Amazon’s cloud company AWS launched Amazon Q

Amazon's cloud company AWS launched Amazon Q

Amazon’s cloud company AWS has launched Amazon Q, a chat tool that allows businesses to ask questions specific to their company. Announced during AWS CEO Adam Selipsky’s keynote at AWS re:Invent, Amazon Q acts like an AI assistant, allowing users to ask questions about their business using their own data.

For example, an employee can ask Amazon Q for the latest guidelines on the use of the company logo or understand another engineer’s code for maintaining an application; Q can display the information instead of the employee having to look at dozens of documents.

Users can access Amazon Q from the AWS Management Console, individual company documentation pages, developer environments like Slack, and other third-party applications. Selipsky noted that the questions asked in Amazon Q “are not used for training on the underlying model.” Amazon Q can work with any model in Amazon Bedrock, the AWS repository of AI models, including Meta’s Llama 2 and Anthropic’s Claude 2.

According to the company, customers using Q typically choose the best model for their company, connect to that model’s Bedrock API, use it to learn their data, policies and workflows, and then deploy Amazon Q. According to AWS, Amazon Q is trained on 17 years of AWS knowledge and can be used to ask questions specific to AWS usage. It can also recommend the best AWS services for a project. Amazon Q is currently only available to users of Amazon Connect, AWS’ service for contact centers.

Eventually, it will be available for other services such as Amazon Supply Chain, which helps customers track their supply chain management, and Amazon QuickSight, a business intelligence platform. Amazon Q for Supply Chain and Business Intelligence in preview. Dilip Kumar, Vice President of Applications at AWS, told The Verge in an interview that each instance of Amazon Q in the AWS service will be different.

In Amazon Connect, Q will be deployed in real time, and in Amazon QuickSight, Q will be deployed in real time, Q provides contact center agents with relevant answers to their questions. “AI is a natural fit for contact centers, supply chain and business intelligence,” Kumar said. Amazon Q in Connect’s price starts at $40 per agent per month, and users can try Amazon Q in Connect “free until March 1, 2024,” according to the AWS Connect website.

According to Selipsky, Amazon Q recognizes security parameters set by the customer, so that employees without access to certain information cannot use the query system to retrieve data they are not authorized to view. Other companies have created similar products. Microsoft’s Copilot offers similar functionality for Windows users, while Dropbox’s Dash can query stored documents. Selipsky said that with the launch of Amazon Q, AWS will offer Bedrock users the ability to put guardrails around the models they use to build AI-powered applications.

These guardrails, currently in preview, will allow companies to ensure that their applications and the models they use to power them comply with data privacy and responsible AI standards. Companies, especially those operating in highly regulated sectors such as finance and healthcare, often say they cannot fence off their data and ensure that the information is not being used to train future models. AWS also has the ability to redact personally identifiable information of customer end-users, which is said to be part of the guardrails, but is not yet immediately available.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back To Top