Question

A media company wants to add automatic content moderation to its user-comment platform. The data science team has limited ML expertise, needs the solution running within two weeks, and wants to avoid managing any training infrastructure. The moderation categories map closely to standard toxic-language and spam detection. Which Google Cloud approach best fits these constraints?

Accepted Answer

Use a pre-built Cloud Natural Language API to classify comment content out of the box

Answer

Provision a TPU pod in Vertex AI and train a custom moderation model from scratch

Answer

Fine-tune an open model from Model Garden on a large labeled moderation dataset

Answer

Build a multi-agent system in Vertex AI Agent Builder to route comments between reviewers