Case Study

True AutoMod

Long-Term Scalability with Artificial Intelligence

True Digital Group is part of the True Corporation conglomerate, and the parent of the TrueID In-Trend platform, which allows registered users to contribute local news stories for a small payment. With over 68,000 regular contributors, the site’s content moderation had to be improved. Together with Fluxus, True made the switch from human moderators to an AI and machine learning-based system. This is how we made that happen and how it helped grow the platform.

Proving Viability

True had set an ambitious target: reaching 8 million contributors. Using human moderators for such numbers was instantly recognized as unachievable, so, Fluxus presented automated moderation as a solution.

The system was designed so an article would need to pass two levels of approval. Firstly, it would provide real-time sentiment analysis to contributors while writing, and run SEO checks. After passing this level, it would progress further, going through AI-based text and image plagiarism checks, profanity checks, and other filters. Upon passing everything successfully, the piece would upload automatically, while getting passed on to a human moderator if certain checks failed. Other features were included like providing feedback for edits or amendments in articles, community moderation through a report function, and automated payment for contributors.

Solutions and Processes

Our team built a profanity filter from a list of community-flagged words, and an SEO evaluator, which analyzed text formatting, keywords, and word counts. Sentiment analysis was implemented using a combination of custom-built services and third-party AI tools. Using already-approved and published content, Fluxus trained the AI model to recognize identifiable traits of an acceptable piece.

The Fluxus team designed their own text plagiarism system for cost efficiency, and implemented AWS Rekognition for image analysis and copyright checks. Other plagiarism tools are in R&D stages such as our ElasticSearch feature to analyze content from all over the Thai internet.

Contextual analysis was also implemented to correctly tag and organize articles.


The innovations Fluxus helped True set in place have allowed for the following cost savings:

• Per-article moderation $0.05 from $2

• Yearly moderation $18,000 from $750,000 (40x saving)

• Per-article text analysis $0.01 from $1

These financial benefits allowed True to effectively scale the platform for the long term. With the success of their In-Trend service in Thailand, plans are in place for launches in Indonesia and The Philippines. Fluxus continued to stay in contact with True throughout 2020 and into 2021 for further AI development on the platform.

To scale your own AI and machine-learning based challenges, get in touch with us today!

Next Case Study: