Terms of Service

1. Agreement Overview

These Terms of Service ("Terms") constitute a legally binding agreement between you and Opus Data, Inc. ("Opus Data," "we," "us") governing your access to and use of our AI training platform, annotation tools, project marketplace, and related services (collectively, the "Platform").

By creating an account, submitting an application, or accessing the Platform in any capacity — whether as an AI Trainer, Enterprise Client, or visitor — you acknowledge that you have read, understood, and agree to be bound by these Terms. If you are using the Platform on behalf of an organization, you represent that you have authority to bind that organization.

2. Platform Description & Services

Opus Data operates a two-sided marketplace connecting AI Trainers with Enterprise Clients who require human-in-the-loop data services. Our services include, but are not limited to:

• •Reinforcement Learning from Human Feedback (RLHF) — ranking and rating AI model outputs for quality, safety, helpfulness, and factual accuracy
• •Data Annotation — text classification, named entity recognition, image labeling, video annotation, audio transcription, and sentiment analysis
• •Domain Expert Evaluation — specialized review of AI outputs in coding, medical, financial, analytics, and scientific contexts
• •Red Teaming & Safety Testing — adversarial prompt testing, bias detection, and safety vulnerability assessment
• •Customer Discovery & Market Research — evaluating AI-generated market insights, user sentiment, and survey analysis
• •Multilingual Services — translation quality assessment, localization review, and cross-lingual annotation in 50+ languages

3. Trainer Eligibility & Onboarding

To apply as an AI Trainer, you must be at least 18 years of age, legally authorized to work as an independent contractor in your jurisdiction, and possess the domain expertise indicated in your application. All trainers undergo a qualification process:

• •Submit a complete application with verifiable professional background, skills, and portfolio links
• •Pass a domain-specific qualification assessment (Coding, Medical, Finance & Banking, Analytics, Customer Discovery, Multilingual, or Science)
• •Agree to our Trainer Code of Conduct, Confidentiality Agreement, and these Terms
• •Complete platform orientation including annotation guidelines, quality standards, and tool training

Opus Data reserves the right to deny, suspend, or revoke trainer access at any time based on qualification results, quality performance, conduct violations, or platform needs.

4. Compensation & Payment Terms

Trainers are compensated as independent contractors on an hourly or per-task basis, as specified for each project. Key payment terms:

• •Rates vary by domain and project complexity: Domain Experts ($40-$100+/hr), Generalists ($20-$35+/hr)
• •Payouts are processed weekly (every Friday) for work completed and approved through the prior Sunday
• •Payment methods: direct bank transfer (US), PayPal, and Wise (international). Minimum payout threshold is $25
• •Opus Data covers all transaction fees — the approved amount is the amount you receive
• •Quality bonuses: 1.2x multiplier at 95%+ quality score, 1.5x at 99%+
• •Trainers are solely responsible for reporting income and paying applicable taxes. US trainers earning $600+ annually will receive IRS Form 1099-NEC

5. Independent Contractor Relationship

Trainers operate as independent contractors, not employees of Opus Data. You set your own schedule, choose which projects to accept, and may perform services for other platforms. Opus Data does not provide employment benefits, insurance, retirement plans, or tax withholding. You are responsible for your own equipment, workspace, internet connection, and any costs associated with performing services. Nothing in these Terms creates an employer-employee relationship, joint venture, partnership, or agency arrangement.

6. Confidentiality & Data Handling

All project data, client information, model outputs, annotation guidelines, and internal platform details are strictly confidential. As a Trainer, you agree to:

• •Never share, copy, download, screenshot, or distribute any project data, AI model outputs, or annotation content outside the Platform
• •Not disclose the identity of Enterprise Clients, project specifications, or any non-public information about AI models you evaluate
• •Use project data exclusively within the Platform annotation interface for the purpose of completing assigned tasks
• •Delete any locally cached or temporarily stored data immediately after task completion
• •Report any data breaches, unauthorized access, or security incidents to security@opusdata.ai within 24 hours

Violation of confidentiality obligations may result in immediate account termination, forfeiture of pending earnings, and legal action including claims for damages.

7. Intellectual Property

All annotations, labels, ratings, evaluations, written feedback, and other work product created by Trainers on the Platform ("Work Product") are the exclusive property of Opus Data and/or the applicable Enterprise Client. By submitting Work Product, you irrevocably assign all intellectual property rights therein. Opus Data retains all rights to the Platform, its technology, tools, interfaces, processes, and proprietary methodologies. Enterprise Clients retain ownership of their source data submitted for annotation.

8. Quality Standards & Account Standing

Opus Data maintains rigorous quality standards. Trainers are evaluated on accuracy (50%), consistency (30%), and completeness (20%) on a rolling 30-day basis. Consequences for substandard performance:

• •Quality score below 80%: 2-week improvement period with calibration tasks and additional review
• •Repeated low scores: restricted project access until successful completion of re-qualification
• •Fraudulent submissions (copy-paste, AI-generated answers, random selections): immediate termination and forfeiture of pending earnings
• •Guideline violations: warnings for first offense, project removal for repeated violations

9. Enterprise Client Terms

Enterprise Clients engage Opus Data under separate Master Service Agreements (MSAs) or Statements of Work (SOWs) that supplement these Terms. Enterprise engagements include SLA commitments for turnaround time, accuracy targets, and volume guarantees. Custom security requirements, dedicated annotation teams, and white-label solutions are available. Enterprise pricing, invoicing, and payment terms are governed by the applicable MSA/SOW. Contact enterprise@opusdata.ai for details.

10. Limitation of Liability

To the maximum extent permitted by applicable law, Opus Data, Inc., its officers, directors, employees, and affiliates shall not be liable for any indirect, incidental, special, consequential, or punitive damages, including loss of profits, data, business opportunities, or goodwill, arising from or related to your use of or inability to use the Platform — regardless of the theory of liability. Our total aggregate liability for any claims arising under these Terms shall not exceed the total compensation paid to you (if a Trainer) or paid by you (if an Enterprise Client) in the twelve (12) months preceding the claim.

11. Dispute Resolution & Governing Law

These Terms are governed by the laws of the State of California, without regard to conflict of law principles. Any disputes arising under these Terms shall first be addressed through good-faith negotiation. If unresolved within 30 days, disputes shall be submitted to binding arbitration administered by JAMS in San Francisco, CA. Each party bears its own costs. Class action waivers apply. Notwithstanding the foregoing, either party may seek injunctive relief in any court of competent jurisdiction for breach of confidentiality or intellectual property provisions.

12. Modifications & Termination

Opus Data may modify these Terms at any time by posting updated Terms on the Platform and notifying you via email or in-app notification. Continued use of the Platform after modifications constitutes acceptance. Either party may terminate the relationship at any time. Upon termination, all pending approved earnings will be paid out in the next regular payout cycle. Sections relating to confidentiality, intellectual property, limitation of liability, and dispute resolution survive termination.

13. Contact Information