DataPacts

Shfa.ai: Medical Data Buyer Profile, Program Terms, Payouts and Reputation

Checked

At a glance

Shfa.ai is a medical data intermediary that de-identifies records held by doctors, clinics and hospitals and connects them to AI licensing deals.

Independent reference. Every buyer is covered on the same terms, every figure carries its source and check date, and no buyer pays to be listed or placed higher.

Request an independent introduction

Tell us what data your company holds. We reply with the buyers that fit and what their terms leave out. Your details reach a buyer only after you agree to a named introduction.

IndependentYour details stay with us until you agreeNot selling is a valid outcome

It states that medical facilities "have been paid $300,000 to $3,000,000", while its own footer calls such figures "representative and potential outcomes only" (checked 2026-11-09). No legal entity, address or governing law appears on its pages.

Every figure on this page is Shfa.ai's own statement with its check date, in the same format as every profile in a reference on AI training data licensing that covers each buyer and intermediary by one method.

The page follows how DataPacts sources and dates every figure: facts come from Shfa.ai's own pages and from the rules it cites, and anything not found is marked as not stated.

What Shfa.ai Does for Medical Providers: De-identification, Then Licensing Introductions

Shfa.ai works on the provider's side: it asks medical providers with years of records for EHR exports, FHIR data, clinical notes and Apple Health data, de-identifies them, and connects the provider to licensing deals that the provider approves.

The record types Shfa.ai names, as checked 2026-11-09:

  • EHR exports
  • FHIR data
  • Clinical notes
  • Apple Health data

Shfa.ai does not state that it buys or takes ownership of records, and no AI lab or other licensee is named in the sources checked. Eligibility is stated as medical providers with years of records.

Who licenses clinical records, at what prices and under which models is covered under medical and clinical data for AI training, which also treats Shfa.ai as an intermediary that handles records and does not buy them.

Shfa.ai Company Identity

Shfa.ai publishes a brand and a website but no legal entity: its home, privacy and terms pages name the operator only as "Shfa.ai" and give no address or governing law, checked 2026-11-09.

Attribute Value Source
Brand Shfa.ai shfa.ai
Website shfa.ai shfa.ai
Legal entity Not published on the home, privacy or terms pages shfa.ai/terms
Address and governing law Not published shfa.ai/terms
Operator wording Privacy page: "We are the controller"; footer: "(c) 2026 Shfa.ai" shfa.ai/privacy
Registry number, country, founding date Not recorded n/a
Role Intermediary: de-identifies records and connects providers to licensing deals shfa.ai
People Not recorded in the sources checked n/a

The brand and website rest on the home page. "We are the controller" and the footer line come from the privacy page. The legal entity name and form, registry number, address, country, governing law, founding date and people are not public in the sources checked, so none is stated here.

Shfa.ai Medical Data Licensing Terms, Dated

Shfa.ai publishes its data wanted, eligibility, process, a payout range and a referral rate, and leaves every license term to the deal; the table lists each term with what Shfa.ai states and the check date, as of 2026-11-09.

Term What Shfa.ai states Checked Source
Program Medical data licensing 2026-11-09 shfa.ai
Data wanted EHR exports, FHIR, clinical notes, Apple Health 2026-11-09 shfa.ai
Eligibility Medical providers with years of records 2026-11-09 shfa.ai
Process Connect in the client's environment; de-identify and red-team; ship with evidence 2026-11-09 shfa.ai
De-identification methods HIPAA Safe Harbor and Expert Determination 2026-11-09 shfa.ai
License terms The client approves every deal and sets terms 2026-11-09 shfa.ai
Payout "Medical facilities have been paid $300,000 to $3,000,000"; footer: "Financial figures describe representative and potential outcomes only" 2026-11-09 (footer re-read 2026-11-16) shfa.ai
Referral 5 % of the deal when a referred doctor, clinic or hospital completes a sale; a link alone earns nothing 2026-11-09 shfa.ai/referral
Licensees Not named 2026-11-09 shfa.ai
Contracting entity Not published 2026-11-09 shfa.ai

Shfa.ai states two terms that favor the provider: the client approves every deal and sets its terms, and the work happens in the client's environment. Scope, exclusivity, deletion, payment trigger and payment timing are not published in the sources checked, so the table cannot be compared row by row with programs that publish license terms.

Set beside the dated rows of program terms compared, the missing license terms are the main gap, and a provider asks for them in writing before any call about records.

What Shfa.ai States Medical Facilities Have Been Paid

Shfa.ai states that "Medical facilities have been paid $300,000 to $3,000,000" for their data, according to its home page checked 2026-11-09, and the same page's footer says "Financial figures describe representative and potential outcomes only".

The range is Shfa.ai's statement, not a confirmed payment by a named party. The two sentences sit on the same site and pull in different directions: "have been paid" describes past payments, while the footer describes representative and potential outcomes. Both are shown here, and neither is averaged or rounded.

No number of facilities, dates, buyers, data volumes or payment timing accompany the range. Across the market, advertised and confirmed figures are kept apart, as set out under reported payouts to data sellers.

Shfa.ai's De-identification and HIPAA Claims, Set Against 45 CFR 164.514

Shfa.ai states that it de-identifies medical facility data under both HIPAA methods, Safe Harbor and Expert Determination, working inside the client's own environment, according to its pages checked 2026-11-09.

