A voice guide to what's actually going on
Every day, something new to worry about. A war. An outbreak. A political crisis. A headline designed to make your stomach drop. Your phone buzzes, your chest tightens, and before you've even read the article, your mind is three catastrophes ahead of reality.
But how much of it is signal — and how much is noise?
The problem
We're drowning in information and starving for clarity. News is a model — a simplified story about something messier and more uncertain than the headline allows. Social media makes it worse: fear is contagious, outrage breeds outrage, and the algorithm doesn't care whether you're informed or just anxious. Most of the time, you're not reacting to what's actually happening. You're reacting to everyone else's reaction.
The philosopher René Girard called this mimetic desire — we borrow our wants, our fears, our outrage from the people around us, often without noticing. Your worry about a crisis might be perfectly rational. Or it might be borrowed panic, amplified by proximity, inflated by repetition, and scaled to your emotional stakes rather than the actual situation.
The question isn't whether to care. It's whether your concern is yours — and whether it's proportionate to what's real.
What Mostly Noise does
Mostly Noise is a voice-powered guide that helps you cut through the noise on any topic you're worried about. You speak to it. It listens. Then it does three things:
It finds out what's actually happening. Using real-time web search, it pulls the latest facts, figures, and reporting — not from its training data, but from the live web, right now. What's known. What's disputed. What's genuinely uncertain.
It examines what's being projected. This is the part most news coverage skips. Are people catastrophising a future that hasn't arrived? Is the panic borrowed — spreading because everyone else is panicking, not because the facts warrant it? Is a political narrative being mistaken for reality? Is your own proximity to the situation changing how loud the signal feels?
It grounds you in the present. What can you actually do, right now? Sometimes there's a practical action. Sometimes there's genuinely nothing — and the honest answer is: pay attention, hold it lightly, and stop performing concern as a substitute for action.
The philosophy
The thinking behind Mostly Noise comes from the upcoming book Mostly Noise by Andy Coughlan, which makes a simple but far-reaching argument: only the present moment is real. The past is a reconstruction. The future is a simulation. Most of our anxiety lives in one of those two imaginary places.
The book traces how projection works — in our stress responses, our desires, our politics, our relationships — and makes the case that the opposite of projection isn't calm. It's engagement: creative, imperfect, ongoing attention to what's actually here.
Mostly Noise puts that framework to work on the news. Not to dismiss your concerns, but to examine them honestly — and to help you tell the difference between a situation that needs your attention and a situation that's mostly noise.
How it's built
- ElevenLabs Conversational AI provides the voice — a natural, conversational agent you speak to directly, with the warmth and wit of a well-read friend who happens to have read everything.
- Firecrawl Search provides the knowledge — real-time web search that returns structured, current content rather than stale training data.
The philosophical framework sits between them: a system prompt that encodes the book's core ideas into a conversational analysis tool, without jargon or lectures.
The situation is, as far as anyone can tell, mostly noise.