New around here? Start here. This category is for quick orientation: what the forum is about, how to post, what “authenticity” means in practice, and the basic rules that keep discussions useful. You’ll also find posting templates, examples of “good” threads, and any announcements that help you get moving fast.

For everything written: essays, blog posts, comments, scripts, emails. This is where we talk about what makes text feel human vs synthetic, why detectors often disagree, and what “proof” actually looks like (draft history, notes, revision trails, style consistency). Expect debates about AI-assisted writing too—where to draw the line, and how to disclose it without drama.

Is this photo real, edited, or generated? This category covers visual tells (lighting, reflections, hands, text, edges), manipulation patterns, and practical checks that don’t rely on vibes alone. Bring tricky examples (described, not generated) and discuss how you’d verify them—especially when images are compressed, reposted, or stripped of context.

Deepfakes, stitched clips, synthetic “talking heads,” and misleading edits. Here we break down what to look for (timing, lip-sync, motion, artifacts), how context gets distorted, and what verification steps actually hold up. It’s also the place for “how would you validate this clip before sharing it?” threads.

Voice cloning, synthetic speech, and edited recordings are getting good fast. This category focuses on what can be checked (odd pacing, breath/noise patterns, abrupt cuts, mismatch with known voice traits) and what’s unreliable. Also includes process topics like how to handle voice notes, calls, and “audio evidence” responsibly.

Practical workflows and repeatable checklists for verification across text, image, audio, and video. Tool comparisons are allowed, but the focus is on methodology: how to combine signals, avoid overconfidence, reduce false positives, and document what you did. If you’re building a “verification playbook,” it belongs here.

The moving front line of AI authenticity. Share big updates—new scams, high-impact incidents, policy shifts, and verification changes—then talk about the implications. Keep it summary + discussion. If you’re dissecting a specific artifact or running checks, use the relevant media category instead.

Receipts and trails: EXIF, timestamps, hashes, content credentials, watermarks, chain-of-custody, and platform metadata (when available). This is the more “forensic” category—what provenance can and can’t prove, how it gets lost in reposts, and how to set up lightweight provenance practices that normal people will actually follow.

Studies, benchmarks, and data on AI detection accuracy. Published findings, methodology discussions, and evaluation of AI detection tools.

AI regulation, academic integrity policies, institutional guidelines, and legal frameworks around synthetic and AI-generated content.