# HEIP > HEIP tells your server whether a real person is actually doing an important action on your website, like a payment, a login or a withdrawal, without asking them to solve a puzzle and without learning who they are. HEIP (Human Entropy Integrity Protocol) is a proof-of-human check for important actions on the web. It answers one question at the moment an action happens: is a real person at the controls right now? It does not identify the person, show a puzzle, or collect biometrics, typed text or personal data. It works alongside login, MFA, KYC and bot management. Integration in one paragraph: the page loads the browser SDK with a public site key (pk_live_…) and calls HEIP.prepareAction("") when a protected form is submitted. It sends sessionId and actionNonce to your server, which calls verify with its secret key (sk_live_…, server only) before committing the action. The answer is ALLOW, STEP_UP or DENY, with reasons. The action token works once. The server SDK fails closed. Pricing: Free covers 10,000 decisions a month with no card. Enterprise is a custom annual contract. ## Product - [Why human presence?](https://heip.dev/why-human-presence): The idea, with everyday examples: knowing who is not knowing whether. - [How it works](https://heip.dev/how-it-works): The flow, what HEIP sees and never collects, policies and how mistakes are handled. - [Security](https://heip.dev/security): Data received and never received, retention, keys and responsible disclosure. - [Pricing](https://heip.dev/pricing): Free and Enterprise plans, what counts as a decision, pricing FAQ. - [Enterprise](https://heip.dev/enterprise): Guided onboarding, security review support and procurement. ## Use cases - [HEIP for digital assets](https://heip.dev/use-cases/digital-assets): Exchanges, brokers, wallets and custodians - [HEIP for fintech and payments](https://heip.dev/use-cases/fintech): Payments, neobanks, lending and payout platforms - [HEIP for AI agents](https://heip.dev/use-cases/ai-agents): AI products and agent platforms that act for users - [HEIP for online platforms](https://heip.dev/use-cases/online-platforms): SaaS, communities and consumer apps - [HEIP for crypto and Web3](https://heip.dev/use-cases/crypto-web3): dApps, token launches, airdrops and NFT mints - [HEIP for iGaming](https://heip.dev/use-cases/igaming): Casinos, sportsbooks, poker and skill games - [HEIP for online marketplaces](https://heip.dev/use-cases/marketplaces): E-commerce, classifieds, rentals and resale - [HEIP for ad tech](https://heip.dev/use-cases/ad-tech): Ad networks, publishers, affiliate and lead-gen platforms - [HEIP for ticketing](https://heip.dev/use-cases/ticketing): Event ticketing, box offices and resale ## Docs: Getting started - [Overview](https://heip.dev/docs): What's in the docs, and how the API is versioned. - [Quickstart](https://heip.dev/docs/quickstart): From a new site to your server's first decision, in observe mode. - [Build with AI](https://heip.dev/docs/build-with-ai): A prompt and project rules for coding agents. ## Docs: Reference - [Server API](https://heip.dev/docs/api): The calls your server makes, their responses and errors. - [SDKs](https://heip.dev/docs/sdk): The browser SDK and the Node.js server SDK. ## Docs: Trust - [Security](https://heip.dev/security): Keys, data, retention and disclosure. - [Privacy](https://heip.dev/privacy): What HEIP collects and never collects. ## Blog - [Before AI agents touch money: ten questions for banks, fintechs and exchanges](https://heip.dev/blog/ai-agents-money-movement-checklist): A practical checklist drawn from the bank principles paper, OWASP, NIST and the October 2026 bank responses. Work through it before deciding how your platform treats customers’ AI agents. - [Amazon v. Perplexity: what the Ninth Circuit’s ruling means for platforms deciding which agents to let in](https://heip.dev/blog/amazon-perplexity-ruling-ai-agents-platform-access): In August 2026, an appeals court lifted an injunction that had barred Perplexity’s Comet agent from shopping on Amazon. The ruling is narrow. Its practical message for banks and platforms is not. - [Deepfakes broke “seeing is believing”. Agents are breaking “logged in means present”.](https://heip.dev/blog/deepfakes-agents-identity-checks): Generative AI has already undermined two assumptions banks relied on: that a face on a call is the person, and that a document is real. AI agents now undermine a third. A look at the evidence, from FinCEN to the FBI. - [NIST SP 800-63B-4 and the minutes after login: what the standard now says about session monitoring](https://heip.dev/blog/nist-800-63b-4-session-monitoring): The 