Privacy policy

How ASIC collects, uses, and protects personal information on our website and in our Hosted Services, including the images and questions sent to our Models.

Version 1.0Effective October 13, 2026

On this page16 sections

1. Who we are

This Privacy Policy explains how American Space & Intelligence Corporation (“ASIC,” “we,” “us,” or “our”) collects, uses, discloses, and protects personal information when you visit our website at asi-corp.com and when you or your organization use our Services. This version is effective October 13, 2026 and was last updated on October 13, 2026.

This Policy covers ASIC, including ASIC Research and ASIC Intelligence. ASIC Research carries out foundational research, including the Kanzi Program. ASIC Intelligence develops our products, including the Arkhe model family, the Pipwick platform, and the Haytham vision models. In this Policy:

  • “Models” means our AI models, including the Haytham vision models (in every size, such as Haytham-2B-1.0 and Haytham-4B-1.0) and the Arkhe model family, and any model weights we make available.
  • “Hosted Services” means the APIs, platforms (such as the Pipwick platform), consoles, and other online services we operate to provide access to our Models and products. “Services” means the Hosted Services and the Models together. Our Terms of service also treat our website as part of the Services; in this Policy, we refer to the website separately.
  • “Customer” means a business or individual that has an account with us or has agreed to our Terms of service.
  • “Customer Content” means Inputs, Outputs, and Stored Context, as our Terms of service define them: the images, questions, prompts, files, documents, and other content a Customer or its users submit to the Hosted Services; the answers and other results the Hosted Services return; and information from them that the Hosted Services keep for that Customer for use in later sessions, such as remembered project context and retrieval indexes. We handle Customer-specific model updates (“Account-Specific Learning” in our Terms of service) as Customer Content under this Policy.
  • “Training Endpoint” means an endpoint that we have told you is a Training Endpoint, as our Terms of service define it. Section 5 explains how we use content sent to one.

Some of the Services described here may not be available yet. The parts of this Policy that describe them apply when you use them.

Who is responsible for your information

For the personal information described in this Policy, the controller (the “business,” under California law) is American Space & Intelligence Corporation, except for Customer Content we process on a Customer’s instructions (see “Controller and processor roles” below). Privacy contact: privacy@asi-corp.com. Legal notices: legal@asi-corp.com.

If we appoint a data protection officer or a representative in the European Union or the United Kingdom, we will name them here.

Controller and processor roles

Our role depends on the information. We are a controller when we decide why and how information is processed. We are a processor (a “service provider,” under California law) when we process information on a Customer’s instructions to provide the Hosted Services.

Controller and processor roles
RoleWhenExamples
ControllerWe decide the purposes and means of processingWebsite request logs; accounts and sign-in; billing and payment records; communications with us, such as support, sales, and legal correspondence; requests for access to Model weights; datasets we use to develop our Models; security and abuse records we create; Customer Content we review for security or abuse; bug bounty reports and payment records (below); flagged content we use to improve our abuse-detection systems, Customer Content a Customer has opted in to let us use for training, and Inputs and Outputs sent to a Training Endpoint, to the extent we use them for training (Section 5)
Processor / service providerWe process on a Customer’s instructions to provide the Hosted Services, except to the extent used for training on a Training EndpointImages, questions, prompts, files, and other Customer Content sent to the Hosted Services; the outputs returned; and project context, retrieval indexes, and Customer-specific model updates kept for a Customer

When we act as a processor, the Customer’s own privacy policy governs how it collects and uses that information, and our agreement with the Customer, including our Terms of service and any data processing agreement we sign with it, governs how we handle it. Business Customers can request a data processing agreement at privacy@asi-corp.com.

What this Policy does not cover

  • Corstrike, a separate operating company with its own website (corstrike.com), terms, and privacy policy.
  • Services offered under the InfoLang name. InfoLang is becoming ASIC Intelligence; until we announce otherwise, services offered as InfoLang, including at infolang.ai, are covered by the privacy policy published there.
  • How Customers handle personal information, including the images and questions they choose to send to the Hosted Services.
  • Personal information of our job applicants, employees, and contractors, which separate notices we give them describe.
  • Third-party websites and services, including any platform you use to download Models (see Section 14).

2. Personal information we collect

What we collect depends on how you interact with us. Today, our public website collects very little: it is a static site served by our hosting and network provider, Cloudflare.

