Build Surveys Directly in Claude and ChatGPT: BlockSurvey Keeps Answers Hidden from AI

AI Assistant Becomes a Survey Factory—But Answers Stay Secret

BlockSurvey, a new product ranked #18 on Product Hunt, tackles a seemingly contradictory need: let Claude, Claude Code, or ChatGPT help you build surveys, while the AI assistant never sees respondents' answers.

The core logic is straightforward: you describe the survey you need in the chat window, and the AI returns a live link you can use. Want to change a question? Just keep the conversation going. All responses are end-to-end encrypted, and the AI side never touches the data. For users who need to launch research quickly and are sensitive about data privacy, this workflow hits a real pain point.

From a product positioning standpoint, BlockSurvey isn't trying to build a smarter survey generator. Instead, it positions AI as a "survey architect"—responsible for building structure, not participating in data flow. This separation of roles is key to understanding the product.

Conversational Survey Building: Editing Questions Like Editing Copy

The traditional survey tool workflow is: log into a dashboard, drag and drop question types, set up logic, publish a link. BlockSurvey tries to compress the first half into a single natural-language description.

According to its Product Hunt page, after connecting BlockSurvey to Claude or ChatGPT, users describe their research goals directly in the conversation, and the AI generates a survey and returns a live link. If the first version of questions isn't quite right, there's no need to switch to another interface—just say "change question three to multiple choice" in the chat to make the edit.

The value of this interaction model lies in lowering the startup cost. For ad-hoc user research or event feedback collection, users don't need to learn a new tool or leave the AI conversation environment. The Claude Code integration hints at developer scenarios—for example, generating a developer experience survey while coding.

However, the product has disclosed limited information so far, making it impossible to judge whether detailed capabilities like survey logic branching and question-type richness match those of professional research tools.

Claude Code is a command-line tool from Anthropic aimed at developers, embedding Claude's capabilities into the terminal environment and supporting direct AI invocation during code editing, debugging, refactoring, and other development workflows. Unlike the web version of Claude, Claude Code runs as a CLI, can read local codebases, execute commands, and respond to natural-language instructions. BlockSurvey's integration with Claude Code means developers can generate surveys without leaving their coding environment—for example, quickly creating a developer experience feedback form during development. This integration also reflects how AI tools are evolving from standalone apps toward "embedded workflows," where AI capabilities are integrated into users' existing toolchains via APIs or plugins rather than requiring users to switch context.

End-to-End Encryption Is Both the Selling Point and the Technical Challenge

"AI can't see answers" is BlockSurvey's most prominent differentiator. The official description states that responses are end-to-end encrypted and inaccessible to the AI assistant.

This design addresses a reasonable concern: if a survey is generated by AI, will the data also flow through AI service providers? In most AI-integrated products, data passing through third-party services is the norm. BlockSurvey chooses to isolate response data from the AI channel, meaning encryption and decryption happen between the user and BlockSurvey, while the AI is only responsible for generating survey structure.

From a privacy engineering perspective, this architecture requires clearly distinguishing two data pathways: "at generation time" and "at collection time." The product hasn't disclosed its specific encryption implementation, but the end-to-end encryption promise alone sets it apart from similar products that "feed survey data to large models for analysis."

One caveat: currently the only source of information is the product's self-description on its Product Hunt page, and no independent third party has verified the encryption mechanism.

End-to-end encryption (E2EE) means data is encrypted at the sender and decrypted only at the receiver, so that no intermediate node (including service providers and third parties) can read the plaintext. In the survey context, E2EE means respondents' submitted answers are encrypted in the browser, and only the survey creator holding the key can decrypt and view them. This is fundamentally different from traditional "transport-layer encryption" (such as HTTPS): the latter only protects data from eavesdropping during transmission, but the server can still access plaintext. BlockSurvey's claim that the AI assistant cannot access answers essentially excludes the AI service provider from key holders, keeping survey structure generation and response data collection on two independent pathways. However, the implementation details of E2EE (such as key management and whether multi-device decryption is supported) directly affect its security strength, and the product has not disclosed this information.

Ranking and Feedback: Early Signals Positive but Sample Too Small

BlockSurvey received 10 upvotes and 1 comment on Product Hunt, ranking #18. This data shows the product got some attention on launch day, but the sample is too small to judge market acceptance.

A score of 10 votes is below average on Product Hunt, which may reflect several factors: privacy-focused surveys are a vertical need with a narrower audience than general-purpose AI tools; the product is newly launched with limited community awareness; or the product demo wasn't compelling enough.

A single comment also can't constitute meaningful user feedback analysis. For users considering trying it, we recommend visiting the official website directly and trying the free tier rather than relying on early voting data.

Product Hunt is a global platform for launching and discovering new products. Users can upvote and comment on products launched that day, and the platform generates a daily leaderboard based on vote counts. Ranking #18 means a mid-to-upper position among products launched that day, but Product Hunt's voting mechanism is heavily influenced by launch timing, community mobilization, and category popularity. 10 votes is a relatively low magnitude—popular products typically get hundreds or even thousands of votes, while vertical tool products often hover around a few dozen. Therefore, a single-day ranking and vote count can only serve as a rough signal of early attention, not a direct equivalent of product quality or market potential. For early-stage products, the more reliable evaluation method is still to try the free tier yourself and check feature completeness.

Who Should Use It, and Who Can Wait

Based on disclosed features, the product suits three scenarios: teams that need to launch internal research quickly, organizations with hard requirements on respondent data privacy, and users already accustomed to working within Claude or ChatGPT.

Scenarios where we'd hold off for now include: large research projects requiring complex logic branching and professional analysis reports; and workflows that need to feed survey data directly into AI for insight analysis—which fundamentally conflicts with the product's privacy design.

BlockSurvey's attempt reveals a product direction worth watching: AI assistants can become the entry point to the toolchain without necessarily becoming part of the data chain. This division—"AI touches structure, not content"—could become a reference model for privacy-sensitive AI integrations.

Whether the product ultimately gains traction depends on the completeness of the survey features themselves and whether the end-to-end encryption promise can withstand technical scrutiny. The current information is enough to keep people watching, but not enough to draw a conclusion.

One-line Summary

BlockSurvey lets AI generate surveys through conversation, but responses are end-to-end encrypted and invisible to AI—exploring privacy-sensitive AI integration.

Paragraph Summary

BlockSurvey is a new product ranked #18 on Product Hunt that lets users generate surveys and get live links by describing them in natural language within Claude, Claude Code, or ChatGPT, with edits also done through conversation. Its core differentiator is end-to-end encryption of response data: the AI assistant only generates survey structure and cannot access respondents' answers, isolating the "generation" and "collection" data pathways. The product suits quick research and privacy-sensitive scenarios, but not workflows requiring complex logic branching or feeding responses into AI analysis. It received 10 upvotes and 1 comment early on—too small a sample to judge market acceptance, and the encryption mechanism has yet to be verified by third parties.