Shfa.ai also states that it red-teams the output and ships it with evidence. These are Shfa.ai's claims, not findings by a regulator or court.

Safe Harbor removes the 18 identifier types listed in 45 CFR 164.514(b)(2), and a 3-digit ZIP code is allowed only where the area has more than 20,000 people. Expert Determination rests on an expert's determination under 164.514(b)(1).

Data de-identified under 164.514(a)-(b) is outside the Privacy Rule (164.502(d)(2)), with eCFR Title 45 up to date as of 2026-11-05; the full rules are set out under HIPAA de-identification, sale of PHI and limited data sets.

The claims do not settle what the provider owes. A sale of PHI that is not de-identified needs patient authorization (164.502(a)(5)(ii)), and the authorization must state that the disclosure results in remuneration (164.508(a)(4)).

Whether Shfa.ai signs a business associate agreement is not in the sources checked, and both that and the qualifications of the expert who signs a determination are questions for the provider to ask.

Before connecting any system, a provider asks which method is used per dataset, who signs the determination, what evidence is delivered and where raw data stays. The seller step is covered under de-identification before licensing data.

Shfa.ai Referral Terms: 5 % of the Deal

Shfa.ai states that a person who refers a doctor, clinic or hospital receives 5 % of the deal when the referred provider's sale is finished, and that a referral link alone earns nothing (checked 2026-11-09).

No cap, payment date or contracting entity is stated for the referral. The trigger means a referrer is paid only on a finished sale, not on an introduction or a booked call. Referral terms across buyers are compared among the referral programs of AI data buyers, and this page carries no referral link.

Shfa.ai Reputation: The Record as of 2026-11-09

No court case, regulator action, breach notice or complaint involving Shfa.ai appears in a primary or named-person source as of 2026-11-09.

The one documented outside statement is a report by the CEO of SimpleClosure, dated 2026-10-01, that SHFA.ai ads promised founders USD 1 million, 3 million and 5 million. SimpleClosure is a competing intermediary, so the report is attributed to its CEO and not adopted here.

Set beside Shfa.ai's own page checked 2026-11-09, the report describes ads aimed at founders with larger sums than the USD 300,000 to 3,000,000 range on the home page.

No deal or payment is confirmed by any primary source, and forum or video results are not evidence. The general checks for an unsolicited offer are set out under whether a data licensing email is legitimate.

Benefits, Risks, Downsides and Alternatives for a Medical Provider Considering Shfa.ai

For a medical provider, Shfa.ai offers de-identification in its own environment and a veto over every deal, against an unnamed counterparty and payout figures the site calls potential, so the practical alternatives are programs that publish their terms or entity, and the option of not licensing at all.

Benefits (as Shfa.ai states them, checked 2026-11-09)

  • The client approves every deal and sets terms.
  • De-identification runs inside the client's environment, with evidence delivered.
  • Both HIPAA de-identification methods are named.
  • Because every deal needs the client's approval, the provider can still decline after de-identification.

Risks

  • No legal entity, address or governing law is published, so the provider does not know whom it contracts with before signing.
  • The payout figures are ones the site itself calls "potential outcomes only".
  • No licensee is named, so the end use and the AI developer are unknown in advance.
  • Business associate status is not on record.

Downsides

  • No license terms (scope, exclusivity, deletion, payment timing) are published to compare; they are not stated in the sources checked.
  • Records with patient identifiers stay subject to HIPAA until they are de-identified.

Alternatives (alphabetical, one claim each, checked 2026-11-09)

  • Gradient Health data partners: a recurring revenue share each time data is used, with a "typical" 20 % partner share in Gradient's own worked example.
  • Labelbox Prism: requires IRB approval or qualified de-identification for medical data and, checked 2026-11-09, states "Smaller, more targeted datasets typically start around $50K".
  • Mercor Data Monetization: lists healthcare among its target industries and states that it covers extraction, anonymization and transfer costs.
  • Protege: states it "provides revenue share payouts to data partners with each use", percentage not public.
  • Not licensing at all.

The case for keeping records is set out under alternatives to selling data, including keeping it.

Shfa.ai Among Company Data Partnership Programs and Health Data Routes

Shfa.ai is one of the company data programs profiled on this site and one of several routes for health records, most of which, like Shfa.ai, handle records for the holder rather than buy them. All of them sit among the company data partnership programs.

More About Shfa.ai

Frequently Asked Questions About Shfa.ai

Does Shfa.ai buy medical data?

No purchase is on record: Shfa.ai de-identifies records for medical providers and connects them to licensing deals that the provider approves. The provider sets the terms (checked 2026-11-09).

How much does Shfa.ai say medical facilities are paid?

Shfa.ai states that medical facilities "have been paid $300,000 to $3,000,000". Its footer says such figures describe "representative and potential outcomes only" (checked 2026-11-09).

Is Shfa.ai HIPAA compliant?

Shfa.ai states that it de-identifies data under HIPAA Safe Harbor and Expert Determination (45 CFR 164.514(b)). No regulator or court finding about Shfa.ai is on record as of 2026-11-09.

Not published: Shfa.ai's home, privacy and terms pages name no legal entity, address or governing law. They were checked 2026-11-09, and a provider asks for the entity before signing.

Does Shfa.ai pay referral fees?

Shfa.ai states 5 % of the deal when a referred doctor, clinic or hospital completes a sale. A link alone earns nothing (checked 2026-11-09).