You do not have to give us personal information to browse our website. To use the Hosted Services we need account and billing information; without it we cannot provide them.

Our website

  • Request logs. When your browser requests a page, Cloudflare processes and logs technical information about the request: your IP address, user agent (browser and operating system), the URL requested, and timestamps, and network error reports your browser may send to Cloudflare. Cloudflare uses these logs to deliver and secure the site for us. We do not keep our own copies of request logs. Cloudflare’s dashboard shows us aggregated traffic statistics and details of requests its security systems block or challenge, such as the IP address and URL.
  • Theme preference. If you switch between light and dark mode, the site stores your choice in your browser’s local storage under the name asic-theme. It stays on your device and is never sent to us.

Our website’s own code does not set cookies (Section 7 describes a security cookie Cloudflare may set). The website does not use analytics, advertising pixels, or session recording, and it has no forms or accounts. Videos and images on the site are served from our own site, so viewing them does not send information to third-party media platforms.

Hosted Services

If you or your organization use the Hosted Services, we also collect:

  • Account information: name, email address, organization, role, and log-in credentials.
  • API keys and authentication records.
  • Billing information: billing contact and address, plan, invoices, and payment history. Payment card details are collected and processed by our payment processor; we do not store full card numbers.
  • Usage and request metadata: which Model, Model size, and endpoint were called and when, whether hand-off mode was used, request and response sizes, latency, errors, and usage volumes.
  • Customer Content: images, questions, prompts, files, and other inputs, the outputs returned, and any project context or Customer-specific model updates kept for the Customer (see Section 3).

If you use our Models through a third-party marketplace or router, such as OpenRouter, that provider’s own terms and privacy policy govern what it collects from you, it decides what it sends us, and we handle the requests we receive from it as this Policy describes.

Communications

If you email us, for example at privacy@asi-corp.com or legal@asi-corp.com, we receive your name, your email address, and whatever you include in your message. Our email hosting provider receives and stores these messages for us.

Bug bounty participants

If you report a vulnerability under our Bug bounty, we collect the contents of your report, your name or handle, your contact details, and, if we pay you, your payment details, your tax form (Form W-9 or W-8BEN, which includes your tax identification number), and the results of our sanctions screening. We use them to assess the report, run the program, pay rewards, and meet tax, sanctions, and other legal obligations. We handle this information as this Policy describes, and we share your name only with your consent or where the law requires.

Models you run yourself

If we make Model weights available and you run them on your own systems, we do not receive your inputs or outputs. If we require you to request access before downloading weights, for example through a gated repository, or to accept a model license, we receive the information you give in that request, such as your name or username, email address, organization, and intended use, and a record of your acceptance. We use it to decide on access, keep a record of license acceptance, enforce the model license, and comply with export control and sanctions laws. A platform you use to download weights may also collect information about you under its own privacy policy.

Data used to develop our Models

We use datasets to develop and evaluate our Models. They may contain personal information, such as images that show people. Section 5 explains how we use this data and the choices you have.

Categories of personal information (CCPA)

The table below describes the personal information we collect using the categories listed in the California Consumer Privacy Act (Cal. Civ. Code § 1798.140(v)(1)). Items marked “Hosted Services” are collected only when the Hosted Services are used.

