1. What outputs in FLORA can be commercially distributed?

Outputs created in FLORA may generally be commercially distributed, provided the user has the rights to any inputs they upload or reference, the output does not infringe third-party rights, and the use complies with the applicable model/provider terms and FLORA’s policies. FLORA does not claim ownership of customer inputs or outputs; however, commercial distribution is not permitted for outputs that unlawfully copy or closely reproduce protected third-party content, misuse a person’s likeness, trademark, or other rights, or are otherwise unlawful.

2. What risks are associated with importing media from social media, Google, or Pinterest?

Content available on the internet may be subject to third party intellectual property rights, and due to the nature of generative AI technologies, there is a risk that a generative AI model may produce output that could be deemed to infringe third party rights. Because you have full control over the content that you upload or input into Flora’s services, you are responsible for ensuring that you do not upload content for which you lack the necessary rights, and for ensuring appropriate use of any resulting output.  Some practical examples of risk levels include:

• Low risk

  • Moodboards

  • Composition reference

  • Color palette reference

  • Loose stylistic inspiration

• High risk

  • “Make this same image with minor changes”

  • “Extend or outpaint this copyrighted photo”

It is worth noting that FLORA has a built in integration with Unsplash, a repository for commercially licensed assets to reference.

Reference: Copyright and Artificial Intelligence, Part 2 Copyrightability Report

3. Do FLORA train on user data?

FLORA does not train any models in house. In FLORA, you can access a suite of models by various providers, and they each have training policies, which can be found here. FLORA pushes to make the full list of subprocessors compliant with our policy of not training on user data, and that is the case today as of Mar 26, 2026. Additionally, there is an in-app feature that allows workspace admins to disable models for the entire workspace as well as a setting to opt-in for new models, which allows customers to fully protect their data.

4. Do customers own the assets that they create in FLORA?

FLORA does not claim any ownership rights in the output. 

Ownership of outputs is determined by the third-party terms and is either allocated to the model provider or end user. As of Mar 26, 2026, none of our third party providers claim explicit ownership of outputs. You can find the breakdown of third-party terms for models on FLORA here.

5. Do you offer indemnification?

• FLORA agrees to indemnify the customer for third-party claims that the Service infringes third-party IP rights. 

• However this indemnification obligation does not extend to the output generated by Flora’s platform. Customer’s right to indemnification for outputs is subject to the terms of the third-party model that Customer used to generate the output. The customer's admin account can access and review all applicable third-party terms through Flora’s platform.

  • Specifically of highlight, Google offers the following indemnification:

    • 1. Training Data Indemnity: We indemnify our customers against third-party IP claims related to the data we used to train our generative AI models

    • 2. Generated Output Indemnity: We also indemnify customers against claims that the unmodified output they generate using our services infringes on a third party's IP rights.

    • It is important to note that this indemnity applies to responsible use of our services. The terms state that the protection does not cover situations where a customer "knew or should have known" that the output was likely to be infringing.

• FLORA does not have a blanket indemnification policy yet that indemnifies user’s outputs. FLORA is actively in discussions with our insurance broker to obtain this.

6. Are you SOC2 compliant?

We are SOC2 Type 1 compliant, and you can access this report in our Trust Portal. We are on track to finish SOC2 Type 2 by June 2026.

7. Do you have SSO?

FLORA provides standard SSO via Google and Microsoft for self-serve customers as well as Enterprise SSO/SAML for enterprise customers.

8. Can I limit the amount of credits each seat can use?

Yes, admins have the ability to set credit limits for any individual in their workspace.