AWS releases generative AI employee assistant

Matthew Burbidge
By Matthew Burbidge
Johannesburg, 29 Nov 2023
Adam Selipsky, CEO of Amazon Web Services, opens AWS re:Invent with the launch of the Amazon Q generative AI tool.
Adam Selipsky, CEO of Amazon Web Services, opens AWS re:Invent with the launch of the Amazon Q generative AI tool.

Amazon Web Services (AWS) has launched Amazon Q, a new type of generative artificial intelligence (AI) assistant that it says is specifically developed for work and can be tailored to a customer’s business.

Speaking at his company’s annual re:Invent conference in Las Vegas this week, AWS CEO Adam Selipsky kicked off a programme filled with discussion around generative AI (GenAI).

Selipsky said while there had been great strides in providing useful GenAI applications for consumers, AWS noticed this usefulness did not extend to employees at their places of work.

“While their [GenAI tools’] general knowledge and capabilities are great, they don’t know your company, data, customers, or operations, and this limits how useful their suggestions can be. They also don’t know much about who you are at work; they don’t know your role, or preferences, what information you use, and what you do, and don’t, have access to.”

Another impediment to adopting GenAI tools within companies is they typically do not come with data privacy or security capabilities, and as a result, said Selipsky, many CIOs and CISOs banned the use of popular chat assistants at the office.

The Amazon Q tool can provide relevant answers to specific questions and generate content by mining a company’s information libraries and enterprise systems.

It’s envisaged that Amazon Q will be used by developers to write code on AWS, and it has now been embedded in the AWS management console and the integrated development environment, through the Code Whisperer coding app and in the team chat function of Slack.

Selipsky provided an example of how it can be used: “You could ask, ‘how do I build a web application with AWS? What are my options?’ It’s going to answer with a list of potential services you could use, such as [app development service] Amplify, [serverless compute service] Lambda, or [elastic compute service] EC2, and then it’s going to offer reasons why you might consider each service.

“From there, you can further narrow down your options through natural language. You could ask: ‘Which of these would be preferred if my app only needs to run for a few seconds, and only has very infrequent traffic?’

Once a service has been selected, said Selipsky, a developer can ask Q how they could get started.

If running into a problem with Lambda, for example, the developer can press the “troubleshoot with Q” button, and it will research the error and suggest how to fix it. It can also troubleshoot network connections, and analyse its configuration and suggest fixes for connectivity issues, he said.

Selipsky added Q will be able to shave hours of work off projects, “with just one good prompt”.

He noted developers will often find themselves “slogging through the muck” that came with migrations, maintenance and upgrades, and will run tests on upgraded applications.

AWS has been using Q internally. “With Amazon Q Transformation, a very small team of Amazon developers successfully upgraded a thousand applications from Java 8 to Java 17 in two days, which is usually how long it will take to upgrade a single application. It’s months, if not years, of development time saved.

“You will not currently find an AI assistant from another major cloud provider today,” according to Selipsky.

The feature is generally available from today.