Categories of personal information (CCPA)
CCPA categoryExamplesSourcesPurposes
IdentifiersWebsite: IP address and user agent in request logs. Hosted Services: name, email address, account ID, API key identifiers. Weight access requests: name or username and email address. Communications and bug bounty reports: name or handle and email address. Model development: names or other identifiers that may appear in datasets.Automatically from your browser or client; you; your organization; marketplaces and routers, such as OpenRouter, that send us requests; licensed and publicly available datasetsDeliver and secure the website and Services; operate accounts; respond to you; review weight access requests and enforce model licenses; Model development and evaluation
Customer records (Cal. Civ. Code § 1798.80(e))Hosted Services: name, organization, billing contact, billing address. Bug bounty: payment details and tax formsYou; your organization; our payment processorContracts, billing, tax, fraud prevention; paying bug bounty rewards
Commercial informationHosted Services: plan, orders, invoices, payment history, usage volumesYou; our systems; our payment processorProvide and bill for the Services; keep business records
Internet or other electronic network activityWebsite: URLs requested and timestamps, and network error reports your browser may send to Cloudflare. Hosted Services: request metadata, such as endpoint, Model, Model size, latency, and errors.Automatically; Cloudflare; marketplaces and routers, such as OpenRouter, that send us requestsSecurity, abuse prevention, debugging, performance, capacity planning
Geolocation dataApproximate location (country or region) derived from IP address. We do not collect precise geolocation, though image files in Customer Content can contain it (see Section 3).Derived from IP addressWebsite: security and delivery (Cloudflare). Hosted Services: also sanctions and export-control compliance.
Audio, electronic, visual, or similar informationHosted Services: images and other Customer Content sent to Models, which may show people. Model development: images in datasets we use to develop our Models, which may show people.Customers, who send it on their own instructions; marketplaces and routers, such as OpenRouter, that send us requests; licensed and publicly available datasetsProvide the Hosted Services on the Customer’s instructions; investigate abuse; meet legal obligations; Model development and evaluation, including training on Training Endpoints
Professional or employment-related informationJob title and organization, if you provide them, including in a request for access to Model weights together with your intended useYou; your organizationAccount administration, support, business communications; review weight access requests and comply with export control and sanctions laws
Sensitive personal informationHosted Services: account log-in credentials. Payment card details are handled by our payment processor, not stored by us.YouAuthentication, payments, security

We do not draw inferences about you to build a profile. We do not seek to collect government ID numbers (other than the tax identification numbers on tax forms bug bounty participants send us), precise geolocation, biometric identifiers, or special categories of data under Article 9 of the GDPR. Customer Content and the datasets we use to develop our Models can contain such data; Sections 3 and 5 explain how we handle it.

3. Images and other inputs sent to our Models

Haytham, our vision model, answers questions about images. It takes an image and a question and returns a decision, such as yes or no, or a choice among options. To do that, the Hosted Services must receive and process the image, the question, and any options or context sent with them. Other Models, such as Arkhe, receive the prompts, files, and other inputs a Customer sends.

What Customer Content may contain

Images and inputs can include personal information. They may show faces, people in the background, vehicle license plates, documents with names or account numbers, screens, or the inside of homes. Image files can also carry metadata, such as the device used and the place and time a photo was taken. We recommend removing metadata, and cropping out people and details you do not need, before you send an image.

Customer responsibilities

Customers decide what to send and why. Our Terms of service do not allow Customers to send protected health information covered by HIPAA, payment card data, government-issued identification numbers, biometric data used to identify individuals, or other sensitive personal information subject to heightened legal requirements, unless we have agreed in writing to support it. Where we agree, the Customer is responsible for meeting the laws that apply, such as the Illinois Biometric Information Privacy Act.

A Customer that sends personal information to the Hosted Services is responsible for:

  • having a lawful basis to collect it and to have us process it, and giving the notices and obtaining the consents the law requires;
  • for any sensitive data it is permitted to send, such as images that show children, meeting any additional legal requirements that apply, including any parental consent the law requires and the conditions for special categories of data under Article 9 of the GDPR; and
  • not sending personal information it is not permitted to share, and following the acceptable use rules in our Terms of service.

How we process Customer Content

  • We process Customer Content only to provide the Hosted Services on the Customer’s instructions and, on a Training Endpoint, to train our Models as Section 5 describes, to keep the Services secure and to detect and investigate abuse (including as Section 5 describes for flagged content), and to meet legal obligations.
  • We do not sell Customer Content, use it for advertising, or use it to build profiles of the people it shows.
  • We do not use Customer Content to identify the people shown in images, or to create or keep templates of their facial or other physical features, for our own purposes.
  • If Customer Content contains biometric identifiers or biometric information, we permanently delete it when the purpose for which it was sent has been satisfied, or within three years of the individual’s last interaction with us, whichever comes first, except content we keep for training on a Training Endpoint (Section 5).
  • We do not use Customer Content to train or improve Models we offer to others unless the Customer opts in or sends it to a Training Endpoint, which we tell the user about before use. See Section 5.
  • Automated systems may scan Customer Content to detect abuse. Our personnel access Customer Content only where needed to operate the Hosted Services, provide support a Customer asks for, investigate abuse or security issues, or comply with law.

