Privacy

This privacy policy sets out how Persona Enterprises Inc. ("Persona", "we", "us" or "our"), the company that operates Iodine, collects, processes and uses your Personal Information through your use of our Services. Persona Enterprises Inc. is responsible for your Personal Information and is located at 251 Little Falls Drive, Wilmington, New Castle County, Delaware 19808.

Definitions

DPO means Data Protection Officer, and is the person specifically appointed by us to deal with your enquiries with respect to this privacy policy and your Personal Information.

API Content means the text you send to our API (including through the console playground) and the audio we generate from it.

Request Metadata means the technical record of an API request, described in "What We Keep for Each Request" below. Request Metadata never includes API Content.

Personal Information means any identifying information about you. This includes, but is not limited to, the following: your name and email address, your organization's name, billing records, the session cookie used to keep you signed in, and Request Metadata.

Services means all services provided by Persona Enterprises Inc. to you under the Iodine name, including (but not limited to) this Website.

Website means this website and all websites, applications and communications that post a link to this Privacy Policy.

Zero Data Retention (ZDR)

Zero Data Retention is enabled by default for every account and every API key. You do not need to opt in, and it costs nothing extra. Under ZDR:

• Your text is not stored. Text you send is held in memory only for as long as it takes to generate the audio, then discarded. It is never written to disk, to a database or to our logs.
• Your audio is not stored. Generated audio is streamed straight back to you. We do not keep a copy, and responses are marked "no-store" so they are not cached along the way.
• Nobody reads or listens to your requests. Because API Content is never kept, there is nothing for our staff to review.

Turning ZDR Off

The owner of an account can turn Zero Data Retention off in the console. We ask them to confirm first, and record who turned it off and when. While ZDR is off:

• we keep the text and generated audio of every completed request, together with its Request Metadata, on our servers;
• we may use that content to train, fine-tune, evaluate and improve our models;
• access to it is limited to the staff who build and evaluate our models;
• in return, requests made while ZDR is off are billed at a discounted rate (currently 15% off your usual rate), shown in the console before you confirm.

Requests that fail or are cancelled are not kept. ZDR can be turned back on at any time and takes effect immediately for new requests. The account owner can also delete everything kept for their account from the console at any time; it is removed from our systems straight away, with one exception described under "Safety Screening" below. Content that has already been used to train a model cannot be extracted from that model, but it is not used in any training run that starts after you delete it. Content kept while ZDR was off is otherwise kept until you delete it or close your account.

If you turn ZDR off, you are responsible for having the right to share that content with us, including the consent of your own end users where the content is theirs. Turning ZDR off never affects cloned voices, which are not used for training under any setting.

Safety Screening of Kept Content

Content kept while ZDR is off is screened automatically for material that may sexualise minors or otherwise break the law. Anything the screening flags is kept out of all training and held for review by a person. Until that review is done, it is not removed when you delete your kept data. If the reviewer finds it harmless, it is handled like the rest of your kept data (and deleted if you asked us to delete it). If they confirm it, we keep it out of training permanently, keep it for as long as the law requires, and may report it to the National Center for Missing & Exploited Children or to law enforcement. Content sent while ZDR is on is never stored, so it is never screened or reviewed.

No Training on Your Data

While Zero Data Retention is on, which is the default, we do not use API Content or Request Metadata to train, fine-tune or evaluate our models. We only train on API Content from accounts whose owner has turned ZDR off (see above). We never train on cloned voices or the reference audio used to create them, whatever your setting.

What We Keep for Each Request

To bill you accurately, show your usage in the console and protect the Services from abuse, we keep Request Metadata for each request:

• a request ID and the time of the request;
• which of your API keys made it;
• the model and voice used;
• the number of characters sent (the count only, not the characters themselves);
• the length of audio generated;
• the time to first byte and the response status.

Request Metadata is kept while your account is open and for as long afterwards as we need it for billing, tax and legal obligations. You can ask us to delete it at any time, subject to those obligations.

Speech to Text

Recordings you send for transcription are converted and transcribed in memory, on speech recognition that we run ourselves rather than a third-party service, and discarded as soon as the transcript is returned. Neither the recording nor the transcript is ever stored, whatever your ZDR setting, and neither is used for training. We keep only the request's metadata: when it was made, which key made it, the audio's length, the transcript's length in characters and the response status.

