During the AWS re:Invent conference in Las Vegas, Amazon unveiled its latest offering, an AI-powered chatbot called Q, designed specifically for AWS customers.
Available for as low as $20 per user per year and currently in public preview, Q has been trained on 17 years’ worth of AWS knowledge, making it capable of answering complex questions such as “How do I build a web application using AWS?” One of the standout features of Q is its ability to provide a list of potential solutions along with detailed explanations to help users make informed decisions.
AWS CEO Adam Selipsky highlighted Q’s versatility during his keynote speech, emphasizing that users can easily chat, generate content, and take actions using the chatbot.
Q’s understanding of users’ systems, data repositories, and operations informs its responses and recommendations.
To configure Q, AWS customers can connect it to organization-specific apps and software, including popular platforms like Salesforce, Jira, Zendesk, Gmail, and Amazon S3 storage instances.
By indexing and learning from connected data and content, Q gains insights into a business’s organizational structures, core concepts, and product names.
Q goes beyond mere question-answering, offering additional capabilities such as content generation and summarization of materials like blog posts, press releases, and emails.
Moreover, it can execute actions on behalf of users through configurable plugins.
These actions include creating service tickets automatically, notifying specific teams through Slack, and updating dashboards in ServiceNow. To ensure accuracy and avoid errors, Q allows users to inspect and validate actions before they are executed.
Q’s integration with CodeWhisperer, Amazon’s service for generating and interpreting app code, enables the chatbot to create tests, generate plans, and provide documentation for implementing new features or transforming code.
Selipsky shared an example of Q being used internally at Amazon to upgrade approximately 1,000 apps from Java 8 to Java 17, including testing, in just two days.
The chatbot’s expertise extends to AWS’s first-party products, such as AWS Supply Chain and QuickSight. Within QuickSight, Q enhances business reports by providing visualization options and answering questions about the data contained within.
In AWS Supply Chain, Q can analyze and answer queries related to shipment delays, offering up-to-date analyses.
Selipsky emphasized that Q prioritizes privacy and security. Administrators can control the information accessed by Q, ensuring that only authorized data is returned to users.
To address concerns about generative AI systems inventing facts, Q’s models are trained on company documents rather than customer data, preventing unauthorized use of sensitive information.
With Q, Amazon aims to provide a comprehensive solution for cloud customers, covering a wide range of use cases from business intelligence to programming and configuration.
Industry analyst Ray Wang considers Q to be the most significant announcement at re:Invent, highlighting its potential to empower developers with AI and drive successful implementations.
As Q’s effectiveness unfolds, it promises to revolutionize the way AWS customers interact with the platform, opening up new possibilities for efficiency, problem-solving, and decision-making.
