Media
3 items
Description
Detect manipulative tactics — ragebait, engagement bait, karma farming — on Reddit, with calibrated confidence and visible evidence. Reddit is engineered to make you react before you think. Ragebait, loaded language, outrage farming, karma traps — it's everywhere, and it's hard to spot in the moment. Gujob AI is a free Chrome extension that shows you exactly what's happening, right as you read. ── HOW IT WORKS ────────────────────────── Browse Reddit as normal. Highlight any post, comment, or thread text. Click the "Analyze with Gujob" button that appears — or press Alt+G. A result card opens in seconds. No page reload, no new tab. That's it. No setup, no settings to configure, no learning curve. ── WHAT IT DETECTS ─────────────────────── Gujob AI runs your selected text through a five-stage analysis pipeline: • Ragebait — content designed to trigger outrage without substance • Engagement bait — posts engineered for reactions, not real discussion • Karma farming — formulaic content optimised purely for upvotes • Loaded language — words chosen to provoke rather than inform • Outrage amplification — exaggeration designed to override critical thinking • Concern trolling — false concern used to derail or dismiss When the signal is too weak to call, Gujob returns "Unclear" instead of guessing. That's intentional — an honest "we don't know" is more useful than a confident wrong answer. ── WHAT YOU SEE IN THE RESULTS ─────────── Every verdict comes with the evidence it's built on — not just a label: • Verdict — Looks Fine, Mild, or Manipulative • Confidence score — calibrated so 70% confident means right ~70% of the time • Severity level — how strong the manipulation signal is • Evidence spans — the exact sentences or phrases that triggered the verdict, tagged with their tactic label (e.g. "Loaded language", "Outrage amplification") • Why we think this — the full reasoning behind the verdict, expandable • Feedback — tell Gujob if the verdict looks right or off ── PRIVACY ─────────────────────────────── Gujob only sees text you explicitly select and click Analyze. It never reads your feed in the background. Gujob does not collect your Reddit username, voting history, browsing history, or the URL of the page you're on. It does not run cookies, fingerprints, or advertising trackers. It does not sell data. Ever. Your text is not used to train models unless you explicitly opt in — and you can revoke that at any time from the extension popup. Full privacy policy: https://gujob.ai/privacy ── FREE TO USE ─────────────────────────── Gujob AI is free. No credit card, no subscription. • Anonymous: 5 analyses per hour — no account needed • Signed in (free): 240 analyses per day — one-click Google sign-in, no charge The signed-in tier exists to prevent abuse, not to lock features behind a paywall. Every feature is available on both tiers. ── WHAT GUJOB DOES NOT DO ──────────────── • Read your feed, inbox, or notifications in the background • Access any website outside of reddit.com • Store your Reddit identity or browsing history • Guarantee every verdict is correct — AI analysis can be wrong. Use it as one lens among many. Gujob AI is an independent tool. It is not affiliated with Reddit Inc.