156 active jobs · updated May 11, 2026
AI training, search-quality rating, content review — async, flexible hours.
What used to be "content moderation" is now overwhelmingly AI-training and search-quality work. Rating search results against quality guidelines, labeling AI training data, providing feedback on LLM outputs (RLHF), evaluating image-recognition accuracy. Median pay sits at $19/hr across 156 active openings, with ceiling at $26/hr for specialty raters (medical, legal, multilingual).
The market is dominated by a handful of large hirers. Appen and Telus International AI are the two biggest publicly-traded players, both running rolling 1099 contractor programs at $14–$22/hr. Lionbridge (often surfaced under the brand TELUS Digital AI Data Solutions after the 2023 merger) covers similar work. Outlier and Scale AI are the rising RLHF-focused players paying $20–$30/hr for subject-matter experts in coding, math, medicine, and law. Smart Crowd (a Lionbridge sister brand) handles the entry-tier work at $14–$18/hr.
About 84% of listings are part-time and self-scheduled (you log in when you want, you log off when you want), making this the second-most flexible category after transcription. About 96% are phone-free. The catch: most projects are intermittent — you may have heavy assignment availability for 3 weeks, then nothing for 2 weeks. Stacking 2–3 platforms is the typical pattern.
Filter listings
Showing 5 of 156 listings
Concentrix · Business
Posted just now · ✓ Verified just now
Kiwimo-Product GmbH · Various
Posted just now · ✓ Verified just now
Concentrix · Business
Posted just now · ✓ Verified just now
Kiwimo-Product GmbH · Various
Posted just now · ✓ Verified just now
Spotify · Various
Posted 623 days ago · ✓ Verified just now
We'll send you new content moderation listings paying $19/hr or more.
Most cushy content-moderation work today is benign — search-quality rating, AI training data labeling, LLM response evaluation. The minority involving distressing user-generated content (typical for social-media trust-and-safety roles) is clearly labeled in our listings and pays a premium of $4–$8/hr above standard. We exclude trust-and-safety roles handling CSAM, terrorism content, or gore from this category entirely; those are specialized roles with mental-health support requirements and shouldn't be sold as "cushy."
Yes — both are large publicly-traded AI data companies (Appen on the ASX, Telus on the NYSE). Pay is reliable when work is available, but project assignments are intermittent — you may go weeks between active projects. Both run rolling 1099 contractor programs in the US, no relocation, no minimum hours, no shift schedule. Glassdoor ratings hover around 3.4–3.6, reflecting the project-availability variance rather than pay reliability.
For subject-matter experts (coders, medical, legal, advanced math): Outlier and Scale AI both pay $20–$45/hr depending on specialty, with Scale running more rigorous skill-assessment gates. Appen pays $15–$25/hr for similar work but has lower entry-skill bars. Generalist RLHF tasks (rating LLM responses on general knowledge): Appen and Outlier both pay $15–$22/hr — Outlier is generally faster to onboard.
Apply to Appen, Outlier, and Telus AI in parallel. Onboarding is independent and 1099 status means none of them require exclusivity. Most experienced raters keep 2–3 active accounts and switch when one platform's project pipeline dries up. Smart Crowd is a useful backstop for entry-level work between specialty assignments.
The work is shifting, not shrinking. Pure flagging and filtering tasks (the old "trust and safety" work) are increasingly automated. RLHF and human-feedback work is growing fast — every major AI lab pays humans to evaluate, correct, and align LLM outputs. The category has nearly doubled in active US listings since 2024. Specialty raters (medical, legal, multilingual) are the most insulated; generalist raters are the most exposed and should plan to specialize within 2–3 years.