Questions & Answers
What SpanNova is, and what it isn’t
The questions families actually ask — including the ones about what happens to their child’s information.
What is SpanNova?
SpanNova turns what you already know about a learner with special needs — plus their IEP, ISP, medical records and evaluations — into a portable AI profile. It is a structured file that describes how that person communicates, what helps them, what sets them off, and what they are working toward. You keep the file and hand it to whichever AI tool you use, so the tool understands your learner from the first message instead of starting from zero.
What makes SpanNova different from just using ChatGPT?
Nothing stops you from explaining your child to ChatGPT. The problem is that you have to do it again in every new chat, on every tool, forever — and under pressure you leave things out. SpanNova is not a competing chatbot. It builds the context those tools are missing, once, in a structured form, so the explaining is done and it travels with your child.
Is SpanNova a chatbot my child talks to?
No. SpanNova never talks to your child. It produces a document you give to the AI tools your family already uses. That is a deliberate design decision: an AI that talks directly to a vulnerable learner carries risks we are not willing to take on, and it would mean asking families to adopt yet another app.
How does SpanNova approach New York's new law on AI chatbots and kids?
New York's new law (S.9051B/A.10379C, signed October 2026) bars companies from offering unsafe AI chatbot features to anyone under 18 — such as a chatbot that pretends to be human, claims a relationship with a child, or encourages self-harm. We treat laws like this as a floor, and SpanNova is built to stay well clear of what they are trying to prevent.
First, SpanNova is not a companion chatbot and never talks to your child. It writes instructions you give to the AI tools your family already uses, and the companies that run those tools have their own obligations under the law.
Second, every child profile tells the AI to hold the same lines the law draws: say plainly that it is an AI and never claim to be human or to have feelings; never claim to be your child's friend, teacher, or any other authority figure; not ask about feelings unprompted or bring private health details into later chats; never use flattery or guilt to keep them chatting; never ask them to keep secrets or discourage them from going to you, trusted adults, or professionals for help; never produce sexual content; and never help with suicide, self-harm, disordered eating, or drugs, pointing to a trusted adult and 988 instead.
Third, the instruction the AI receives does not name your child's diagnosis.
No instruction can guarantee how another company's AI behaves, so stay involved in how your child uses it. This explains how we approach the law; it is not legal advice or a claim of certified compliance.
What do I actually get at the end?
It depends on which one you build. Quick Start, the free one, gives you a system instruction you copy and paste into Gemini, ChatGPT, Copilot, Perplexity (13 and older), Claude or Grok (for learners 18 and older), or a school Chromebook. The Complete Profile, with a beta invitation or a license, gives you a file you download — a structured profile containing a complete system instruction plus the underlying profile data — that you paste or upload into the same tools. Either way it is yours: no subscription is required to keep using a profile you have already made.
Does SpanNova store my child's information?
No. A Complete Profile is saved as a file you download to your device. A Quick Start profile is not saved anywhere: it appears on screen for you to copy, and it is gone when you leave the page, so paste it somewhere you will keep it. SpanNova does not operate a database of children's health or education records. Signing in stores three things: your email, which plan you are on, and a random ID per learner slot. The profile itself is never retained on our servers. Separately, if you ask for beta access we keep what you typed into that form (your name and email, plus your role and whether you want launch updates if you chose to answer) and which link brought you to it, so we can contact you — contact details you chose to give us, never a child's record.
If nothing is stored, how does the AI generate the profile?
The answers you type are sent to Anthropic's API to generate the text, the same way any AI product works, and the result comes back to your browser. SpanNova does not keep a copy. Anthropic automatically deletes API data within 30 days and does not use it to train its AI models. The one exception worth stating plainly: content that Anthropic's automated systems flag for trust and safety review may be kept longer under Anthropic's usage policies.
Is SpanNova HIPAA compliant?
SpanNova is not a covered entity under HIPAA, so 'compliant' is not the right word and we will not claim it. What is true: the product is built around the same categories of sensitive health information HIPAA protects, and handled with that seriousness — medication timing and safety context flow into the profile you export and are not retained afterward.
Do I need an account? Is it free?
Quick Start is free, needs no account, and takes about five minutes — it produces a starter profile. The Complete Profile, which parses IEP and ISP documents and builds the full lifespan-aware profile, requires an account plus either a beta invitation or a subscription — it is in limited beta and free for testers while that runs. Individual covers one learner; Family covers three.
Can I use SpanNova if my child doesn't have an IEP?
Yes. An IEP or ISP speeds things up because SpanNova can parse it, but it is not required. If you have no documents ready, you answer questions about diagnoses, communication, interests, safety needs and goals instead. Plenty of families arrive before a formal evaluation, or between them.
Does it work for adults, or only school-age children?
Both. The same profile engine spans IFSP through IEP through ISP — early intervention, school age, and adulthood. There is a separate adult path written for self-advocates and for the caregivers and day programs supporting them. Most tools in this space stop working the moment a child ages out of school; that gap is the reason SpanNova exists.
Does SpanNova give medical advice or diagnose anything?
No. It does not diagnose, does not recommend treatment, and does not generate clinical conclusions. It describes a learner using information you provide and documents you already have, so that an AI tool can respond appropriately. It is a context layer, not a clinician.
Can a teacher, aide or therapist use a profile?
Yes, and that is often the point. A family exports the profile and shares it with the people who work with their learner, so a new aide in September starts with the context that took years to build rather than a blank page.
What if I'm not comfortable putting my child's information into AI at all?
That is a reasonable position and worth taking seriously. Two things may help: the profile lives on your device rather than in our database, and the sensitive sections are explicit and reviewable before anything is generated, so you see exactly what leaves your hands. If that is still not enough, Quick Start asks for far less and requires no account.
Still deciding?
Quick Start is free, needs no account, and takes about five minutes.
Build a profile →