Is DataAnnotation Legit? What It Pays and What to Watch For

A man with headphones approves an answer card while a friendly robot assistant holds a shield

Yes, DataAnnotation is legitimate. It’s run by Surge Labs, better known as Surge AI, a large San Francisco company that sells AI training data, and it pays through PayPal. The catches: you get one try at the starter assessment and often no reply, your access can end at the company’s discretion, and a lawsuit over whether its workers should really be contractors is still open. Here’s what we found on its own pages, in court records and in news reports.

DataAnnotation at a glance

Run bySurge Labs, Inc. (Surge AI), named in DataAnnotation’s own terms
What you doRate and compare chatbot answers, flag errors, write hard prompts, review code
PayGeneral projects from $25 to $50+ an hour; coding from $40 to $150+ an hour (its FAQ)
PaidBy PayPal, in US dollars, a few days after you request a payout
To joinOne starter assessment (about an hour) and an ID check
Cost to joinNone. It says it will never ask you for money
Trustpilot3.9 from 1,975 reviews (read October 18, 2026)

Who runs DataAnnotation

DataAnnotation doesn’t name a company on its homepage beyond a “by surge” logo, which is why so many people ask. Its Terms of Use (updated July 14, 2026) cover data labeling services for Surge Labs, Inc., and Surge AI’s own site says it manages applications and projects through DataAnnotation. In 2023 The Verge reported that the two appeared to be the same company, when Surge’s CEO would neither confirm nor deny it. The terms now settle it.

Surge was founded in 2020 by Edwin Chen and is based in San Francisco. Forbes reported $1.2 billion in revenue for 2024 and, in September 2026, that the company has no venture funding.

How DataAnnotation works

  • The starter assessment: most take about an hour, and you can only take it once.
  • The work: checking chatbot answers for accuracy, comparing AI outputs, flagging factual errors, testing AI images, writing hard prompts and reviewing code.
  • Who they want: fluent English writers. Its FAQ asks for a bachelor’s degree or equivalent experience, though its generalist job page says no degree is required. Coders, STEM specialists, licensed professionals and native speakers of other languages get higher-paying projects.
  • Checks: an ID check through a company called Persona, with a photo ID and a selfie.
  • Where: it says it’s available in 50+ countries but publishes no list.
  • The terms: you’re an independent contractor, 18 or older, with no minimum hours.

What DataAnnotation pays, and how you get paid

Its own pages don’t agree on rates. The homepage says up to $75 to $125+ an hour. Job pages say up to $50 an hour for generalists, up to $150+ for software engineers and up to $40 for language roles. The FAQ gives $25 to $50+ for general projects in one answer and $25 to $30+ in another. Plan on the lower end until you see real projects in your dashboard.

  • Payouts: by PayPal, in US dollars. It says deposits arrive within a few days after you request them.
  • Taxes: you’re a contractor. It asks for a W-9 or W-8 tax form, and you handle your own taxes.
  • Rejected work: work it rejects, for any reason at its discretion, isn’t paid.
  • Fees: it says it charges no sign-up fees. PayPal’s own fees may apply.

Complaints and lawsuits

  • “I never heard back”: the most common complaint, and the FAQ confirms it: DataAnnotation can’t reply to every applicant, and silence likely means you’re still under review or weren’t accepted.
  • Losing access: its FAQ says poor or problematic work may cost you access, and its terms let it end your account for any reason or none. We found no reliable reporting on specific cases.
  • Wage lawsuit (open): Cavalier v. Surge Labs, filed May 20, 2025 in San Francisco Superior Court, claims workers are wrongly treated as contractors and aren’t paid for training time, Inc. reported. Surge’s CEO told Forbes the suit is without merit. We found no ruling or settlement. DataAnnotation’s current terms send disputes to private arbitration and waive class actions.
  • Non-payment: we found no reliable reporting of DataAnnotation failing to pay workers.

How to spot a fake DataAnnotation offer

  • DataAnnotation says it will never ask you for money: it pays you, not the other way around.
  • The only way to apply is its official sign-up flow on dataannotation.tech. No recruiter can “fast-track” you.
  • It warns you not to engage with direct messages from anyone outside its official channels.
  • Got a “DataAnnotation” job offer by text or a chat app? Paste it into our free job scam checker before you reply.

Should you sign up?

If you write well in English, or you have coding or subject expertise, DataAnnotation is a real platform from a large, established company, and its published rates are toward the high end for this work. Take the starter assessment seriously, since you only get one try, and don’t sit waiting for a reply: apply to two or three platforms at once. Compare it with Outlier, Mercor, Alignerr and others on our AI training jobs page.

Sources

Checked against DataAnnotation’s own pages, court records and news reports on October 18, 2026. Terms and pay change often, so check its site before you apply.

Related guides

Frequently asked questions

Is DataAnnotation a scam?

No. It is run by Surge Labs, Inc., which its own terms name, and it says it never charges workers. Anyone asking you to pay for a DataAnnotation job is not DataAnnotation.

How long does DataAnnotation take to respond?

Its FAQ says you usually hear within a few days, and that silence likely means your application is still under review or was not accepted. It cannot reply to every applicant.

Can I retake the DataAnnotation starter assessment?

No. Its FAQ says you can only take the starter assessment once. Most take about an hour.

How does DataAnnotation pay?

Through PayPal, in US dollars. It says deposits arrive within a few days after you request a payout.

Who owns DataAnnotation?

Surge Labs, Inc., also known as Surge AI, a San Francisco AI data company that Edwin Chen founded in 2020.

Scroll to Top