Browse trending topics

Navigate to Trends in the sidebar. Sarah sees a live feed of trending topics sourced from hundreds of data sources across 50 countries. She applies two filters to narrow the results:

• Category -- Technology (one of 22 available categories)
• Country -- United States

Each trend card displays a score (0-100), a lifecycle badge, and a source count. Sarah scans for trends with a "peaking" badge and a score above 70 -- these have the strongest momentum and the richest source material available for article generation.

Sarah spots "OpenAI releases GPT-5 API" with a score of 87, lifecycle "peaking", and 12 sources across multiple platforms. That is her story.

Open the trend and start generation

Sarah clicks the trend card to open the detail panel. She reviews:

• Score chart -- A real-time chart showing the score climbing over the past 6 hours
• Source breakdown -- 12 sources from 5 different platforms, confirming broad interest
• Volume history -- Mentions spiking in the last 2 hours

A trend with 5+ sources from different platforms guarantees diverse material for the article. Satisfied, Sarah clicks "Write Article from Trend". This pre-fills the topic with the trend's title and context, giving the system a head start -- the knowledge base already contains pre-cached sources from the continuous background analysis.

Choose Agent mode

Sarah sees two generation modes. She picks Agent.

For her first article, Agent mode is the clear choice. The extra 45 seconds pays off with significantly richer content and more citations.

Configure generation parameters

Before generating, Sarah sets three parameters:

• Tone -- journalistic (other options: neutral, analytical, casual, formal)
• Length -- medium (~1,000 words) (other options: short ~500, long ~2,000)
• Language -- English (can differ from the source language of the trend)

She clicks "Generate" to start the process.

Watch the real-time generation stream

IANews streams the article to Sarah's browser in real time. She sees the generation progress through several phases:

1. Source Discovery -- IANews searches for the most relevant sources for the topic
2. Intelligent Search -- The system selects the best passages from collected content
3. Writing -- The AI generates the article paragraph by paragraph. Text appears in the editor in real time as it is produced
4. Citations -- Inline references [1], [2], etc. are inserted, linking each claim to its source

The entire Agent process typically completes in under 2 minutes. A progress indicator shows which phase is active. Sarah watches the article materialize in the editor, paragraph by paragraph.

Review in the editor

Once generation completes, Sarah lands in the AI-powered editor. She reviews the article:

Check citations -- She hovers over a [3] badge and sees a card with the source title, publisher name, publication date, a reliability score, and a text snippet. Clicking the badge opens the original source in a new tab. The Source Panel at the bottom groups all citations with color-coded reliability indicators.

Adjust the headline -- The generated headline is solid but generic. Sarah rewrites it to add her editorial angle: "GPT-5 API Launches With Reasoning Mode -- What Developers Need to Know."

Add a hero image -- She clicks the image inserter, searches "artificial intelligence" in the image library, and selects a relevant photo from thousands of free, high-quality images.

Use the AI sidebar -- Sarah selects a thin two-sentence paragraph about pricing and opens the AI sidebar. She clicks "Expand" and the AI adds a detailed comparison with previous pricing tiers. She then selects the conclusion and clicks "Adjust tone" to make it more forward-looking.

Publish the article

Sarah clicks "Publish" in the top-right corner and sees her available publishing targets:

The article status changes to "Published" and Sarah can track it from the Articles page. Total time from opening IANews to published article: under 5 minutes.