Your students will ask an AI about tonight’s reading. Ours has to show them the page.

Humy is a web-based educational platform — successor to Hello History — that lets students hold in-depth conversations with 1,200+ AI historical figures. A figure can be given an archive of curated primary sources and scholarly references, which it searches during the conversation and quotes with a reference back to the document students can open.

Humy runs on the same frontier models as any AI tool. The difference is where the answer comes from: a Humy simulation searches an archive your team curates, not the open web. When it cites a passage, our system copies the words straight out of your document and checks them letter by letter. The AI never types a quote itself, so a quote cannot be made up, and students can open the source and read the typed-out text beside the scanned page.

Yes. Humy works with Canvas, Blackboard, and Moodle. You can embed it in course pages via iframe, share a direct link, or print a QR code for lecture hall access. SSO with school accounts is available on request, via Auth0 by Okta.

Five: Interview (live conversation with a figure), Debate (up to ten figures at once), Simulation (students role-play inside a historical moment), Open Questions (short answer / SAQ), and Essay (DBQ, LEQ, FRQ, concept application). Each assignment is generated, editable, and auto-graded against a grading guide you write. Source grounding applies inside assignments too, so the same evidence discipline holds when the work counts toward a grade.

Every student conversation is logged in full. Faculty can review transcripts to verify originality. You can also disable copy-paste, cap the number of attempts per question, and require named access. Student data is never used for AI training.

That is the most common way faculty use it. You load the readings you actually assign — the set text, the article on the week's topic, the documents in the course reader. The figure answers from those and quotes the page a student is meant to have read. Because a version carries its own instructions and reading level, the same reading list can support a Socratic tutor for a survey course and a harder interlocutor for a seminar. A professor can set this up alone, without a department decision.

Either. A single instructor can start free and upgrade a personal account at $129 per year, and departments and institutions can license centrally with pricing scoped to faculty and student volume. A growing number of courses run the other way round: the professor decides the tutor is part of the course and students take their own subscriptions, the way they would buy a set text. Ask us on a call and we will lay out both.

Yes. Upload biographies, primary sources, transcripts, images, and video, and attach up to 50 sources to a single figure. Scans without a text layer are transcribed page by page, including historical handwriting, up to 1,000 pages per source; transcription uses credits and you approve an estimated cost first. Each source can stay private to your account or be shared across your whole institution. You can also publish multiple versions of the same Humy — a graduate-seminar edition and an intro-survey edition — each with its own reading level, instructions, and language.

That is what our university pilots do. Three institutions are piloting collection-based figures now: National Louis University (Chicago), the University of Wisconsin–River Falls, and the Private University of Education, Diocese of Linz in Austria. At National Louis we built an AI simulation of the university's founder, Elizabeth Harrison, from 25 of her writings, photographs, and documents supplied by the university library. These are built with our team rather than self-serve — if your library holds a collection worth teaching from, email hello@humy.ai and we will scope a pilot with you.

Flexible — based on faculty and student volume. We work with individual departments and with institution-wide deployments. Contact us for a quote scoped to your needs.

Yes. Students can access Humy through anonymous links — no accounts, emails, or passwords required. Where you do require named access, FERPA and GDPR controls apply. Source retrieval runs on our own servers, so student messages are not sent to an additional third-party service to query search indexes. Student data is never used for AI training.

Yes, and it is one of the reasons institutions widen from a single department. A guide made from your prospectus, your aid explanations and your own archive answers a prospective student at the hour they are actually reading, shows the passage it used, and says plainly when your material does not cover the question. It can also introduce the institution through the people and the moments that made it, which a support bot cannot do. Be aware that admissions is a newer deployment for us than teaching, so ask what we have run and what we have not.

No. It can be a person, an event, or a place, and all three are built the same way: a name, a short description, and the documents you attach. A seminar can question the figure on the reading list, step inside the convention or the trial the week turns on, or ask about the site an excavation report describes. Historical figures are the best-known case and what our ready-made library of 1,200+ is stocked with, but they are one case, not the limit.

Yes. Individual faculty can start with a free instructor account. When you're ready to expand to your department, our team will help with onboarding, custom Humies, and a quote.