Meet Sage: An AI Tutor Built on Verified Credibility

Sage is Tutorwise's free AI tutor, and the thing that separates it from a general-purpose chatbot isn't the model underneath — it's the marketplace it lives in. Every human tutor on Tutorwise carries a computed, checkable credibility score rather than a self-written bio, and Sage is held to that same bar: it answers from a curated knowledge base, tracks what a student has genuinely learned, says when it isn't sure, and hands off to a verified human the moment a screen can't do the job. If you're deciding whether to trust an AI tutor with your child's education, that's the real question — not whether the AI sounds fluent, but whether you can check what's behind it.

Why an anonymous AI tutor is a problem

A general chatbot is a genuinely useful homework tool. Ask it to work through a quadratic, translate a paragraph of French, or summarise a set text, and it will oblige in seconds. The trouble surfaces only when you try to check its working. You don't know what it was trained on, whether its explanation lines up with the exam board your child actually sits, or what happens the moment it's confidently wrong. Nobody stands behind the answer.

That's tolerable for a one-off homework question. It's a different problem across a school year, where a small, uncorrected error compounds and a fluent wrong explanation can plant a misconception that takes months of a real tutor's time to undo. The traditional fix has been a human you can vet — qualifications, references, a track record you can ask about. Vetting a person properly takes time and access, though, and that's exactly why good tuition has historically gone to families who could afford it or happened to know the right person. According to the Sutton Trust's 2023 polling on private tuition, roughly three in ten young people in England have had private tuition at some point — which means most families haven't, often for reasons of cost and access rather than lack of need.

An AI tutor solves the access half of that. It doesn't automatically solve the trust half — you can now get an instant answer, but you still have no way to check who or what produced it. Sage is built specifically to close that second gap, not just the first.

What "verified credibility" actually means here

Sage sits inside Tutorwise's own trust layer, Credibility as a Service (CaaS), and it's worth being precise about what that system does, because a generic chatbot has nothing equivalent to point to.

On most tutoring directories, the tutor writes their own profile: their claimed qualifications, their chosen framing of a star rating, and the parent is left taking their word for it. CaaS replaces that self-description with a computed score, built across six weighted categories of real, checkable signal: Delivery — the quality of tuition actually delivered — carries 40% of the score, Credentials 20%, Network 15%, Trust 10%, Digital engagement 10%, and Impact 5%. Delivery sits heaviest deliberately: what a tutor actually does for a learner matters more than what they say about themselves.

Verification isn't a badge a tutor can buy — it's points earned inside the Trust category. A completed enhanced DBS check, a verified identity, a finished onboarding flow: each adds to the score. Underneath all of it sits a hard gate — nobody gets a credibility number at all until they've either verified their identity or completed onboarding, so an unverified stranger has no score to hide behind. The number also recalculates close to real time as new evidence lands, so it reflects who a tutor is now, not who they were when they first signed up.

The upshot: when a parent looks at a Tutorwise tutor's profile, they're not reading a bio the tutor wrote about themselves. They're reading an earned score built from signals the platform has actually verified. That's the standard the word "verified" means on this platform — and it's the standard Sage is built to meet as well.

The engineering that makes Sage checkable

Sitting near verified humans only gets Sage halfway there. The AI agent itself is engineered differently from a general-purpose model, and that engineering is where the trust is actually earned rather than asserted.

It answers from a maintained knowledge base, not raw model memory. Instead of generating from whatever a foundation model happens to have absorbed, Sage retrieves from a knowledge base written and reviewed for the curriculum it's teaching to, then grounds its answer in that material. That's the difference between a tutor working from the actual specification and one working from a half-remembered version of it — and it keeps explanations closer to what a child is really being taught in class.

It models what a student has actually mastered, not what they've merely seen. Sage runs a spaced-repetition mastery model on top of prerequisite chains — you can't be judged secure on simultaneous equations until fractions and rearranging are solid — and it deliberately estimates on the conservative side: the lower bound of what the evidence supports, not the flattering read. Sage would rather under-credit a student's grasp of a topic and keep practising than declare victory early. That caution is intentional; it's how a careful human tutor operates, and it's built into the model rather than left to hope.

It knows when to stop and ask, not bluff. Sage carries a confidence threshold on its own answers. When confidence drops below that bar, the response is to slow down, ask a clarifying question, or route the learner to a human — not to sound fluent regardless. An AI tutor that can admit uncertainty is a materially safer one for a child to rely on than one that's confident and wrong.