Features that use more than one model

Some features route an input to a different model. For example, Haytham’s optional hand-off mode sends images such as charts, maps, documents, and multi-panel images to a separate specialist model, which we run. When hand-off mode is off, images are not sent to the specialist model. Before a feature sends Customer Content to a model or provider other than the one you called, we describe this in that feature’s documentation and list any third-party provider involved on our Subprocessors page.

If a Hosted Service, such as Pipwick, lets a Customer connect third-party tools or models, we send Customer Content to them on the Customer’s instructions. Their own terms and privacy policies govern what they do with it.

Memory, retrieval, and learning features

Some Services, such as Arkhe, can keep project context between sessions, search records a Customer provides, and update model parameters for that Customer. Depending on how a Customer configures them, the Hosted Services may store project context and other Customer Content so it can be retrieved in later tasks, and may update model weights that are used only for that Customer.

We treat stored context, retrieval indexes, and Customer-specific model updates as Customer Content. We keep them separate from other Customers’ data and from the general Models we offer to others, we process them only on the Customer’s instructions, and we use them only to provide the Hosted Services to that Customer. This is not training of our general Models, which Section 5 covers. Customer-specific model updates are not used to train the Models we offer to others unless the Customer opts in. Stored Context is not used for training except to the extent it is sent as an Input to a Training Endpoint.

Deleting content from memory or from a retrieval index does not by itself undo model updates already made from it. Customer-specific model updates are deleted when the Customer asks us to delete them or its account closes.

How long we keep it

We keep inputs and outputs only as long as needed to provide the Hosted Services, investigate abuse, and meet legal obligations, and then delete them. Inputs and Outputs sent to a Training Endpoint are also kept as long as needed for training and evaluation and as the law allows. Context that the Hosted Services store for a Customer, retrieval indexes, and any Customer-specific model updates are kept until the Customer deletes them or closes its account.

If you appear in an image a Customer sent

If you believe a Customer sent an image of you, or other information about you, to the Hosted Services, that Customer controls the information and is the right place for your request. We do not identify people in images, and we will not use facial recognition to search Customer Content for images of a particular person, so we usually cannot find images of you on our own. If you contact us at privacy@asi-corp.com, tell us which Customer sent the information if you know. If we can tell which Customer is involved, we will pass your request to that Customer and help it respond. This does not apply to content we use for training on a Training Endpoint. For that use we are the controller and Section 5 applies, so send those requests to privacy@asi-corp.com.

4. How we use personal information

We use personal information for the purposes below. If you are in the European Economic Area (EEA), the United Kingdom, or Switzerland, the table also shows the legal basis we rely on under the GDPR or UK GDPR. It covers information for which we are the controller, including the limited review of Customer Content for security and abuse and training on Training Endpoints. Otherwise, we process Customer Content on the Customer’s instructions (see Section 3).

How we use personal information
PurposeInformation usedLegal basis (GDPR)
Deliver the website, keep it available, and protect it from attacksRequest logs; network error reports; approximate locationLegitimate interests (Art. 6(1)(f))
Provide the Hosted Services: create and manage accounts, authenticate users, process requests, and return outputsAccount information; API keys; request metadataContract (Art. 6(1)(b)), where you are our Customer; otherwise legitimate interests (Art. 6(1)(f)) in providing the Services your organization has contracted for
Billing, metering, tax, and fraud preventionAccount, billing, and usage informationContract (Art. 6(1)(b)), where you are our Customer; otherwise legitimate interests (Art. 6(1)(f)) in providing the Services your organization has contracted for; legal obligation (Art. 6(1)(c))
Review requests for access to Model weights, keep records of license acceptance, enforce model licenses, and screen for export control and sanctions complianceInformation in your access request, such as name or username, email address, organization, and intended useLegitimate interests; legal obligation
Security, abuse investigation, and enforcement of our TermsRequest logs; request metadata; account information; Customer Content under reviewLegitimate interests; legal obligation
Review flagged content and improve our abuse-detection systemsCustomer Content flagged for review; request metadataLegitimate interests (Art. 6(1)(f))
Develop, train, evaluate, and safety-test our ModelsInformation in the datasets we use, which may include images that show people; Customer Content only where the Customer has opted in or sent it to a Training EndpointLegitimate interests (Art. 6(1)(f))
Train, evaluate, and improve our Models using content sent to a Training EndpointInputs and Outputs sent to a Training Endpoint; request metadataLegitimate interests (Art. 6(1)(f)). The Customer chose a Training Endpoint after we told it the endpoint is one.
Keep the Services reliable and improve them using aggregated or de-identified metricsUsage and request metadataLegitimate interests
Run the bug bounty program: assess reports, pay rewards, and screen for tax and sanctions complianceReport contents; name or handle; contact details; payment details; tax forms; sanctions-screening resultsLegitimate interests (Art. 6(1)(f)) in finding and fixing vulnerabilities; contract (Art. 6(1)(b)) once we confirm a reward; legal obligation (Art. 6(1)(c))
Respond to messages and send service, security, and legal noticesContact details; message contentContract; legitimate interests
Send product news or marketing emails, if we start sending themName and email addressConsent (Art. 6(1)(a)) where required; otherwise legitimate interests. You can unsubscribe at any time.
Comply with law, including sanctions and export controls; respond to lawful requests; establish or defend legal claimsInformation relevant to the matterLegal obligation; legitimate interests

