Video summary

DESCOBRI UMA EMPRESA QUE PAGA EM DÓLAR PARA BRASILEIROS TREINAREM IA (SEM APARECER E PELO CELULAR)

Main summary

Key takeaways

Business

Business Opportunity Overview (What the Video Is Pitching)

  • The video claims you can earn money from home by training/evaluating AI for global AI companies.
  • It states that payments are made in USD or EUR with weekly payouts.
  • The pitch frames AI testing as an ongoing operational need: AI isn’t perfect, so companies require human evaluation to:
    • assess AI outputs
    • label data
    • improve language accuracy (including Portuguese)

Companies Highlighted (Execution-Focused)

1) Data Annotation

Positioning

  • “Connect human genius with artificial intelligence” by hiring people to evaluate, translate, and improve AI outputs.

Scale / Traction Metrics Mentioned

  • Paid over $150M to collaborators (trainers/testers)
  • 120M completed entries
  • 100,000+ annotators worldwide
  • 55+ languages supported

Compensation / Earning Targets Mentioned

  • $5 to $40/hour (language-specific roles)
  • Another cited range: $25 to $40/hour (task/difficulty dependent)

Work Model / Operations

  • Remote work
  • Flexible hours
  • Treated as self-employed (not registered as an employee)
  • Weekly payments
  • Described process:
    1. Apply
    2. Qualify via an evaluation (video/voice test)
    3. Work
    4. Get paid
  • Emphasis on evaluation rather than a traditional interview.

Example Tasks

  • Evaluate English ↔ Portuguese translations for:
    • fidelity
    • fluency
    • tone (described as “like an editor”)
  • Identify issues in intonation/tone, then:
    • provide an explanation
    • produce an improved version when the model output isn’t sufficient

Hiring Funnel / Segmentation

  • Beginner/generalist roles vs specialist roles (e.g., Portuguese specialization)
  • Claim: Portuguese may have more vacancies than English due to lower supply of qualified Portuguese annotators

2) Outlier

Positioning

  • A platform/company that “powers the world’s most advanced AI,” hiring people to train AI.

Difference vs Data Annotation (As Stated)

  • “A little more advanced” (minor differentiation), while still offering some beginner-friendly roles.

Compensation Examples / KPIs Mentioned

  • $15/hour for an “English conversation evaluator” (example given)
  • “More knowledge = higher pay” (no strict formula, but price-per-task scales with expertise)
  • Higher-paid categories referenced (examples, exact hourly not consistently stated):
    • “French conversation evaluator” (~higher, exact rate not consistently given)
    • “Korean 31”
    • “Thai”
    • Other specialized roles (e.g., educators, math knowledge for AI training)

Operations and Selection

  • Less “interview-like” than the first platform; described as more straightforward than typical interviews
  • Emphasis on selecting people for expertise, with top earners suited to harder specialties

3) Remotasks (Described as the Third Option)

Framing

  • Targeted at people with more knowledge (higher difficulty, higher pay).

Compensation Examples / Range Mentioned

  • $30 to $80/hour for video editing (example)
  • “English teacher for diction classes” example: $65/hour
  • Psychology specialist example: $100 to $300/hour

Concrete Specialist Job Types Mentioned

  • Audio specialist (engineer implied)
  • Senior legal consultant

Rationale Provided

  • Specialists help build more accurate “knowledge banks” for AI capabilities (example used: expert-backed/doctor-like guidance for medical contexts)

Frameworks / Playbooks Used (Explicit or Implicit)

No formal frameworks (e.g., OKRs/SWOT/GTM) are presented. However, the video uses a selection-and-scaling marketplace playbook:

  • Supply segmentation
    • Beginners vs specialists (language knowledge, academic background, domain expertise)
  • Qualification gating
    • Apply → evaluation/assessment → qualify → begin tasks
  • Task-to-pay matching
    • Harder/specialized tasks pay more
  • Market logic / demand assertion
    • AI adoption creates ongoing demand for human-in-the-loop QA and labeling

Key Metrics / Numbers Captured (As Stated)

Data Annotation

  • $5–$40/hour
  • $25–$40/hour (another cited range)
  • $150M+ paid to collaborators
  • 120M completed entries
  • 100,000+ annotators
  • 55+ languages
  • Weekly payouts

Outlier

  • Example role: $15/hour (conversation evaluation)
  • Claim: higher expertise → higher pay (multiple category examples; no consistent universal hourly range)

Remotasks

  • $30–$80/hour (video editing example)
  • $65/hour (diction/English teaching example)
  • $100–$300/hour (psychology example)

Actionable Recommendations (How to Start, Per the Video)

  • Choose a platform based on your profile:
    • If you have no knowledge: start with Data Annotation (claims include beginner options)
    • If you have academic/domain expertise: consider Outlier or Remotasks
  • Follow the onboarding steps:
    • Create account → apply → pass the evaluation/test → begin tasks → earn via weekly payouts (for Data Annotation)
  • Stand out by specialization:
    • The video argues that rarer skills/languages/domains lead to faster approval and more projects

Leadership / Mindset Tactics Used to Drive Action

  • Strong “do it vs complain” framing:
    • “Locomotives” = apply, act, build success
    • “Wagons” = skeptical, waiting for proof, quitting early
  • Call-to-action:
    • Apply immediately and avoid over-worrying that “it’s too good to be true”

Presenters / Sources Mentioned

  • Thiago (main presenter)
  • Rian (co-presenter)
  • Warren Buffett (quoted to support the “opportunity is dismissed as scams” mindset)
  • Company sources named:
    • Data Annotation
    • Outlier
    • Remotasks

Original video