It's the only one of Tutorwise's AI agents built and operated under a regulated compliance framework. Sage's design is measured against the UK Department for Education's five-area safety framework for generative AI in education — covering filtering, factual accuracy, data protection, governance, and emotional-influence safeguards. That isn't a marketing line; it's a checklist the product has to answer against, and it's a bar none of Tutorwise's other AI agents currently have to clear, because none of them work directly with children.

Beyond those four pillars, Sage carries features built specifically because it serves under-18 learners: full personal-data redaction with output validation, age-bracket adaptation to how it explains things, wellbeing-signal detection with safeguarding event logging, and prompt-level adjustments across eleven categories of learning difference for SEN/SEND students. None of this makes Sage infallible — no AI tutor is — but it's the gap between a tool engineered to be checked and one that's merely engineered to sound convincing.

Built safe for young learners, not bolted on afterward

Because Sage's users include children, safety isn't a layer added at the end — it's a constraint the product is designed around from the start. Sage is a study companion, not a private messaging channel: it keeps help on the platform, steers explanations back to the curriculum, and escalates anything sensitive to a human rather than improvising a response. When a situation needs more than an AI can responsibly give, Sage's job is to hand off to a verified human tutor, not to hold on and attempt it anyway.

The goal is simple to state: a child gets patient, always-on help with the subjects they're actually studying, and a parent keeps the reassurance that real, verified, accountable people stand behind the platform Sage operates inside.

Sage as the first step, not the whole journey

The most useful way to think about Sage is as the opening move in a chain that can end with a human tutor, not as a replacement for one. Sage is there at eleven at night before a mock exam when no tutor is available. It works through a stubborn proof, explains exactly why an essay lost marks, or drills vocabulary as many times as a student needs, without ever losing patience.

When the sticking point outgrows a single session — a topic that keeps not landing, exam technique that genuinely needs a human eye, the confidence that only comes from a real relationship — Sage points towards tutors whose credibility score you can actually inspect. The move from instant AI help to a verified human happens without leaving the platform and without starting the trust decision from zero: the score has already done the vetting before the introduction is made.

What this means for you

If you're a parent, Sage gives your child curriculum-grounded help on demand, inside a marketplace where the humans are vetted by a system you can inspect rather than a set of self-written claims. If you're a student, Sage is a patient tutor that never sleeps and never makes you feel judged for asking. And when the moment comes to bring in a person, the hardest part of that decision — working out who to trust — has already been done, by a score built from signals the platform verified, not by a stranger's word.

Trust in education shouldn't require a leap of faith. On Tutorwise, it's a number you can inspect, and Sage is built to the same standard as the humans who share the platform with it.

Frequently asked questions

Is Sage free to use? Yes. Sage is Tutorwise's free AI tutor — you can start asking questions and working through topics without paying, and you can use it alongside a paid human tutor when that's what a topic needs.

How is Sage different from a general AI chatbot? A general chatbot answers from whatever its underlying model happens to hold, with nobody accountable behind the answer. Sage answers from a curated, curriculum-grounded knowledge base, tracks what a student has genuinely mastered using a spaced-repetition model, flags when it isn't confident, and sits inside a marketplace of verified human tutors it can hand a student to. It's built to be checked, not just to sound fluent.

Can Sage replace a human tutor? No — and it isn't meant to. Sage is the always-available first step: instant help, curriculum-grounded explanations, and unlimited patience. When a student needs a human — for exam technique, a topic that keeps not landing, or the confidence a real relationship builds — Sage points towards verified tutors whose credibility score is visible.

How does Tutorwise verify its tutors? Every tutor is scored from 0 to 100 by Tutorwise's Credibility as a Service (CaaS) system, across six weighted categories — Delivery (40%), Credentials (20%), Network (15%), Trust (10%), Digital engagement (10%) and Impact (5%). An enhanced DBS check, verified identity, and completed onboarding all earn points inside the Trust category, and no tutor receives a score at all until they've verified their identity or completed onboarding.

Is Sage safe for children to use? Sage operates under the UK Department for Education's five-area safety framework for generative AI in education — covering content filtering, factual accuracy, data protection, governance, and emotional-influence safeguards — and carries features built specifically for under-18 users, including personal-data redaction, age-appropriate responses, and wellbeing-signal detection that escalates to a human when needed.

---

Related reading

• Free GCSE Maths Help: Try Sage, Our AI Tutor
• How Tutorwise Scores Tutor Credibility: CaaS Explained
• What Does a DBS Check Tell You About a Tutor?
• Do Private Tutors Need Insurance? An Honest Guide