AI Meets AEM

In recent months, there has been an awakening in the general public about some of the practical uses of Artificial Intelligence (AI) in our lives. In fact, you probably heard about Chat GPT from your friends, your kids or a podcast. You may have even used it yourself.

Before we share some of the ideas we have come up with, we do want to mention that there is a price to using AI in the ways we mention in this article. The availability of a free version of Chat GPT may have some people thinking that using other AI products won’t cost a thing. Before installing or connecting any tool, always check the cost. Knowing your estimated usage will help you price it out.

What is Chat GPT?

Chat GPT is an Open AI chat tool. Quite simply, it is just a bot that lets a user talk to it or ask it questions while it responds. People everywhere can chat with a machine and get responses containing information that runs the gamut from insightful to fabricated. To be fair to the bots though, the information they output is only as good as the information they receive…from humans. And anyone who has spent any time on the internet knows that the information we share is not always correct!

Chat GPT should not be confused with what we are going to talk about for the rest of this blog or even its relevance to usage of AI in content management systems or AEM. We’re interested in what’s powering it.

From a technical perspective, the proliferation of more accessible AI tools results in endless potential and my colleagues and us have already come up with some exciting ways to answer the question, “How can we use AI in AEM?”

Doesn't AEM have AI?

Adobe recently introduced Generative AI which builds upon AI work they have been doing for a decade. This will enable content creators to benefit from products they add to applications including Firefly which is in beta now and Sensei which you may already be familiar with.

The intelligence layer Adobe Sensei can tag images in AEM Assets for example, perform basic edits in Adobe Creative Cloud and help deliver personalized experiences to customers. These all save time and effort while adding great value to businesses. And, there is more that AI can do but these are just examples of what has already existed in AEM.

How can you use AI with a CMS?

In addition to AI products that are native to a CMS, there are countless ways to integrate AI with your CMS and many of them have yet to be discovered. Some ideas are listed below.

Text

Use case examples: translation, content generation, Tweet generation, Chatbot responses, speech to text

Common Services: Open AI, Amazon Lex (chatbot), Azure Cognitive Services Translator (translation), Copywriting Generator by Simplified Content (content creation), Whisper by Open AI (speech to text)

Images

Use case examples: Image generation, creation of metadata including alt-text and tags, image delivery (personalization)

Common Services: EveryPixel (image recognition), Midjourney (image generation), Dall-E by Open AI (image generation)

Search

Use case examples: Auto-suggest, results prioritization, results relevancy (personalization), understanding of natural language, aggregation of data from resources

Common Services: yext, Azure Cognitive Search, Amazon Kendra

We thought about this question in terms of workflows and tasks that we ask AEM to perform and developed three features and are coming up with more ideas constantly. We’ll be featuring three of these ideas in upcoming blogs and an Evolve session so look out for those soon.

• Generation of Summary and Keywords for Text
• Generation of a Tweet
• Generation of Alt-text for an image

3|SHARE would love to develop your next set of Adobe Experience Cloud features or talk about the best approach to achieving your goals with AEM. If you are looking for an AEM development partner or have a question we can answer, contact us.