Cloned Voices

When you clone a voice, through the API or the console, we keep the reference clip you upload (converted to a standard audio format) and its transcript on our servers, so the voice can be used in your requests. If you don't provide a transcript, we create one using speech recognition that we run ourselves, not a third-party transcription service; it is stored with the voice and shown to you. A voice's recording may identify the person speaking, so we treat it as sensitive personal information:

• it is only ever used to speak in that voice for your account, and is never available to other customers;
• it is never used to train, fine-tune or evaluate our models;
• you must confirm, each time you clone a voice, that the speaker has given permission for their voice to be cloned, and we record who confirmed it and when;
• when you delete a voice, its recording and transcript are deleted from our servers immediately; all of your voices are deleted when you close your account.

If you are the person in a recording that someone cloned without your permission, contact us at business@iodine.dev and we will remove it.

Infrastructure and Security

• Encryption. All traffic to the Services is encrypted in transit with TLS. Stored data is encrypted at rest.
• Hashed credentials. API keys, sign-in links and session tokens are stored only as SHA-256 hashes, so they cannot be read back even by us.
• Restricted access. Access to production systems is limited to the engineers who operate them, and service secrets are kept in a dedicated secrets manager rather than in code or configuration files.

Account and Billing Information

We only collect and use Personal Information to the extent necessary to provide you with the Services. When you create an account we collect your email address, and optionally your name and your organization's name. We use a single session cookie to keep you signed in to the console; we do not use advertising or tracking cookies. Payments are handled by Stripe: we never see or store your full card number, only the references Stripe gives us to charge the payment method you saved.

Service Providers

We use a small number of providers to run the Services. None of them receive API Content for storage, and none use your data to train models.

• Stripe processes payments and invoices.
• Resend delivers sign-in emails.
• Cloudflare hosts this website and the console, and carries network traffic to our servers.
• Google Fonts and jsDelivr serve fonts and stylesheets used by this website.

Personal Information Retention

While ZDR is on, API Content is not retained at all (see "Zero Data Retention" above). We keep other Personal Information for the shortest time necessary to provide you with Services and to meet all our legal and compliance obligations. To determine this retention period, we take into account (i) the nature of the Personal Information gathered; and (ii) the nature of our legal and compliance obligations. All Personal Information no longer required by us is destroyed and/or erased.

Personal Data from Third Parties

We may, from time to time, obtain Personal Information from third parties to enable us to better tailor our Services to you (Third Party Personal Information). When we obtain Third Party Personal Information, we will notify you within one month. If we use or share Third Party Personal Information, we will notify you immediately.

Your Rights

With respect to Personal Information we hold, you have the following rights:

• Access: You may request from us access to your data that we hold on you.
• Rectification: If the data we hold on you is inaccurate, you may request that we correct it. If the data we hold is incomplete, you may request that we complete it.
• Erasure: Subject to certain conditions, you may request that we erase all of the data we hold on you.
• Restrictions: Subject to certain conditions, you may request that we restrict the processing of data we hold on you.
• Portability: Subject to certain conditions, you may request that we transfer all the data we hold on you to a third party (including yourself).
• Objections: Subject to certain conditions, you may object to us processing the data we hold on you.

Social Media

You may choose to engage with our Services through a variety of social media sites and platforms. When you interact with our Services through these social media platforms, you may allow us to receive Personal Information on how you are interacting with our Services. This will depend on your privacy settings on each social media site and platform.

Minors

We do not knowingly collect information on children under the age of 16 ("Child"). If you become aware that a Child has provided us with Personal Information, please contact us. If we become aware that we have collected Personal Information from a Child without their parent's verifiable authorization to access our Services, we will take steps to remove that information from our servers.

Third Party Links

Our website contains links to other third-party websites not owned or managed by Persona Enterprises Inc.. This privacy policy applies to this website only. If you click a link to a third-party website, the privacy policy of that website will apply. We highly recommend that you read the privacy policies of other websites as they may be different from ours.

Updates

We are constantly reviewing our privacy policy and procedures to ensure that they meet best practices. This policy was last updated on November 09, 2026.

Contact Us

If you have any questions or suggestions about our privacy policy or want to know more information about Personal Information we hold, please contact us at business@iodine.dev, or write to us at:

Persona Enterprises Inc.
Attn: Privacy
251 Little Falls Drive
Wilmington, New Castle County, Delaware 19808
United States

Complaints

If you believe that we have breached our obligations to you under this privacy policy or relevant privacy laws, please contact us at business@iodine.dev. If you are not satisfied with our response, you have the right to lodge a complaint at any time to your relevant authority.