Is micro1 a legit company?

micro1 is an AI training company that The Next Web reported on October 9, 2026 will spend $1bn over 12 months buying and licensing company data, with capital from Citi and Hercules Capital. It made a competing bid for Spirit Airlines' records. We couldn't confirm completed company data payouts, and micro1 doesn't publish exclusivity terms.

How much does micro1 pay per hour?

That question is about micro1's contractor work (its "Expert opportunities"), not the company data program, and we didn't check contractor rates for this article. For companies, micro1 lists data partnership tiers of $100K-500K, $500K-2M and $2M+, and says payments typically range from $150K to over $2M.

Can I get paid for training AI?

There are two routes at micro1. Individuals can apply for contractor work, and companies can license operational data through the Enterprise Data Partnership. Anyone 18 or older can also refer a company and earn $50,000 for deals up to $700,000, or 7.5% of larger deals, once the company's data is transferred and approved.

micro1 Data Partnership Review 2026: Data, Controls and Terms

What micro1's Enterprise Data Partnership takes, how it handles PII and retention, and which terms it doesn't publish, checked October 25, 2026

micro1’s Enterprise Data Partnership licenses historical operational data (internal communications, SOPs, project histories, CRM records and support tickets) from companies with 30+ full-time employees. micro1 says it removes PII “when relevant,” keeps data in isolated pipelines for agreed periods and may apply synthetic rewrites. Stated company pay starts at $100K, and micro1 says it pays within 30 days of invoicing once it approves the data, per its pages read October 25, 2026.

Onfolio may earn a referral fee if you sign up to Polyshares through our links. This does not affect our editorial assessment.

micro1 is an AI training company that, per The Next Web (October 9, 2026), uses de-identified company data to build reinforcement learning environments and will spend $1bn over the next 12 months buying and licensing operational data, with Citi and Hercules Capital providing the capital. Most “is micro1 legit” searches are about its contractor jobs and AI interviews. This review covers the company data program from an ops point of view; for other programs, see companies that buy business data for AI.

What micro1 takes

micro1’s FAQ says the data “can include internal communications, SOPs, project histories, CRM records, support tickets, and other business documentation,” and that it wants material that reflects “how work is actually performed.” Its page gives industry examples (software and engineering, finance, customer support, sales and CRM, legal, operations) but doesn’t name specific software systems.

Data type	Named by micro1	What micro1 says is stripped or excluded
SOPs, knowledge bases, process documentation	Yes	PII “removed when relevant”
CRM records	Yes	Same; contact-level handling not described
Project histories, QA processes and support tickets	Yes	Same
Internal communications	Yes, no systems named	Same; micro1’s CRO told Business Insider Slack and email records matter (see below)
Decision-making processes	Yes, as part of the most valuable datasets	Not published
Excluded categories	Not published	Agreed per partnership; micro1 identifies “licensable datasets” during discovery and pricing

The page names internal communications without naming systems, and micro1’s chief revenue officer, William Almeida, told Business Insider (October 13, 2026) that developers want “long-horizon” data. He said that if a company won’t sell its Slack, CRM or email records, “the rest of the corpus is less valuable.” Expect scoping conversations to reach into chat, CRM and mail.

One public example: the agency MyPortal Marketing licensed Figma and Canva files, client onboarding checklists and Google Ads material to micro1, and said it didn’t sell client-specific material (Business Insider, October 13, 2026). That split, firm-made assets in and client material out, is the pattern the licensing section below follows.

How the data gets out of your systems

micro1 says you transfer data “through our APIs or other approved methods,” but it doesn’t say whether that means connectors, OAuth access or exports your team produces. Its FAQ says you don’t need to anonymize before transfer because “micro1 has processes for de-identifying and transforming sensitive information.” That implies raw data leaves your systems and de-identification happens on micro1’s side; confirm it in writing. The published process has four stages.

Discovery and pricing. micro1 identifies licensable datasets and tailors an offer “based on the nature, scale, and value of your data.” Companies that arrive through a referral go through “an initial AI interview.”
Contracting. Licensing terms, which micro1 says include “a commitment not to resell or distribute your data until personal and identifying information has been anonymized.”
Delivery and acceptance. You transfer the data and micro1 reviews it for completeness.
Payment. micro1 has 30 days to review and approve the data, then pays within 30 days of your invoice.

