API Access Terms

These API Access Terms ("API Terms") supplement the general Terms of Service at /legal/terms and the Acceptable Use Policy at /legal/aup. They apply to any access to or use of the AI2 Design API, the eighteen-endpoint surface published at /api-access, and any related documentation, SDK, sample code or output. Where these API Terms conflict with the general Terms of Service with respect to the API only, these API Terms govern.

By subscribing to AI2 Design early-access updates, by accessing or using the Service in any capacity, or by making any payment to BEY AGENCY LTD, you acknowledge that you have read, understood, and agree to be bound by this API Access Terms. You acknowledge that you have had the opportunity to review this document with legal counsel of your choice if desired. You acknowledge that violation of this document may result in immediate suspension or termination of access, forfeiture of any amounts paid (which are non-refundable), reporting to appropriate authorities, criminal prosecution, and civil liability.

The API surface published at /api-access (eighteen endpoints across Extraction, Inspiration, Agent, Launchpad and Tooling categories) is in private beta until general availability ("GA"). Access is granted in waves to founding accounts. Endpoints are released progressively as they stabilise; you may sign up for a tier and reserve quota before all endpoints in that tier are live, and we will not bill you for any tier until at least one endpoint relevant to that tier is live for your account.

We make no representation that any specific endpoint will be released by any specific date. The roadmap published at /api-access is forward-looking and may change at our reasonable discretion based on engineering, security or commercial factors.

Available tiers, monthly request quotas, rate limits (requests per minute), feature access matrices and prices are published at /api-access and may be updated from time to time. Founding rates that you accept at signup are locked for one (1) year of GA and one (1) renewal cycle thereafter, after which standard published rates apply.

Your monthly quota is pooled across the entire endpoint surface. Quotas reset on the first day of each calendar month at 00:00 UTC. Unused quota does not roll over and is not refundable in cash or credit. Where you exhaust your quota, we may return HTTP 429 with a Retry-After header until the next reset, until you apply an a-la-carte top-up, or until you upgrade your tier; we will not auto-charge any overage without your explicit consent.

We enforce per-tier rate limits to protect the Service for all customers. Bursts that exceed your tier's published RPM are throttled with HTTP 429 responses. Sustained abuse, automated scraping outside of declared use cases, or any pattern we determine in good faith to harm Service stability may result in immediate suspension of API credentials at our sole discretion, with no refund or credit.

All API outputs — including but not limited to extracted design tokens, palettes, typography rules, motion timings, brand-tier flags, accessibility audits (WCAG and APCA), curator briefs, agent prompt kits, theme generators and starter scaffolding — are provided on an "AS IS" and "AS AVAILABLE" basis as best-effort engineering output. Curator notes are editorial opinion, not legal, accessibility, contractual, medical or compliance advice.

You are responsible for verifying any output before relying on it for production, regulatory, accessibility or commercial decisions. We expressly disclaim all warranties of accuracy, completeness, fitness for a particular purpose and non-infringement to the maximum extent permitted by applicable law.

During private beta, the API operates on a best-effort basis with no guaranteed uptime. We may schedule downtime, throttle, suspend, or rollback any endpoint at any time without prior notice in order to maintain Service stability. No refund, credit or other compensation is offered for downtime, throttling, or endpoint changes during the beta period.

After general availability, the Enterprise tier (and only the Enterprise tier) includes a 99.9% monthly uptime commitment as documented in the executed Enterprise order form. Uptime is measured against successful HTTP 2xx responses for the production /v1/extract/* and /v1/inspiration/* endpoints, computed monthly, and excludes (a) scheduled maintenance announced at least 48 hours in advance, (b) any force majeure event as defined in /legal/terms section 14, (c) any failure caused by Enterprise customer's own infrastructure, integration code, credentials misuse, or violation of these API Terms or the Acceptable Use Policy, (d) any failure of any third-party CDN, DNS, hosting, captcha or anti-bot service that we do not directly operate, and (e) the first thirty (30) minutes of any single incident. Where measured uptime falls below 99.9% in a calendar month, the sole and exclusive remedy is a service credit equal to ten percent (10%) of that month's prorated Enterprise fee, applied to the next renewal invoice. Service credits are not redeemable as cash, are not transferable, and lapse if the Enterprise order is not renewed. No tier other than Enterprise carries any uptime commitment.

Token history retention varies by tier as published at /api-access (currently: Maker 7 days, Builder 30 days, Studio 365 days). At the end of the applicable retention window, your historical token snapshots are permanently deleted from our active systems, with no recovery option. Backups for disaster-recovery purposes are kept for up to 30 additional days and are not user-accessible. Enterprise customers may negotiate custom retention windows in the executed order form.

You are responsible for exporting and storing any extracted output that you wish to retain beyond the applicable retention window. We are not responsible for output you fail to export.

We reserve the right, at our sole discretion and without prior notice, to suspend or terminate your API access for any breach of these API Terms, any Acceptable Use Policy violation, any chargeback, any threat to Service stability, or any conduct we consider harmful to the Service or other Users. Upon termination of API access, any outstanding paid amounts remain non-refundable and your right to call any endpoint immediately ceases.

Liability under these API Terms is governed by, subordinate to, and aggregated into the single combined cap set out in /legal/terms Section 11. The cap defined there is calculated across all paid offerings of the Service — sponsorships, licences, API access, AI-generated outputs and any successor offering — taken together; purchasing API tiers does not multiply the cap, and nothing in these API Terms increases or supplements the cap.

All fees and amounts paid to BEY AGENCY LTD for the Service — including but not limited to sponsorship fees, license fees, and subscription fees — are final and non-refundable, except where expressly required by applicable mandatory law. No partial, pro-rated or full refund is offered for unused time, cancelled sponsorships, editorial disagreements, feature changes, service downtime or dissatisfaction.

We may update this document from time to time to reflect changes in our services, applicable law, or business practices. When we make material changes, we will update the effective date at the top of the document. During early access, we notify active email subscribers of material changes. After general availability, registered users will additionally receive in-product and email notification. Continued use of the Service after an update constitutes acceptance of the revised document.

This document, and any dispute or claim (including non-contractual disputes or claims) arising out of or in connection with it or its subject matter or formation, shall be governed by and construed in accordance with the laws of England and Wales. The parties irrevocably submit to the exclusive jurisdiction of the courts of England and Wales.

Questions about these API Terms may be sent to hi@bey.agency or hello@ai2.design.