2025 revision of NIST’s authentication guidelines names session monitoring as a way to reduce risk after sign-in, and attaches a privacy warning to it. A plain-language reading for teams deciding what to do about agents in valid sessions. - [The OWASP Top 10 for Agentic Applications, read by a fraud team](https://heip.dev/blog/owasp-agentic-top-10-for-fraud-teams): OWASP’s agentic Top 10 was written for people building agents. Four of its ten risks are also a fraud team’s problem the moment a customer’s agent signs in. Here is the list, translated. - [Half of organisations saw an AI agent do something nobody asked for: inside the 2H 2026 agentic security survey](https://heip.dev/blog/half-of-organisations-ai-agent-unintended-action): Salt Security’s survey of more than 380 professionals found 49.8% had confirmed or suspected an unintended agent action, and only 17.5% could stop a misbehaving agent automatically. What the numbers measure, and what they mean outside the enterprise. - [Bots are now most of the web: reading the 2026 Bad Bot Report for financial services](https://heip.dev/blog/bad-bot-report-2026-financial-services): Thales’s 2026 Bad Bot Report found 53% of internet traffic was automated in 2025, with financial services taking a quarter of bot attacks and nearly half of account-takeover attacks. Here is what the numbers do and do not say. - [Prompt injection meets online banking: what agentic-browser research tells banks](https://heip.dev/blog/prompt-injection-agentic-browsers-online-banking): An AI agent inside a logged-in banking session takes instructions from whatever it reads. Security researchers have shown how that goes wrong, and the companies building these browsers say it will not be fully fixed. - [Web Bot Auth, Trusted Agent Protocol, Agent Pay, AP2 and Poppy: what agent protocols prove, and what they leave open](https://heip.dev/blog/agent-identity-protocols-human-presence): A year of agent protocols has given the web ways to name an agent, scope what it may do and record what the user asked for. One question is still left to each platform: was a person there when it mattered? - [Building Trust in Agentic Commerce: a close reading of the six-bank principles paper](https://heip.dev/blog/six-banks-agentic-commerce-principles): Bank of America, Capital One, ING, NatWest, CommBank and ASB published shared principles for AI agents that buy things. The paper is short and non-binding. Its specifics are more pointed than the coverage suggested. - [Read-only is the new default: why banks are splitting “see” from “do” for AI agents](https://heip.dev/blog/read-only-ai-agents-banking): JPMorgan, Citi and Schwab all landed in the same place in October 2026: agents may look, but not touch. The split is sensible. It is also temporary, because the useful part of an agent is the touching. - [You gave an AI agent your bank password. Is what it does “unauthorised”?](https://heip.dev/blog/regulation-e-ai-agents-shared-credentials): US consumer protection for bank transfers turns on one word: authorised. Credential-sharing with AI agents puts that word under strain. Here is what Regulation E and the CFPB’s guidance actually say. - [Unvetted AI agents inside US bank accounts: what JPMorgan, Wells Fargo and four other firms have said](https://heip.dev/blog/us-banks-unauthorised-ai-agents-customer-accounts): In one fortnight, six of the largest US banks and brokers said they are blocking or restricting outside AI agents. Here is what each one said, what is confirmed, and what is still unknown. - [Wells Fargo’s new AI-agent terms, clause by clause: who pays when an agent gets it wrong](https://heip.dev/blog/wells-fargo-ai-agent-terms-liability): From 24 November 2026, Wells Fargo’s online access agreement treats an AI agent’s instructions as the customer’s own. We read the new section line by line and ask what it settles, and what it leaves open. - [Why human presence assurance is now a required control](https://heip.dev/blog/why-human-presence): For twenty years, “logged in” meant “a person is here”. That assumption has quietly stopped being true. Here's what has to replace it. ## Optional - [Full context for AI tools](https://heip.dev/llms-full.txt): Key facts, FAQs and use cases in one plain-text file. - [Developers](https://heip.dev/developers): Overview, and a prompt for coding agents. - [About](https://heip.dev/company): Why HEIP exists and its principles. - [Status](https://heip.dev/status): Live status of the API, SDK and dashboard. - [Contact](https://heip.dev/contact): Book a demo, talk to sales, report a security issue. - [RSS feed](https://heip.dev/feed.xml): New posts.