Before participation, micro1 says the two sides confirm four controls: U.S. and allied buyers only, data segregation, de-identification before sharing with third parties, and defined retention periods. Contracting is the point where an ops lead should put the extraction method, the de-identification location and the excluded systems into the agreement, because none of them are fixed on the public page.

micro1 says companies can review representative samples before prepared data is used. It also says “certain datasets may undergo synthetic rewrites” that preserve workflow structure while reducing ties to the original records.

Privacy and security controls

These are micro1’s own claims from its data partnership page and FAQ, read October 25, 2026.

Control	What micro1 says	Ask for in writing
PII removal	PII “is removed when relevant”	A definition of “relevant,” the fields removed and where the removal runs
Access	Restricted to a limited number of authorized team members	Named roles with access and access logs
Isolation	Isolated pipelines that prevent commingling with other engagements	Architecture summary or security questionnaire
Pre-transfer anonymization	Not required; micro1 de-identifies after transfer	Where that processing runs and who can see raw data
Retention	Data kept only for agreed periods; raw data “deleted according to an agreed schedule”	The retention period in days or months
Sample review	Companies can review representative samples before use	A sign-off right on samples before any use
Deletion	On request or at the end of an engagement, “according to the terms of the engagement”	A deletion certificate and what happens to models already trained
Synthetic rewrites	Some datasets may be rewritten	Which datasets, and whether rewritten versions count as your data
Ownership	Companies retain ownership of underlying data	The license scope and any rights over derived datasets
Security certifications	Not published	A SOC 2 report or equivalent
End buyers and sub-processors	Not named; buyers limited to labs and enterprises “based in the U.S. and allied countries”	Named end buyers and a sub-processor list

A Pebblous blog analysis (October 2026, secondhand) notes that micro1 hasn’t published end-buyer names, exclusivity wording or liability and indemnity clauses. Those match what we couldn’t find on micro1’s page on October 25, 2026.

Commercial terms in brief

Read October 25, 2026, on micro1’s data partnership and referral pages.

Term	What micro1 says
Stated pay	$100K-500K (qualified partnerships), $500K-2M (large datasets or multiple teams), $2M+ (highly unique proprietary data); payments “typically ranging from $150K to over $2M”
Pay basis	Dataset size and volume, workflow complexity, data quality, domain expertise and uniqueness; each partnership evaluated individually
Payout estimator	Pebblous (secondhand) reports an illustrative $50,284 to $92,776, below the lowest $100K tier
Payment timing	30 days to review and approve after transfer, then paid within 30 days of invoicing
Exclusivity	Not published
Fee to seller	Not published
Eligibility	30+ full-time employees, mature operations, substantial historical data; US first, then other Western markets; primarily English; no fixed minimum data volume

micro1 also runs a referral program: anyone 18+ can earn $50,000 for deals up to $700,000, or 7.5% of the deal value above that, with no limit on referrals. The referral bonus enters payout approval once the referred company’s data has been transferred and approved.

What data you can and can’t license

micro1 doesn’t publish a list of excluded data, so the boundary is set in your agreement. The MyPortal example shows one workable split for an agency: firm-made assets and process documents in, client-specific material out.

Usually firm-owned and licensable once scrubbed	Usually not yours to license, or high risk
SOPs, QA checklists and internal knowledge bases	Client deliverables and documents under NDA
Your own onboarding checklists and templates	Client ad accounts and client ESP lists
Internal project histories with client names removed	Client Slack Connect channels
Your own CRM pipeline and sales process records, with contacts stripped	Contact-level CRM personal data
Design files you made for your own brand	HR systems, compensation, health records and call recordings

The rules behind the right-hand column apply whatever micro1’s scoping says. CCPA/CPRA has covered employee and B2B contact data since January 1, 2023, and GDPR covers EU and UK people.

NLRA Section 7 protects employee discussions of pay and conditions, which often sit in Slack. Slack’s API terms (August 2025) bar bulk export and LLM training via the API, per Hunton, and Google Workspace API policies restrict training non-personalized models. Recorded calls carry state consent rules, and the FTC said in February 2024 that retroactively changing privacy terms to allow AI training can be deceptive.

