Flora logo

Flora

flora.ai
Moderate Risk
Updated August 25, 2026

Flora is a generative AI creative workspace (image/video/audio generation) aimed at professional creatives and marketing/advertising teams. The consumer Terms and Privacy Notice are recent (November 2025 and June 2026) and notably user-favorable on the key AI questions: users own both their Inputs and Outputs, and Flora explicitly commits not to train its own or third-party AI systems on User Data. However, the consumer posture carries meaningful risk: broad Aggregated De-Identified Data reuse, a binding arbitration clause and class-action waiver, a $100/12-month liability cap, silence on concrete retention schedules and deletion SLAs, session replay technology, and 'sale' of Personal Data acknowledged under state law. Compliance certifications are largely absent from the consumer documents themselves (the SOC 2 claim appears only in enterprise marketing copy). Review carefully before using with sensitive or proprietary data on the consumer tier.

AI Transparency Facts

Independent analysis by TermsWatchdog · © 2026 TermsWatchdog

Input Data Ownership

Low Risk

Users retain ownership of everything they submit; Flora takes only a service-provision license. Inputs are expressly defined as User Data and remain the user's property, with the license limited to operating the Services.

Confidence
88%

Output Data Ownership

Low Risk

Flora assigns all rights in AI-generated Output to the user. The one caveat is that Output may not be unique and users have no rights to similar materials generated for other users, which is inherent to generative AI rather than a vendor land-grab.

Confidence
85%

Training Data Usage

Low Risk

The Terms explicitly state Flora does not train its own AI systems on User Data and does not permit third-party model providers to train on it either. This is an unusually clear no-training commitment for consumer terms, though Flora may still use Aggregated De-Identified Data for improving its products.

Confidence
85%

Data Retention & Deletion

Moderate Risk

The Privacy Notice describes retention only in open-ended, purpose-based terms with no fixed schedules or deletion SLAs. Deletion is possible by contacting support, but account termination may destroy Content and recovery is not guaranteed, so users bear the risk of data loss.

Confidence
70%

Third-Party Data Sharing

Moderate Risk

Flora discloses Personal Data to service providers (hosting, AI/ML providers, analytics, security), other users, and third-party services, and acknowledges some disclosures may constitute a 'sale' under state law. Sharing is disclosed and largely tied to service delivery, but the acknowledgment of a 'sale' and business-transfer clause warrant caution.

Confidence
72%

Opt-Out Rights

Moderate Risk

The policy provides some opt-out mechanisms: cookie/analytics opt-out, session replay opt-out, Google Analytics opt-out, and an arbitration opt-out within 30 days. However, there is no described mechanism to opt out of core data collection or the 'sale' of data beyond cookie settings, and DNT is not honored.

Confidence
68%

Compliance & Certifications

High Risk

The consumer Terms and Privacy Notice name no certifications and cite only GDPR lawful bases and COPPA/CCPA-style 'sale' concepts. No SOC 2, ISO, or attestation is claimed in the consumer documents themselves (the SOC 2 Type 1/2 claim appears only in enterprise marketing copy, and even there GDPR is described as forthcoming). Against the universal baseline plus marketing/adtech frameworks, the consumer documents are largely silent.

Confidence
65%

Model Explainability & Auditability

High Risk

The documents provide no transparency into model behavior, no explainability commitments, and no enterprise auditing capability. They only warn users that AI output may be inaccurate or biased and must be independently verified.

Confidence
60%

Security Practices & Breach History

Moderate Risk

The Privacy Notice describes generic 'appropriate physical, technical, organisational and administrative' safeguards but no specifics (encryption, pen testing, bug bounty) in the consumer documents. A Trust Center and Status page are referenced. No breach history is disclosed.

Confidence
62%

Enterprise vs. Consumer Risk Delta

Moderate Risk

The consumer Terms distinguish free vs. paid use (free users limited to internal purposes; certain content rights depend on plan) and the Privacy Notice notes B2B users' employer is the controller with Flora acting as processor. Enterprise-specific protections are not detailed in the consumer documents themselves, leaving the delta partly ambiguous.

Confidence
60%

Human Review of User Inputs

Moderate Risk

No explicit blanket right to have staff read prompts is asserted, but Flora reserves broad rights to remove Content and reserves access for support, security, debugging, and legal purposes. For B2B users, Flora may share account contents (including communications) with the customer organization.

Confidence
55%

Regulatory & Litigation Exposure

Moderate Risk

The Privacy Notice describes cooperation with government and law enforcement to meet legal requirements, and the Terms impose binding arbitration and a class-action waiver. No active litigation is disclosed. The arbitration/class-waiver structure limits user remedies.

Confidence
70%

PII & SPI Data Inventory

Moderate Risk

Flora collects a defined set of PII: name, email, username, payment/transaction data, device/IP data, web analytics, ZIP code, job title, and IP-based geolocation, with disclosed purposes and GDPR lawful bases. No sensitive categories are intentionally solicited, but users may voluntarily upload any content (free-form Inputs) that could include SPI, and precise geolocation is limited to IP-based data.

Confidence
72%

Policy–Product Currency

Low Risk

The Privacy Notice is dated June 1, 2026 and the Terms November 17, 2025 — both within 12 months of the August 2026 analysis date. The Terms substantively address the product's AI-first nature, including generative Output, Third-Party AI Systems, and training data, matching the product surface (multi-model image/video generation). Coverage of the visible AI capabilities is demonstrated.

Confidence
78%

Cross-Document Consistency

Low Risk

Two consumer documents were supplied (Terms of Service and Privacy Notice) and no contradictions were found between them. The no-training commitment, ownership terms, and data-sharing descriptions are consistent across both documents.

Confidence
65%

You've read all 15 risk ratings for Flora. Create a free account to see the exact policy wording behind each rating.