You pay for AI where it pays. Nowhere else.

For the person in charge of a piece of work that comes back in a French SME, and who can decide to change it: the owner or managing director, the operations director, a branch manager, a technical director. Most often in a company of 50 to 500 people. The trade matters little: what qualifies is the work, not the sector. You do not need to be the AI specialist: that is the starting assumption, not a gap.

The rung that concerns you depends on what is blocking. A decision you have to make and defend — that is Counsel. A piece of work that is still done by hand and waits on the same two or three people — that is the project. If ten files out of ten are handled by applying the procedure to the letter, or if the method belongs to a software vendor or an accreditation body whose agreement is needed to change it, neither is for you, and we will tell you so.

You describe a piece of work that comes back, or your business if you do not yet know which one to pick. We ask five questions: how often it comes back and how long it takes, who exercises judgment in it, what a mistake costs when it gets through, where your past jobs live — proposals, costings, lessons learned — and in what shape, and who can say yes.

The call ends with a straight answer: the piece of work we would start with, what it costs you today per year, estimated with your figures, and what we would do in your place. No slide deck, no diagnosis to buy, no surprise quote: a quote only comes afterwards, and only if both sides want to continue. The person who would do the work is the one who takes the call.

The twenty minutes cost nothing. The three reference files cost €4,500 excl. VAT: that is the first instalment of the project, and if you continue, it is deducted from the price of the software. The software in place is quoted at the end of those three files. We cannot price it before, because we do not know the work until we have looked at it with you. The follow-up is included: a register that belongs to you.

One signature covers everything, and you can stop at the end of the three files, with no penalty. You keep your annotated files, your three numbers and the plan.

Before we start, we also fix what this work costs you today (real lead times, rework, rejected files) and what will count as a success. Those are the numbers you will follow the software against afterwards. You are buying software in place: it is on your premises or on sovereign infrastructure, it can evolve without us, and there is no subscription to our presence.

Two answers, because they are two different things.

The solution we build for you runs on your premises or on sovereign infrastructure. Your files do not have to leave your company.

We choose each AI model according to what your files contain. What is confidential goes through models installed on your premises or on sovereign infrastructure: you keep control of them, and their cost does not change at a vendor's whim. A commercial model can serve on a step where it does better, provided it sees nothing confidential. It then works on documents that are not confidential, or on files from which we have first removed the names and whatever you want to protect. You decide what goes out, and the record says which model served at each step.

Counsel, the online product, is more exposed and we would rather say so: the text of the decisions you submit to it is analysed in the United States by Google LLC (Gemini API) and Tavily Inc., processors within the meaning of Article 28 of the GDPR, who do not use it to train their models. The rest — account, reading, billing — is hosted in France with OVH and does not leave the European Economic Area. The full list of subcontractors is in our privacy policy.

L'Ordre du jour (free) is the monthly edition: what AI puts on your table, by function, kept for what you will have to decide — thousands of articles read, a dozen kept, and each edition names what it set aside. One-to-many.

Counsel (Solo) is anchored on the decision: you describe an open decision, and each generation brings up the published evidence in the field, ranked by its direct bearing on your situation. Every claim points to a word-for-word passage of the source article, and each item is honestly labelled as direct, adjacent or tangential to what you are deciding. It reads like a consultant's synthesis, not a news feed.

Because it is not content we deliver to you: it is an argued position on your situation, which cites a real source for every claim, and which is re-argued when the facts move. If it does not pay for itself from the first decision it changes, it is not made for you, and you have fourteen days to find out.

As many as you really need. No quota counter, no charge per decision. We size capacity for the typical Solo user (1 to 5 active decisions); if you run more in parallel, the Team plan gives you shared context across seats and is built for that load.

You archive it in your folder and free up mental space. The Counsel history stays — you can reopen any past decision, export it as Markdown for your team, or re-argue it if the context changes. No billing change either way; Solo is a single flat price.

No. Refreshing is unlimited — your LLM usage is not metered. We made that choice so you never get a nasty surprise on the bill; one price, all decisions, all refreshes.

A one-off question-and-answer surface that complements your Counsel anchored on the decision. You ask a question — “what does the evidence say about putting client files through an AI tool?” — and Counsel answers from the public corpus over a 180-day window, with word-for-word citations and an honest confidence level (high / medium / low). Different from your daily Counsel: no stored context, no decision required, a single answer. Included in Solo and Team.

A chatbot writes first, then attaches links that seem to support what it said. Counsel reads first, then writes only what it can quote word for word.

On the approach. Perplexity and ChatGPT pull snippets from the open web at the time of the query, generate an answer from them, then attach citations after the fact without checking — often beside what the source actually says. A 30-second test: click the links in your last Perplexity answer. Half do not say what they are credited with. Counsel does the opposite: selection from a carefully chosen AI corpus rated for quality, structured extraction at ingestion, scoring by an LLM judge against your precise decision, a synthesis built only from the extracted passages, grounding checked passage by passage, public marking of the remaining uncertainty. It invents no tool, no figure and no citation — and when it cannot ground a sentence, it says so in the document itself.

On usage. Your decisions, questions and watched subjects are kept and automatically re-checked against each new article. The system learns from your explicit signals (useful / off-topic / pinned) and from your clicks, tracks what changed between two Counsels (new, kept, promoted, dropped), and names the decisions with no recent signal — “agent framework: no signal for 41 days” — exactly where the reversals that come six months too late are born.

It is not a better prompt. It is a different architecture.

Team plans give every seat unlimited Counsel decisions, plus shared context across the whole workspace. €499/month flat, or €399/month billed annually (€4,788/year, -20%) — up to 5 seats either way. A workspace administrator manages members and billing. Shared decision context and team-wide feedback loops are available as soon as two or more seats are active.

Yes. Cancel from your account settings. Annual plans keep access until the end of the billing period. Monthly plans stop at the next cycle. You never lose your Counsel history or your favourites.

All major bank cards (Visa, Mastercard, Amex) via Stripe. Annual plans are charged once, upfront. Monthly plans renew on the same day each month. Invoices issued in the EU include reverse-charge VAT where applicable.

We offer a 40% discount on Solo to verified students, researchers at academic institutions and registered non-profits. Write to us at contact@aissential.tech from your institutional address.