LinkedIn AI Detector

Estimates whether LinkedIn posts were likely written by AI or a human. Scores are AI-generated guidance, not proof.

LinkedIn AI Detector analyses the posts and comments you see in your LinkedIn feed and gives each one an instant AI-likelihood score, so you can tell at a glance whether something was written by a real person or generated by an AI tool. How it works As you scroll through LinkedIn, a small row of five icons appears next to each post author. Robots (🤖) represent AI-written content, humans (👤) represent human-written content. The more robots you see, the more likely the post was AI-generated. Hover over the icons to see the exact percentage score and a one-sentence explanation of why. Why install it? LinkedIn is increasingly flooded with AI-generated content — posts that look polished and insightful but were written by ChatGPT or similar tools in seconds. This makes it harder to find genuine voices, real experiences, and authentic professional insight. LinkedIn AI Detector gives you a quick signal to help you cut through the noise and focus on the content that actually comes from real people. Features • Automatic scoring of posts as you scroll — no clicks needed • Optional on-demand mode: a "🔍 Check for AI" pill appears on each post or comment, and you click it only when you want to check • Comment analysis — score individual comments on posts, including nested reply threads • Separate controls for posts and comments: Auto, On-demand, or Off independently • Sentiment or neutral colour mode, choose whether robots appear in warning colours or neutral grey • Configurable minimum text length, skip very short posts that aren't worth analysing • Results are cached for the session, so re-rendered posts never make a duplicate API call • Works with "Load more comments" and dynamically loaded content Scoring guide 0–20% → Human written (5 humans) 21–40% → Likely human (1 robot) 41–60% → Uncertain (2 robots) 61–80% → Likely AI (3 robots) 81–90% → Mostly AI (4 robots) 91–100% → AI written (5 robots) Important — requires your own Claude API key This extension uses Anthropic's Claude Haiku model to analyse text. You must supply your own API key from Anthropic's platform (console.anthropic.com). The key is stored securely in your browser using Chrome's encrypted storage and is only ever sent directly to Anthropic's API, never to any third-party server. Cost is very low: roughly $0.001 per analysis (one tenth of a cent). Analysing 100 posts costs about 10 cents. New Anthropic accounts receive free credits that are more than enough to get started. Privacy • Your API key is stored locally in Chrome's encrypted sync storage • Post text is sent directly from your browser to api.anthropic.com, it does not pass through any intermediary server • The extension collects no personal data and has no backend of its own A note on accuracy Scores are AI-generated estimates provided for guidance only. No AI detection method is 100% accurate — results can be wrong in both directions. A high score does not prove a post was AI-written, and a low score does not guarantee it was human-written. Use scores as one signal among many, not as definitive proof.