ChatGPT is politically compromised, psychologically addictive, and replacing real human thought. Learn to spot AI content, reclaim your attention — and help build what comes next.
We write every issue ourselves. No bots, no ghostwriters. Drop your email and say hi.
Powered by Substack · No spam · Unsubscribe anytime
It's not just a chatbot. It's a political actor, an addiction machine, and an environment destroyer.
OpenAI's leadership donated $25M+ to Trump's super PAC. Your subscription funds political influence operations.
Psychiatrists are documenting AI psychosis — users losing grip on reality after extended chatbot use. ~500K weekly ChatGPT users show signs of mania.
1 in 5 Americans have used AI for romantic companionship. A 37,000-member Reddit community exists for people dating ChatGPT.
OpenAI lobbies for fast, unregulated expansion of data centers — no matter how it blows up electricity prices or burns fossil fuels.
AI-generated slop is flooding the internet, destroying trust in authentic human work, art, writing, and video.
OpenAI agreed to let the Pentagon use its tech for "any lawful purpose" — including mass surveillance and autonomous weapons.
Simple rules anyone can use — no tech knowledge needed. Train your eye.
AI-generated faces often have eyes that don't blink naturally in video, or that have inconsistent gaze direction. Look for glassy, overly symmetric eyes.
EasyAI images of hands are notoriously bad. Look for extra fingers, merged digits, or unnaturally smooth knuckles. Still a reliable tell in 2025.
EasyAI text overuses phrases like "certainly", "of course", "it's worth noting", and "delve into". Sentences are perfectly structured — too perfectly.
EasyAI videos use a distinctive, smooth background blur that looks different from real camera depth-of-field. Edges of subjects often look "melted" into the background.
MediumAI-cloned voices have zero micro-hesitations. Real speech has tiny uhms, breaths, and rhythm breaks. Perfectly fluent = suspect. Also check for flat emotional range.
MediumDrop any suspicious image into Google Images or TinEye. AI images often have no origin trail — no original source, no photographer, no date context.
Advanced5 quick questions. Train your instincts before you scroll your next feed.
100% human-made. No AI narration, no AI visuals. Curated weekly by our team.
We're building a community of people who value authentic human content, want to reduce AI dependence, and are building toward a future where technology serves people — not corporations.