AI content is everywhere on social media, especially LinkedIn
AI‑generated “slop” is rapidly saturating LinkedIn and other major platforms, leaving many users feeling that feeds are now dominated by low‑effort, LLM‑written posts and comments. While some see tools like Pangram’s AI detector as a way to preserve human spaces online, others doubt such detectors’ accuracy and argue that LinkedIn content was vacuous long before AI. Several contributors describe abandoning LinkedIn or aggressively filtering their feeds, yet concede the site remains hard to avoid for networking and job hunting in an already bleak market.
Perception of LinkedIn and AI “slop”
- Many see LinkedIn as “AI‑slopped wasteland” or “dumpster fire” that was already full of corporate platitudes; LLMs mainly multiplied the volume, not changed the character.
- Feed is described as unusable: obvious AI posts, emoji‑laden “thought leadership,” contrived moral stories, and AI responses even in comments.
- Some say AI content is actually an improvement over prior corporate/bro hustle posts.
Broader Internet and “Dead Internet” Feel
- Users report similar slop on Reddit, X, YouTube, Instagram, Medium, Substack, and in product/service reviews.
- Sense that bots, engagement farms, and LLM‑generated text/images are crowding out human signal and making participation less appealing.
- A “dead internet theory” vibe: people assume many others online are bots or AI.
LLM Style Bleeding into Human Language
- Multiple commenters notice people unconsciously mimicking LLM phrasing and structure (listicles, “why it matters,” certain idioms).
- Some deliberately avoid AI‑ish tropes they now find grating; others find themselves adopting them after heavy AI use.
- Discussion that this is just another turn in language evolution, now co‑shaped by AI training and RLHF.
LinkedIn’s Remaining Utility for Work
- Strong split: some deleted accounts and rely on email, others say nearly all their jobs or projects over many years came via LinkedIn.
- Complaints about fake/“ghost” jobs, spammy recruiters, political/rage content, and poor job search UX, but acknowledgment that recruiters still heavily use the platform.
AI Writing: Voice, Thinking, and Access
- One camp: outsourcing writing to AI erodes personal voice and the thinking that writing forces; see it as ethically and cognitively harmful.
- Counter‑camp: most people write poorly or in a non‑native language; LLMs let them sound competent, reach audiences, and “find a voice” they couldn’t otherwise.
- Concern that AI gives an illusion of competence and may stunt real skill development.
AI Detection (Pangram) Debate
- Some praise Pangram as a valuable classifier that rarely mislabels genuine human text and could help preserve human spaces.
- Others call it “snake oil”: report false positives on their own writing, point to biases against non‑native English, and note how easily minor edits can fool detectors.
- Disagreement over claimed 0.01% false positive rate; several are highly skeptical.
Coping Strategies and Alternatives
- Suggestions: aggressively mute/“not interested” AI‑ish content, keep very small LinkedIn networks, or leave corporate platforms entirely.
- Interest in RSS, blogs, webrings, human‑verified or smaller communities, and browser plugins to filter slop.