---
title: "Recraft | Ethical Stack"
url: https://ethicalstack.ai/products/recraft/
catalog_version: f6baff2119d0e9158520da7fbe4232011dac79f25bd08757436d7a1dbd593db2
---

# Recraft

Hosted image-generation and design platform with raster/vector output, editing tools and an API.

Canonical profile: https://ethicalstack.ai/products/recraft/

Scope: Recraft proprietary image models through web/API; training disclosure in the California AB2013 statement applies to Recraft proprietary models offered to California users, not third-party models accessible through the platform.

Published overall: 3.2; 6/6 dimensions rated.

Round the equal-weight mean of available numerical ratings to one decimal; require at least four of six for an overall score. BDS status stays outside the mean; for eligible profiles, a boycott target caps the final score at 2/10 and a pressure target subtracts one point (minimum 1). Not listed and not checked do not adjust the score.

## Energy practices

Finding: 2/10

Assessment: https://ethicalstack.ai/products/recraft/#energy

Recraft is a hosted image service; no product-scoped energy report, facility electricity mix or documented compute-saving control appears in the cited sources.

Reason: 1 + 1 reporting + 0 electricity + 0 controls = 2/10. This grades documented practices and transparency.

Argument: The service’s web/API and third-party model routes are described, and terms note iterative features may consume variable credits. Credits are a billing unit, not an energy measurement; no facility mix or task-level workload footprint is disclosed.

Type: argument; confidence: medium

Scope (dimension): Recraft proprietary image service and API; third-party provider features have separate processing paths.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Serving region, model-specific energy, electricity sourcing and product-level physical resource controls remain unreported.

```json
{
  "method": "practices-transparency-v1",
  "scope": "Recraft proprietary image service and API; third-party provider features have separate processing paths.",
  "components": {
    "disclosure": {
      "score": 1,
      "reason": "Recraft describes its service and API as hosted generation, but does not publish request-level energy accounting.",
      "sourceUrls": [
        "https://www.recraft.ai/",
        "https://www.recraft.ai/legal/privacy"
      ]
    },
    "electricity": {
      "score": 0,
      "reason": "The reviewed product and legal pages do not map Recraft requests to a named serving facility or electricity mix.",
      "sourceUrls": [
        "https://www.recraft.ai/",
        "https://www.recraft.ai/legal/privacy"
      ]
    },
    "demand": {
      "score": 0,
      "reason": "The service terms discuss variable iterative credit use but do not document a physical resource-control setting or measured reduction.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/terms"
      ]
    }
  },
  "actualImpactMeasured": false
}
```

- [Recraft product overview](https://www.recraft.ai/) (primary; checked not recorded)

- [Recraft privacy policy](https://www.recraft.ai/legal/privacy) (primary; checked not recorded)

- [Recraft terms of service](https://www.recraft.ai/legal/terms) (primary; checked not recorded)

## Training data

Finding: 3/10

Assessment: https://ethicalstack.ai/products/recraft/#training-rights

Recraft's California notice names broad data categories and says the corpus mixes licensed, fair-use, public-domain and uncopyrightable material; no named source inventory or source-creator remedy is disclosed.

Reason: 1 + 1 source transparency + 1 permission + 0 creator controls = 3/10, with the permission cap applied.

Argument: The AB2013 statement covers Recraft proprietary image models offered in California and describes publicly available, third-party/user, and internally generated data in a corpus of billions of images. Its privacy opt-out concerns Recraft users' submitted content; the reviewed materials do not show an opt-out or removal process for creators whose works are in training sources.

Type: argument; confidence: medium

Scope (profile): Recraft proprietary image models through web/API; training disclosure in the California AB2013 statement applies to Recraft proprietary models offered to California users, not third-party models accessible through the platform.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Named-source inventory, source-level permission basis, creator opt-out/removal/retraining path, and scope outside the California notice are not established by the reviewed sources.

```json
{
  "method": "training-rights-evidence-v1",
  "scope": "The stated training release or selected model",
  "components": {
    "provenance": {
      "score": 1,
      "reason": "The California AB2013 statement lists publicly available, third-party/user-provided and internally generated data categories, including billions of images, but not named datasets.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/ab2013"
      ]
    },
    "permissions": {
      "score": 1,
      "reason": "Recraft reports a mixture that includes licensed, public-domain and fair-use material, without a per-source permission ledger.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/ab2013"
      ]
    },
    "creatorControls": {
      "score": 0,
      "reason": "The reviewed AB2013 and privacy disclosures specify controls for Recraft users' submitted content, not an opt-out, removal request, or retraining remedy for creators of source-corpus works.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/ab2013"
      ]
    }
  },
  "routeDependency": null,
  "modelReference": null
}
```

- [Recraft training data disclosure](https://www.recraft.ai/legal/ab2013) (primary; checked not recorded)

## Your data & privacy

Finding: 4/10

Assessment: https://ethicalstack.ai/products/recraft/#privacy

Recraft’s new policy says individual-account input may be used to train Recraft models by default, with an account opt-out; API and Group Workspace inputs are excluded.

