We launched AI translation on the stage at Mx3 in London

Labrador CMS has integrated artificial intelligence since the APIs from OpenAI became publicly available in May of this year. Now all customers of Labrador can run AI features on GPT 4 Turbo.

During a live demo on stage at Media Makers Meet - Mx3 in London on December 7th, Labrador CMS launched full automated AI translation of entire articles, including metadata, fact boxes, image captions, titles, tags, and all other text fields.

- This opens up entire new workflows and even new business ideas for cross-language publishing. This will help publishers get more distribution to their valuable content, said Labrador CMS CEO Jan Thoresen.

Structured data

Related story:

Is Labrador CMS a headless CMS? Yes, Indeed

Today, 259 websites are using Labrador in 10 countries, and the AI integration works in all languages.

Labrador CMS is built with structured data at its core, and we deliver continuous updates to all customers. We call this Beyond Headless. You get all the features you would expect from a headless system, and get access to lots of features on top of that. 

In this article, we have used our AI solution to create tags, write SEO titles and descriptions, write captions for the main image at the top of the page, for social media titles, and for a summary of the article itself.

141 languages available

The entire article has then been translated into a range of languages. We have included links to these articles below this article. Customers have 141 languages to choose from, and the translation works between all these languages. With 141 languages, there are 19,740 possible language pairs for translation.

All the languages available are listed in the fact box below the body text.

We received a lot of feedback after the presentation, both verbally and on LinkedIn.

- Your presentation was great; you have a great product

And.

- A Super impressive system!

Growing fast

Labrador CMS is a Norwegian startup and has launched more than one site every week this year. The total traffic from the servers is 2 billion pageviews per month.

The software is only sold to media companies, and our team connects the software to paywalls, print systems, InDesign, ad systems, analytics, mobile apps, video, and e-paper.

Since Labrador CMS is delivered as SaaS, all integrations will be system-to-system integrations that are stable and maintaned.

We import all the articles from the customer's old publishing platform into Labrador CMS and make sure all old URLs are redirected and still work after the migration.

New functionality every week

The system has been built to be easily integrated with any service, so when OpenAI was launched in November 2022, we were tripping on their API waiting list.

And then we integrated OpenAI and delivered it to all our customers. Now, they are all on Chat GPT-4.

We are also working to integrate different AI models. Support for Microsoft Azure OpenAI, Amazon Bedrock, and Anthropic Claude 2.1 are around the corner. When we get access to the Google Gemini API in a few weeks, we will add that too.

AI for productivity and quality

Labrador CMS has implemented AI to enhance productivity and ensure the quality of the content in articles. AI is solely used to refine the text that has already been input by journalists. Customers do not share any data with other customers, and content is not shared with OpenAI.

See this article translated below this article.