported for nameless software

The remaining Amazon Q Enterprise performance and options stay unchanged.

Clear up

When you find yourself completed with the answer, clear up the assets you created.

Conclusion

On this publish, we launched Amazon Q Enterprise nameless person entry mode and demonstrated the way to create, configure, and check an nameless Amazon Q Enterprise software utilizing each the console and AWS CLI. This thrilling characteristic extends enterprise-grade Amazon Q Enterprise generative AI capabilities to your nameless audiences with out requiring authentication, opening up new potentialities for enhancing buyer experiences on public web sites, documentation portals, and self-service data bases. This characteristic is on the market by a consumption pricing mannequin that costs primarily based on precise Chat and Chatsync API utilization and index storage prices nonetheless relevant.

By following the implementation steps outlined on this publish, you possibly can shortly arrange an Amazon Q Enterprise software tailor-made in your exterior customers, secured with acceptable IAM insurance policies, and able to embed in your end-user-facing functions.

To be taught extra about this nameless entry characteristic, see the Amazon Q Enterprise Person Information. For detailed steering on embedding Amazon Q Enterprise in your internet functions, see Add a generative AI expertise to your web site or internet software with Amazon Q embedded. For those who’re excited by constructing fully customized UI experiences with the Amazon Q Enterprise API, take a look at Customizing an Amazon Q Enterprise internet expertise.

In regards to the authors

Vishnu Elangovan is a Worldwide Generative AI Resolution Architect with over seven years of expertise in Utilized AI/ML. He holds a grasp’s diploma in Knowledge Science and makes a speciality of constructing scalable synthetic intelligence options. He loves constructing and tinkering with scalable AI/ML options and considers himself a lifelong learner. Outdoors his skilled pursuits, he enjoys touring, collaborating in sports activities, and exploring new issues to unravel.

Jean-Pierre Dodel is a Principal Product Supervisor for Amazon Q Enterprise, chargeable for delivering key strategic product capabilities together with structured information help in Q Enterprise, RAG. and total product accuracy optimizations. He brings intensive AI/ML and Enterprise search expertise to the crew with over 7 years of product management at AWS.