Here are my slides for the talk "Successful strategies for implementing AI content workflows":
https://ai-sidekick.app/drupal-ai-online-meetup-slides
In considering our approach to AI-generated content, the emphasis should be on the quality of the text rather than the means by which it was created.
It's important to recognize the distinction between AI that generates content independently and AI that is employed to articulate and polish a user's pre-existing ideas effectively. While the first can be problematic, the second is merely better grammar correction.
Just to prove my point, AI mostly wrote this based on: `Write a comment: Focus on text quality, not how created; Differentiate AI can be used to make up content, or to formulate users' ideas nicely - the second is good.` Having a first draft and then refining it is much easier for me than writing from scratch.
PS: If someone were actually able to provide correct and helpful comments with an LLM and RAG, that would be wonderful. I wasn't able to do this yet.
While AI is a gimmick in most commercial products, there’s a significant move toward being integrated into the daily workflows of editors in WordPress.
An overview:
- Many solutions like Wix solve only the blank page problem by creating some generic example content based on your industry. These tools are not very interesting.
- WordPress is noticeably making the biggest strides in integrating AI in useful ways. WordPress itself released a Jetpack AI to generate content and even full articles in Gutenberg. At the same time, Elementor, one of the biggest site builders in WordPress, released their Elementor AI integration, which also writes small code snippets for editors.
- CMS-independent solutions like Notion, Microsoft Office and Fraser are developing faster and show some impressive results
I’m sure there are three distinct areas where AI will make an impact on the way editors work over the next month:
- Automatic, task-optimized integrations will ease the editors’ work. For example, the CMS can automatically generate a good meta description for a page based on an optimized prompt. These implementations require a combination of classical programming and AI models.
- Brainstorming and Writing Assistance will help both people who occasionally update their website and professional editors who create content all day. There will be audience-specific solutions for both. Integrating the CMS-knowledge like exiting content, visitor counts other metrics can offer a real advantage over generic tools like Notion.
- Automating non-text creation work. I imagine in a year you will copy your Word text inside Drupal and it will choose which how to create an layout the page, automatically create a draft for you, maybe even choosing example images from the media library.
Hello Michael,
do you have custom CSP rules, or if you use Drupal Security Kit, please add the wildcard URL https://*.ai-sidekick.app to the frame-src field.
Does that solve your issue?
Hello Meike,
At the moment, the concept is that your editors are trusted folks since they can also change what your website says. So, if someone overuses it for personal things, the website may run out of credits for the month.
On our enterprise plans, we can configure a custom user limit for your editors. Right now, there isn't a user interface for this, but we can do it on the backend. Sound like a solution?