Risks for the ops team

“When relevant” is undefined. PII removal isn’t stated as universal on the page. Get the de-identification standard in the contract.
Scope creep. micro1’s CRO says Slack, CRM and email make the corpus more valuable. Decide your excluded systems before the discovery call.
Client confidentiality. Many 2026 agency contracts carry a “no client data used to train AI” warranty (RICHT law firm, October 18, 2026).
Unpublished terms. Exclusivity, end buyers and liability terms aren’t public, so your team can’t compare them without a draft agreement. Published price evidence across buyers is collected in how much company data is worth to AI.
Referral incentives. Pebblous (secondhand) points out that referral payouts can create pressure to move inside a company before legal review is finished.
Deletion vs trained models. micro1’s deletion runs “according to the terms of the engagement.” Deleting data doesn’t remove it from models already trained.
Reputation. micro1 bid for Spirit Airlines’ records, a sale that drew a union objection and a letter from 100+ members of Congress to Google and Spirit (Business Insider, October 13, 2026).
Short deadlines. Business Insider reported that 48-hour “exploding offers” are common across this market; it didn’t attribute them to micro1.

Pre-export checklist

Inventory the documentation, CRM and project systems you’d consider licensing, with years of history for each.
Mark every folder, channel and record set that belongs to or concerns a client.
Write your excluded systems list (HR, payroll, call recordings, client workspaces) before the discovery call.
Request micro1’s de-identification standard and where it runs, in writing.
Ask whether your data will be synthetically rewritten and who owns the rewritten version.
Get the retention period, deletion process and trained-model position into the agreement.
Ask for the acceptance criteria, exclusivity terms, end buyers and liability clauses before signing.
Confirm your export method complies with Slack and Google Workspace terms.

Deciding whether to sell at all? onfolio.com’s micro1 data partnership review covers it from the owner’s side: whether it’s legit, what you might get and what can go wrong.

Sources

micro1 data partnership page, read October 21 and 25, 2026: https://www.micro1.ai/data-partnerships
micro1 company referral page, read October 21 and 25, 2026: https://www.micro1.ai/company-referral
The Next Web, “micro1 commits $1bn to buy company data for training AI agents,” October 9, 2026: https://thenextweb.com/news/micro1-1bn-enterprise-data-ai-training-citi-hercules
Business Insider, “AI labs are fueling a wild market for your Slack DMs,” October 13, 2026: https://www.businessinsider.com/ai-training-employee-data-emails-slack-dms-2026-10
Pebblous blog on micro1’s company data partnerships (secondhand), October 2026: https://blog.pebblous.ai/blog/micro1-company-data-partnerships-rights/en/
RICHT law firm on agency client data, October 18, 2026
Hunton on Slack API terms, August 2025
FTC guidance on retroactive privacy policy changes, February 2024

micro1’s data partnership page

FAQ

Is micro1 a legit company?: micro1 is an AI training company that The Next Web reported on October 9, 2026 will spend $1bn over 12 months buying and licensing company data, with capital from Citi and Hercules Capital. It made a competing bid for Spirit Airlines' records. We couldn't confirm completed company data payouts, and micro1 doesn't publish exclusivity terms.
What is the micro1 data partnership?: It's micro1's program for licensing operational data from companies, called the Enterprise Data Partnership on its page. It wants historical operational data, such as internal communications, SOPs, project histories, CRM records and support tickets, from companies with 30+ full-time employees. micro1 says it uses de-identified data to build reinforcement learning environments, per The Next Web.
How much does micro1 pay per hour?: That question is about micro1's contractor work (its "Expert opportunities"), not the company data program, and we didn't check contractor rates for this article. For companies, micro1 lists data partnership tiers of $100K-500K, $500K-2M and $2M+, and says payments typically range from $150K to over $2M.
Can I get paid for training AI?: There are two routes at micro1. Individuals can apply for contractor work, and companies can license operational data through the Enterprise Data Partnership. Anyone 18 or older can also refer a company and earn $50,000 for deals up to $700,000, or 7.5% of larger deals, once the company's data is transferred and approved.