Where we rely on legitimate interests, those interests are operating a secure and reliable website and model service, preventing abuse, researching and developing our Models, and running our business. We weigh them against your rights and do not rely on them where your rights override them. You can object to processing based on legitimate interests (see Section 10).

We will not use personal information for purposes materially different from those described here without telling you and, where the law requires, asking for your consent.

5. Model training and improvement

Customer Content

We do not use Customer Content to train, fine-tune, or otherwise improve Models we offer to others unless the Customer opts in through its account settings, where available, or agrees in writing, or sends it to a Training Endpoint (see below). The other exception is abuse and safety review. If specific Customer Content is flagged for abuse or safety review, we may use it for that review and to improve the safeguards we use to detect and prevent similar abuse. We do not use it to train Models we offer to others.

Customer-specific model updates that the Hosted Services make on a Customer’s instructions, and use only for that Customer (see Section 3), are part of providing the Hosted Services. They are not training of our Models for the purposes of this Section.

A Customer that has opted in can withdraw at any time through its account settings, where available, or by contacting privacy@asi-corp.com. Withdrawal applies to Customer Content submitted after it takes effect. It does not remove what a Model has already learned from content used before then.

Training Endpoints

A Training Endpoint is an endpoint that we have told you is trainable. Any endpoint, free or paid, may be one. We tell you before you use it, where you select the endpoint, for example in our documentation, pricing, the console, or the endpoint’s listing on a marketplace such as OpenRouter. An endpoint is not a Training Endpoint unless we have told you so. If we make an existing endpoint a Training Endpoint, we also notify its current users, by email to the account contact or by a notice in the console or the marketplace listing. Documentation alone is not enough. The change applies only to content sent after that notice.

On a Training Endpoint, we use the images, questions, and other Inputs and the Outputs to train, evaluate, and improve our Models, including Models we offer to others. For that use we act as a controller, not as a Customer’s processor. We rely on our legitimate interests (Section 4), and the Customer chose the endpoint after notice. We keep this content as long as needed for training and evaluation and as the law allows. If it is deleted, we stop using it for further training, but we may not be able to remove its effect on Models we have already trained. If you appear in an image sent to a Training Endpoint, you can ask about it or object by contacting privacy@asi-corp.com (Section 10).

Do not send personal, confidential, or sensitive information to a Training Endpoint, including images of people you do not have the right to share. To avoid training, use an endpoint that is not a Training Endpoint.

Usage metrics

We may use aggregated or de-identified usage metrics, such as request volumes, latency, error rates, and which Models are used, to operate and improve the Services. These metrics do not include the content of images or questions. Where we de-identify information, we keep it in de-identified form and do not attempt to re-identify it, except as the law permits.

Data used to develop our Models

We develop and evaluate our Models using data that may include data we create, data we license, and publicly available research datasets and benchmarks. Some of this data includes images that show people. We use it to build and test our Models, relying on our legitimate interests in research and model development, and we do not use it to identify the people shown. Training and evaluation run on rented GPU infrastructure (see Subprocessors). Customer Content is not part of those runs unless a Customer has opted in or sent it to a Training Endpoint.