Reason: 4/10 reflects a documented opt-out and a no-training API/Group route, balanced against individual-account training by default and account/gallery retention.

Argument: The privacy policy effective for new accounts from 20 October 2026 and existing accounts from 3 November 2026 distinguishes consumer and organizational routes: individual AI input/output can train proprietary models unless opted out; API and Group Workspace content is excluded. External providers for third-party image/video and chat/voice features are subject to zero-retention or equivalent no-training terms.

Type: argument; confidence: medium

Scope (profile): Recraft proprietary image models through web/API; training disclosure in the California AB2013 statement applies to Recraft proprietary models offered to California users, not third-party models accessible through the platform.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Policy effective date depends on account creation; inputs/assets remain in accounts until deletion, and Free Tier assets in the community gallery may remain visible afterward.

- [Recraft privacy policy](https://www.recraft.ai/legal/privacy) (primary; checked not recorded)

## Openness

Finding: 2/10

Assessment: https://ethicalstack.ai/products/recraft/#openness

The web service is closed and plan-specific; free outputs have narrower rights than paid-subscription assets.

Reason: 2/10 reflects a managed proprietary service with limited tier-dependent output rights and no reviewed open-weight deployment path.

Argument: Current terms say Free Tier assets are owned by Recraft and licensed for personal use only, while paid users own their generated assets subject to restrictions. The reviewed materials do not establish downloadable open weights or self-hosted deployment for the proprietary Recraft models.

Type: argument; confidence: medium

Scope (profile): Recraft proprietary image models through web/API; training disclosure in the California AB2013 statement applies to Recraft proprietary models offered to California users, not third-party models accessible through the platform.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Terms and ownership differ between Free Tier, paid, workspace and API routes; feature and price availability may vary.

- [Recraft terms of service](https://www.recraft.ai/legal/terms) (primary; checked not recorded)

## Worker protections

Finding: 1/10

Assessment: https://ethicalstack.ai/products/recraft/#labour

Recraft’s current privacy and terms explain account processing, model training settings, credits and content rights, but provide no workforce conditions or worker-remedy measures.

Reason: 1 + 0 + 0 + 0 = 1/10. Documented worker protections; missing disclosure is an evidence gap.

Argument: The reviewed Recraft privacy notice, AB2013 training disclosure and service terms discuss user data and product operations; they do not report employee or contractor pay, conditions, representation or independent labour assurance. This narrow public-evidence gap is not a finding of mistreatment.

Type: argument; confidence: medium

Scope (profile): Recraft proprietary image models through web/API; training disclosure in the California AB2013 statement applies to Recraft proprietary models offered to California users, not third-party models accessible through the platform.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Employee, contractor, annotation and moderation workforce conditions for Recraft remain unestablished in these public sources.

```json
{
  "method": "worker-safeguards-evidence-v1",
  "entity": "products/recraft",
  "company": "",
  "scope": "The named developer or operator",
  "components": {
    "conditions": {
      "score": 0,
      "reason": "Reviewed Recraft sources do not disclose worker pay, employment conditions or contractor protections.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/privacy",
        "https://www.recraft.ai/legal/ab2013",
        "https://www.recraft.ai/legal/terms"
      ]
    },
    "voice": {
      "score": 0,
      "reason": "The cited Recraft materials do not identify worker-voice, grievance or remedy processes.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/privacy",
        "https://www.recraft.ai/legal/ab2013",
        "https://www.recraft.ai/legal/terms"
      ]
    },
    "assurance": {
      "score": 0,
      "reason": "No independent workforce assurance is published in the cited Recraft sources.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/privacy",
        "https://www.recraft.ai/legal/ab2013",
        "https://www.recraft.ai/legal/terms"
      ]
    }
  },
  "disclosureLimited": true
}
```

- [Recraft privacy policy](https://www.recraft.ai/legal/privacy) (primary; checked not recorded)

- [Recraft training data disclosure](https://www.recraft.ai/legal/ab2013) (primary; checked not recorded)

- [Recraft terms of service](https://www.recraft.ai/legal/terms) (primary; checked not recorded)

## Military ties

Finding: 7/10

Assessment: https://ethicalstack.ai/products/recraft/#military-contracts

No military relationship or product-specific military safeguards are established in the reviewed Recraft sources.

