unquestion

unquestion is an AI-powered conversation builder designed to replace static forms with adaptive dialogue. A creator arranges questions, publishes the conversation through a public URL or embeds it on a website, and the AI can answer conversationally and ask follow-up questions while the platform records the resulting answers in structured tables.¹²

The product targets workflows such as customer feedback, newsletter signup, event registration, demo qualification, cancellation surveys, feature requests, waitlists and customer support.¹³ Results can be monitored as complete or partial responses and exported as CSV.

The core product idea is coherent: traditional forms are predictable but rigid, while open chat is flexible but produces messy transcripts. unquestion tries to combine adaptive dialogue for the respondent with structured rows for the operator.

The operator does not have to turn every conversation into a clean record manually. The platform is explicitly designed to preserve structured answers, real-time submission visibility and CSV export.¹

Trend Hunter’s exact-product coverage describes the same thesis: adaptive follow-ups can gather more detail than a static form while still returning structured data rather than an unbounded chat transcript.⁴ That confirms product positioning, not comparative reliability or conversion performance.

The Terms state that AI inference is provided by Cerebras and that the provider may change.² The Privacy Policy says inputs submitted to conversational forms are sent to Cerebras for inference.⁵

Cerebras’s own current cloud/privacy material says service content is not used to train or fine-tune models under the relevant arrangements and describes limited retention behavior.⁶ This is useful provider-boundary evidence, but it does not reveal one fixed foundation model that can be benchmarked as “the unquestion model.”

The Terms explicitly warn that published responses are generated by AI without human review and may be inaccurate, false, inappropriate, biased, inconsistent or outdated.² The creator is responsible for testing conversations, monitoring behavior and adjusting prompts before and after publication.

That makes the product inappropriate as an unsupervised source of high-stakes legal, medical, financial or other professional advice. The Terms expressly state that AI output is not professional advice.²

unquestion provides an OAuth-authorized MCP server through which compatible third-party AI clients can interact with account data.² Once that information is transmitted to an authorized third-party client, that client’s privacy, retention and security rules also apply.⁵²

For organizations, MCP access should therefore be treated as a privileged integration rather than a convenience toggle.

unquestion does not use a recurring subscription model. The Terms say credits are sold in one-time packs, with one credit consumed per conversation.²

Credits are non-transferable, have no cash value and expire 12 months after purchase. Fees and credits are generally non-refundable except where law requires otherwise.² The homepage says a first conversation can be created for free.¹

The credit model is attractive for intermittent use because there is no monthly commitment. However, the public pages reviewed did not expose current pack prices, so total value cannot be compared cleanly with form-builder subscriptions. Buyers should also verify how partially completed sessions and retries are counted before using the tool at scale.

Published conversations can be accessed through a unique public URL without authentication until the creator unpublishes them, and they can be embedded on external sites.²

The Privacy Policy distinguishes roles clearly: Unquestion LLC is controller for account and operational data, while the customer is controller for end-user responses collected through the customer’s published conversations and unquestion acts as processor.⁵

The Terms prohibit collection of HIPAA-regulated health information, payment-card information, Social Security numbers or government IDs, certain data from children, and data the customer is not legally entitled to collect.² This materially narrows the set of sensitive-intake use cases for which the product should be considered.

The Privacy Policy identifies Unquestion LLC, a Wyoming LLC, as operator and lists account information, conversation designs/system prompts, respondent text, interaction metadata, IP/device information, analytics and diagnostics among the data categories.⁵

It names Supabase, Stripe, Loops, PostHog, Langfuse and Upstash among key subprocessors and states that all conversational inputs are sent to Cerebras for inference.⁵ It also says PostHog operates in cookieless mode and the service does not use advertising cookies or cross-site behavioral tracking.

Account information is deleted immediately after account deletion, while conversation data and responses are deleted within 90 days. Backups and logs may remain for up to 180 days, while billing records can be retained as legally required.⁵²

The policy also provides data-access, correction, deletion, export and objection mechanisms.⁵

The Privacy Policy describes HTTPS/TLS, authentication, access controls, monitoring and webhook-signature verification.⁵ The Terms do not promise uninterrupted availability and permit rate-limit changes.²

No mature public assurance layer such as a completed SOC 2 or ISO certification was identified in the reviewed material. That is an evidence/assurance limitation rather than proof of an implementation flaw.

Trend Hunter provides exact-product editorial coverage but no hands-on score, comparative completion study or user-review population.⁴ There is still little independent evidence on follow-up quality, extraction accuracy, respondent completion rates, latency or reliability.

unquestion is best suited to small teams, founders, product managers and marketers who want an adaptive alternative to a static survey or form while still receiving structured results.

Good candidate workflows include customer feedback, feature requests, demo qualification, cancellation interviews, event registration, waitlists and low-risk support intake.¹³

It is a weaker fit for regulated or highly sensitive intake, high-stakes advice, workflows that require deterministic question wording, or enterprises that cannot adopt a vendor without stronger security-assurance artifacts.

Before publishing a conversation, teams should test representative and adversarial inputs, inspect follow-up behavior, provide appropriate respondent notices, avoid prohibited sensitive data and review any authorized MCP client separately.