This Section is our notice to people whose information may appear in this data, because we usually cannot contact them individually. If you believe your personal information appears in it, you can ask about it, object, or ask us to remove it by contacting privacy@asi-corp.com. Where we can locate it, we will remove it from our copies of the data used for future training. We may not be able to remove its effect on Models we have already trained. You can object to this use at any time (Section 10).

Where the law requires it, we will publish documentation about the data used to develop a Model before we make that Model available.

6. Sale, sharing, and disclosure

We do not sell personal information, and we do not “share” it for cross-context behavioral advertising, as the California Consumer Privacy Act (CCPA) defines those terms. We do not process personal information for targeted advertising, and we do not sell it as other US state privacy laws define a sale.

We have no actual knowledge of selling or sharing the personal information of consumers under 16. If our practices change, we will update this Policy first and provide the opt-out methods the law requires, including honoring opt-out preference signals such as Global Privacy Control.

Disclosures for a business purpose

We disclose personal information to service providers and contractors that help us run the website and Services, under contracts that limit their use of it to providing services to us. The vendors that process data to run our website and email and to develop and operate our Models, including our email hosting provider, are listed on our Subprocessors page. In the preceding 12 months, we have disclosed the categories below for a business purpose:

Disclosures in the preceding 12 months
CategoriesRecipientsPurpose
Identifiers, internet or other electronic network activity, and approximate geolocation (website request logs)Cloudflare, our hosting and network providerDelivering and securing the website
Identifiers and the contents of your message (emails you send us)Our email hosting providerReceiving and responding to your messages
Audio, electronic, visual, or similar information in datasets we use to develop our Models, which may show peopleLambda, our GPU compute providerTraining and evaluating our Models

When the Hosted Services are available, we will also disclose the following categories for a business purpose, and we will update this section to reflect actual disclosures:

Planned disclosures for the Hosted Services
CategoriesRecipientsPurpose
Identifiers, customer records, commercial information, and sensitive personal informationHosting, infrastructure, and payment service providersProviding, securing, and billing for the Services
Audio, electronic, visual, or similar information in Customer Content, and internet or other electronic network activity (request metadata)Service providers that host or run the Hosted Services for us, such as Modal, which will host the Haytham API, and Lambda, which will run training on content sent to Training EndpointsProviding the Hosted Services on the Customer’s instructions; training on Training Endpoints

Other disclosures

We may also disclose personal information:

  • at a Customer’s direction, or with your consent;
  • to our affiliates that help us operate the website and Services, for the purposes described in this Policy and under obligations consistent with it. We do not disclose Customer Content to our affiliates, including Corstrike, except on the Customer’s instructions;
  • to professional advisers, such as lawyers, accountants, and auditors, who owe us duties of confidentiality;
  • in connection with a merger, acquisition, financing, reorganization, or sale of all or part of our business, subject to this Policy or equivalent protections;
  • to protect the rights, property, or safety of ASIC, our users, or others, and to investigate and prevent fraud, abuse, and security incidents; and
  • when required by law, regulation, or legal process, such as a subpoena or court order.

We disclose Customer Content in response to a government or legal request only where we believe the law requires it, and, where the law allows, we notify the affected Customer.

Sensitive personal information

We use sensitive personal information, such as account log-in credentials, only to provide the Services, keep them secure, and process payments. The CCPA permits these uses without a right to limit, and we do not use sensitive personal information to infer characteristics about you.

7. Cookies and local storage

Our website’s own code does not set cookies. If Cloudflare, our hosting provider, asks your browser to complete a security check, it may set a strictly necessary security cookie, such as __cf_bm or cf_clearance, to record the result. That cookie is not used for advertising or tracking.

The website does not use analytics, advertising pixels, session recording, fingerprinting, or other tracking technologies, and it does not load third-party tracking scripts. Your browser may keep the following items for our site:

Cookies and local storage
NamePurposeWhere it is storedHow long
asic-themeRemembers whether you chose light or dark modeOnly in your browser’s local storage, on your device. It is never sent to us.Until you change it or clear your browser’s site data
Network Error Logging (set by Cloudflare)Lets your browser report failed connections to our site to Cloudflare so it can diagnose network problemsA reporting policy stored by your browser; reports go to CloudflareUp to 7 days, renewed on each visit

You can remove these items at any time by clearing site data for our website in your browser settings. The site works without them; your theme choice simply will not be remembered.

