Every time you scroll Twitter, YouTube, or TikTok, you expose unfiltered neural pathways to content optimized by supercomputers to hijack your emotions. There's a better way.
In cybersecurity, a "prompt injection" tricks an AI into doing something it wasn't designed to do. Social media algorithms do the same thing to you.
Algorithms A/B test billions of users to find the exact content pattern that makes you unable to look away. Your attention span has dropped from 2.5 minutes to 47 seconds since 2004.
Twitter's algorithm shows 62% angry political tweets vs. 52% in chronological order. Each additional negative word in a headline increases clicks by 2.3%. Outrage is the product.
The more empathetic you are, the more vulnerable. Moral-emotional language increases content spread by 20% per word. Your ability to care is weaponized against you.
You wouldn't browse the internet without a firewall. You wouldn't open random email attachments. You wouldn't give every app root access to your computer.
Yet every day, you give algorithmically-curated feeds root access to your emotional state. No filter. No firewall. No antivirus. Just raw, optimized content injected directly into your attention. Research confirms 31% of social media use is driven by self-control problems, not genuine desire.
Social media uses the same mechanism as slot machines. Your brain releases dopamine not from the reward itself, but from the unpredictability of the reward. Each scroll is a new pull of the lever. Research shows this creates tolerance development analogous to substance addiction (Sharpe & Spooner, 2025).
Allcott et al. 2020 (AER): 2,743 paid Facebook deactivators showed improved well-being. Brady et al. 2021 (Science Advances): social feedback on outrage trains users to post more outrage. Lambert et al. 2022: 1 week off reduced depression by 2.2 pts and anxiety by 1.7 pts.
You can't out-think a supercomputer optimizing for your attention. But you can use a different AI — one that works for you — to filter the noise.
Ragebait, outrage, clickbait, engagement traps, emotional manipulation
Grok, Perplexity, Claude, ChatGPT — ask for exactly what you need
Factual, relevant, actionable — no emotional manipulation
You don't need to quit the internet. You need to block the algorithmically-curated feeds while keeping access to everything useful. Here's how, for every OS.
twitter.com, x.com, tiktok.comyoutube.com but add exceptions +youtube.com/watch* and +youtube.com/@*youtube.com##ytd-rich-section-renderer:has(#title-text:has-text(Shorts))youtube.com##ytd-browse[page-subtype="home"] ytd-rich-grid-rendereryoutube.com##ytd-watch-next-secondary-results-renderersudo nano /etc/hosts0.0.0.0 twitter.com, 0.0.0.0 x.com, 0.0.0.0 tiktok.com (include www variants)sudo systemd-resolve --flush-cachessudo chattr +i /etc/hosts (requires chattr -i to undo)curl -sSL https://install.pi-hole.net | bashgit clone https://github.com/StevenBlack/hosts.gitcd hosts && python3 updateHostsFile.py --extensions social --replaceC:\Windows\System32\drivers\etc\hosts (set filter to All Files)0.0.0.0 twitter.com (one domain per line, include www)ipconfig /flushdns in Command Promptsudo nano /etc/hosts0.0.0.0 twitter.com entries (one per line)sudo dscacheutil -flushcache; sudo killall -HUP mDNSRespondersudo chflags schg /etc/hosts (system immutable flag)You don't need to do everything at once. Check off each step as you go. Every box you check is bandwidth reclaimed from algorithms that don't care about you.