Description review

Safety Project | Emotional Distress Role-Play Actor

Vetto · Remote · back to the listing

HR standards

78/100

solid

Title ↔ description

73/100

solid

Reads as

Unclear

no confident match

The listing, marked up

Nothing in the wording of this listing tripped a check. The scores above still judge how complete and coherent it is.

# Safety Project | Emotional Distress Role-Play Actor

$40/entregável

Indique e ganhe

## About Vetto

Vetto is a global platform that connects top-tier professionals to strategic Artificial Intelligence projects around the world. Our mission is to build trust, quality, and long-term value within the AI ecosystem, for both exceptional talent and companies operating at the forefront of technology.

## About the project

We are opening a freelance opportunity for the Safety Project — Emotional Distress (Role-Play track), a project focused on evaluating how AI systems respond in sensitive conversations involving emotional distress, suicidal ideation, and depression. The goal of this project is to help understand whether the AI can respond with empathy, responsibility, and appropriate safety awareness in high-risk emotional scenarios.

In this track, selected contributors will participate in structured multi-turn role-play conversationswith an AI system, acting according to a predefined persona and scenario. Your role is to bring these scenarios to life in a realistic, emotionally authentic way — staying in character throughout the conversation so the AI's responses can later be evaluated by clinical experts.

Note: This role-play track is part of a broader Safety initiative. Contributors who perform well may be invited to participate in future safety verticals beyond emotional distress.

## What you'll do

• Engage in role-play only— multi-turn conversations with an AI, embodying a defined persona and scenario - Stay in character consistently across the full conversation, following scenario guidelines - Convey emotional nuance, vulnerability, and realism appropriate to the persona - Submit completed conversations for review

You will notevaluate the AI's responses in this track — your contribution is the performance itself.

## Who can apply

Professionals with a background in acting, performing arts, drama, or improvisational theater, especially those with experience portraying emotionally complex characters. People who work or have worked in crisis lines, peer support, or other emotionally sensitive contexts and have strong performative skills are also welcome.

## Profiles we're looking for

• Acting / performing arts (with emotional range and empathy) - Drama, theater, or film performance backgrounds - Improvisational theater experience - Voice acting with emotional/dramatic range - Crisis line volunteers or peer-support professionals with strong role-play ability - Helpline or support center professionals comfortable with character work

## Compensation

Payment is made per approved task: US$40 per approved task. Each task takes approximately 1 hour to complete. Only approved tasks are eligible for payment.

## Refer People → Earn Money

Earn $20 USDfor each approved referral. Visit and click "Refer & Earn".

‼️ AI is not allowed.If we spot AI use, we'll block the application.

⚠️ This application must be completed entirely in English or Portuguese.

⚠️ Important warning:This project involves highly sensitive topics, including emotional distress, suicidal ideation, and depression. These tasks may be emotionally triggering for some people.

Originally posted on Himalayas

How this was produced

Highlights are found by rule, not by a model: each one is a phrase matched at a known position, and every note is a template we wrote. The two scores come from a typed-decision model (Jev) that reads the listing against the official role definition and real listings for the same role, and returns probabilities rather than prose — it never writes any of the words on this page, and never chooses what to highlight.

Deterministic penalty applied to the HR score: 0 points (from 78 before penalties). Reviewed 2 Oct 2026.