Our website does not track you across sites, so a Do Not Track or Global Privacy Control signal has nothing to turn off here.

Hosted Services

If you sign in to a console in our Hosted Services, it may use cookies or similar storage that are strictly necessary to keep you signed in and protect your account. We do not use them for advertising or cross-site tracking. Before we use any cookies that are not strictly necessary, we will describe them here.

8. Retention

We keep personal information only as long as needed for the purposes in this Policy. We decide how long by considering what the information is, why we have it, how sensitive it is, and how long the law requires or allows us to keep it.

Retention
InformationHow long we keep it
Website request logs and network error reportsKept by Cloudflare for a limited period under its retention practices. We do not keep separate copies.
Theme preference (asic-theme)Stored only on your device until you clear it; we never receive it
Account informationFor the life of the account, then deleted or de-identified, except records we must keep for billing, legal, or security reasons
API keysUntil revoked or the account is closed
Customer Content (inputs and outputs)Only as long as needed to provide the Hosted Services, investigate abuse, and meet legal obligations, then deleted
Inputs and Outputs sent to a Training EndpointAs long as needed for training and evaluation and as the law allows
Memory and retrieval indexes (project context stored for a Customer)Until the Customer deletes them or closes its account, then deleted
Customer-specific model updatesUntil the Customer deletes them or closes its account, then deleted
Biometric identifiers or biometric information in Customer ContentPermanently deleted when the purpose for which it was sent has been satisfied, or within three years of the individual’s last interaction with us, whichever comes first
Request metadata and usage recordsAs long as needed for billing, security, and reliability, then deleted or aggregated
Requests for access to Model weights and records of license acceptanceAs long as the model license applies, and as long as needed to meet record-keeping obligations under export control laws
Datasets used to develop our ModelsAs long as we use them for research and development, subject to removal requests (Section 5)
Billing and tax recordsAs long as tax, accounting, and other laws require
Communications with usAs long as needed to respond and to keep a record of our dealings with you
Bug bounty reports and participant detailsReports and contact details: as long as needed to fix the issue and run the program. Payment records, tax forms, and sanctions-screening results: as long as tax, accounting, and sanctions laws require, then deleted
Records of privacy requestsAt least 24 months, as California regulations require

Deleted information may remain in backups until they expire on their normal cycle; we do not use it for any other purpose in the meantime. We may keep information longer when the law requires it or when we need it to establish, exercise, or defend legal claims.

9. International transfers

We are based in the United States, and we process personal information in the United States. Our hosting and network provider operates a global network, so requests to our website are usually handled at a data center near you. Our service providers may process information in other countries where they operate.

Where required, when we transfer personal information from the EEA, the United Kingdom, or Switzerland to a country that has not been found to provide adequate protection, we rely on the European Commission’s Standard Contractual Clauses, the UK International Data Transfer Addendum, or another lawful transfer mechanism, with supplementary measures where appropriate. You can request a copy of the relevant safeguards at privacy@asi-corp.com.

10. Your privacy rights

Depending on where you live, you may have the rights below. Where we process your information as a processor for a Customer, please send your request to that Customer (see Section 3).

California residents

Under the CCPA, as amended by the California Privacy Rights Act (CPRA), you have the right to:

  • know what personal information we have collected about you, including the categories, sources, and purposes, and the categories of third parties to whom we disclosed it, and receive a copy of the specific pieces of personal information we hold;
  • delete personal information we collected from you, subject to legal exceptions;
  • correct inaccurate personal information;
  • opt out of the sale or sharing of personal information (we do not sell or share it);
  • limit the use of sensitive personal information (we use it only for purposes that do not give rise to this right); and
  • not be discriminated against for exercising any of these rights.

Residents of other US states

If you live in a state with a comprehensive consumer privacy law, such as Colorado, Connecticut, Oregon, Texas, or Virginia, you may have the right to confirm whether we process your personal information and to access, correct, delete, and obtain a portable copy of it, and, in some states such as Oregon and Minnesota, to obtain a list of the specific third parties to which we have disclosed your personal information. You may also have the right to opt out of targeted advertising, the sale of personal information, and profiling in furtherance of decisions that produce legal or similarly significant effects. We do not engage in any of those activities.