Reason: 7 base + 0 safeguard adjustment = 7/10 (0 of 9 safeguard points; every 3 adds 1). No military relationship was established in the cited sources. This ranks above known defense ties without claiming a verified clean record.

Argument: The official sources reviewed for Recraft describe its creative route, model access and user terms but do not establish defense work, a military customer, or binding operational restrictions. This product-level disclosure gap does not resolve unreviewed corporate or downstream relationships.

Type: argument; confidence: medium

Scope (profile): Recraft proprietary image models through web/API; training disclosure in the California AB2013 statement applies to Recraft proprietary models offered to California users, not third-party models accessible through the platform.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: Company-wide defense customers and downstream uses of Recraft were not exhaustively established.

```json
{
  "method": "military-ties-priority-v2",
  "entity": "products/recraft",
  "company": "",
  "scope": "The named developer or operator",
  "components": {
    "restrictions": {
      "score": 0,
      "reason": "Reviewed Recraft sources do not state a binding military-use restriction for this product route.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/privacy",
        "https://www.recraft.ai/legal/ab2013",
        "https://www.recraft.ai/legal/terms"
      ]
    },
    "oversight": {
      "score": 0,
      "reason": "No military-specific oversight or enforcement is documented in the cited Recraft sources.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/privacy",
        "https://www.recraft.ai/legal/ab2013",
        "https://www.recraft.ai/legal/terms"
      ]
    },
    "transparency": {
      "score": 0,
      "reason": "The reviewed Recraft sources do not disclose a military relationship for this product or operator.",
      "sourceUrls": [
        "https://www.recraft.ai/legal/privacy",
        "https://www.recraft.ai/legal/ab2013",
        "https://www.recraft.ai/legal/terms"
      ]
    }
  },
  "disclosureLimited": true,
  "relationship": "not-established",
  "relationshipReason": "The cited Recraft product and policy materials do not establish a military contract, customer or deployment; that limited finding is not proof no such relationship exists.",
  "score": 7,
  "baseScore": 7,
  "safeguardPoints": 0,
  "safeguardAdjustment": 0,
  "rawScore": 7,
  "cap": null
}
```

- [Recraft privacy policy](https://www.recraft.ai/legal/privacy) (primary; checked not recorded)

- [Recraft training data disclosure](https://www.recraft.ai/legal/ab2013) (primary; checked not recorded)

- [Recraft terms of service](https://www.recraft.ai/legal/terms) (primary; checked not recorded)

## BDS targets

Finding: Not listed

Assessment: https://ethicalstack.ai/products/recraft/#israel-palestine

The company is not named in the two reviewed official BNC campaign sources.

Reason: Campaign category, excluded from numerical scoring.

Argument: This entry records only whether the named company appears in the BNC consumer guide or No Tech campaign. Not listed is a limited source check, not a finding of endorsement, compliance or absence of supplier and investment relationships.

Type: argument; confidence: medium

Scope (profile): Recraft proprietary image models through web/API; training disclosure in the California AB2013 statement applies to Recraft proprietary models offered to California users, not third-party models accessible through the platform.

Research reviewed: not recorded; evidence date: not recorded

Unknowns: The check does not cover every campaign, subsidiary, investor, supplier, local listing or future target change.

- [BNC consumer boycott guide](https://bdsmovement.net/Guide-to-BDS-Boycott) (primary; checked not recorded)

- [BNC No Tech campaign](https://bdsmovement.net/no-tech-oppression-apartheid-or-genocide) (primary; checked not recorded)

## Reading the evidence

- Assessments are provisional editorial judgments. Cite the profile and the underlying sources, and retain confidence, scope and unknowns.

- Unknown, not-applicable and model-dependent are different states. Missing evidence does not establish good or bad conduct.

- Company-level evidence does not establish use of a particular product or model. Account plan, model, host and route can change an assessment.

- Energy scores describe documented practices and transparency, not measured request emissions. Zero component points can mean an evidence gap.

- Published overall scores require at least four of six numerical dimensions. Preserve coverage and BDS adjustments; BDS itself is categorical.

- No named BDS target found is limited to the reviewed campaign sources and is not approval. Personalized priorities do not change the published overall score.

- Review and source-check dates describe the research, not this deployment. Fetch current primary sources when the decision requires current verification.

Methodology: https://ethicalstack.ai/methodology/