TwoSided

Every story has two sides.
Now you see both.

TwoSided is a global news aggregator that uses AI to show you the left and right perspectives on every major story — side by side, without taking sides itself.

See our sources

The problem with news today

Most news apps make the problem worse, not better.

News is filtered

Your feed shows you articles from sources that match your existing views. Over time you see less of what challenges you.

Context is missing

A single article rarely tells the whole story. You need to visit five sites to understand what actually happened.

Bias is invisible

Most readers can't tell whether a story leans left or right — or whether their own source covers it at all.

How TwoSided works

From raw articles to a balanced perspective in five steps.

01

We fetch from 130+ sources

Every few hours our worker pulls articles from RSS feeds across the political spectrum — BBC, Fox News, Al Jazeera, The Hindu, Reuters, and 125 more — spanning 25+ countries and 10+ languages.

02

AI clusters related articles

Using sentence-transformer embeddings (all-MiniLM-L6-v2), we group articles about the same event into a single story cluster — so ten articles about the same trade deal become one story with ten perspectives.

03

We tag and score each story

Each cluster gets a category (Politics, Tech, Climate…), bias counts (how many left/center/right sources covered it), and a coverage score reflecting how globally the story was reported.

04

AI writes the left and right perspectives

We use Groq (Llama 3) to read all the articles in a cluster and write two neutral summaries — one reflecting what left-leaning sources emphasise, one reflecting what right-leaning sources emphasise. You see both.

05

You get the full picture

The feed shows you stories that matter, ranked by coverage breadth. Click any story to see the AI perspectives, source breakdown by bias, and links to original articles.

Where AI is used

We use AI as a tool for organisation and synthesis — never to generate news, invent facts, or take a political position.

Story clustering

Sentence embeddings group articles about the same event across sources

Category tagging

LLM classifies each cluster into Politics, Tech, Economy, etc.

Left perspective summary

LLM synthesises what left-leaning sources emphasise in a story

Right perspective summary

LLM synthesises what right-leaning sources emphasise in a story

Impact analysis

LLM identifies who is affected and how, per topic area

Neutral summary

A balanced overview written without taking sides

Powered by Groq (Llama 3.1) for text generation and sentence-transformers (all-MiniLM-L6-v2) for semantic clustering.

Who uses TwoSided

🎓

Students & researchers

Understand multiple viewpoints before forming an argument

📰

Journalists

See how differently the same story is covered across the political spectrum

🗳️

Informed citizens

Cut through partisan framing and understand what actually happened

🌍

Global readers

Follow stories with international coverage from your country's perspective

💼

Business & policy

Track how regulatory or economic news is perceived across different markets

Our commitment to transparency

  • Every article links back to the original source — we don't host or rewrite news
  • Source bias labels are based on established media bias research
  • AI-generated perspectives are clearly labeled — you always know what's AI and what's original
  • State-controlled media sources are flagged
  • Coverage gaps are shown — if only one side covered a story, we tell you

Ready to see the full picture?