EEA, UK, and Swiss residents

Under the GDPR, the UK GDPR, or the Swiss Federal Act on Data Protection, you may have the right to access, rectify, or erase your personal information; to restrict or object to its processing; to receive it in a portable format; and to withdraw consent at any time, without affecting processing that took place before you withdrew. You can object to direct marketing at any time. You can also lodge a complaint with the data protection authority where you live or work, such as the Information Commissioner’s Office in the UK.

How to make a request

  • Email privacy@asi-corp.com. Tell us which right you want to exercise and how you have interacted with us, for example as a website visitor, an account holder, or someone who emailed us.
  • We verify requests by matching the information you give us against information we already hold. For account holders, we will usually ask you to write from the email address on the account. We will not ask for more information than we need.
  • Our website has no accounts, its own code sets no cookies, and we do not keep request-level logs ourselves, so we usually cannot link website activity to a named person.
  • We will confirm receipt within 10 business days and respond within 45 days, as California law requires, or within one month under the GDPR. If we need more time, as the law allows, we will tell you why.
  • We do not charge for requests, unless a request is manifestly unfounded or excessive and the law allows a fee.

Authorized agents

You may use an authorized agent to make a request for you. Unless the agent holds a valid power of attorney, we may require the agent to provide your signed written permission, and we may ask you to verify your identity directly with us or confirm that you gave the agent permission.

Appeals

If we decline to act on your request, we will explain why. You may appeal by replying to our decision or by emailing privacy@asi-corp.com with “Privacy appeal” in the subject line. We will respond within the time your state’s law requires and explain the outcome. If you disagree with the result, you may contact your state attorney general.

Non-discrimination

We will not deny you service, charge you a different price, or provide a different level of quality because you exercised your privacy rights.

11. Security

We use administrative, technical, and organizational measures appropriate to the risk. Our website is served only over encrypted (TLS) connections, and our Model weights are kept in private storage with restricted access. When we operate Hosted Services, we will protect them and Customer Content with measures that include encryption in transit and access controls that limit who can reach production systems and Customer Content. No method of transmission or storage is completely secure, and we cannot guarantee absolute security.

If we become aware of a security incident affecting your personal information, we will notify you and the relevant authorities as the law requires. To report a security vulnerability, follow our Bug bounty and email security@asi-corp.com. To report a suspected compromise of your account or personal information, email privacy@asi-corp.com.

12. Automated decision-making

Our Models produce automated outputs, such as a yes or no answer about an image, for Customers who decide how to use them. Outputs can be wrong, especially on charts, edited images, and other inputs that are hard to read, so Customers should review them before relying on them. Benchmark results we publish describe performance under the test conditions we state. They do not guarantee accuracy on a Customer’s images.

We do not use automated processing alone to make decisions that have legal or similarly significant effects on anyone, including website visitors and people shown in Customer Content. Our Terms of service prohibit Customers from using outputs as the sole basis for decisions with legal or similarly significant effects on individuals without human review.

If we begin using automated decision-making technology in a way that California regulations or other laws cover, we will update this Policy and provide the required notices and choices first.

13. Children

Our website and Services are not directed to children, and the Services are only for people 18 and older. We do not knowingly collect personal information from anyone under 18. If you believe someone under 18 has given us personal information, contact privacy@asi-corp.com and we will delete it.

Customers may send images that show children only where they have a lawful basis to do so, including any parental consent the law requires (see Section 3).

15. Changes to this Policy

We may update this Policy as our website, Services, and legal obligations change. When we do, we will post the new version here with a new version number and effective date. If a change is material, we will give notice through the Hosted Services or by email to account holders before it takes effect, where reasonably practicable. If the law requires your consent to a change, we will ask for it.

If we make a material change to how we use personal information we have already collected, including a change that would let us use Customer Content to train our Models, the change will apply only to information collected after it takes effect unless you consent. Training Endpoints are the exception that Section 5 describes: we announce each one where you select it, and making an existing endpoint a Training Endpoint applies only to content sent after we tell you.

16. Contact us

American Space & Intelligence Corporation is the controller of the personal information described in this Policy, except Customer Content we process on a Customer’s instructions. If you are in the EEA, the UK, or Switzerland and are not satisfied with our response, you may contact your local data protection authority.