Every term a vendor might say at you, in plain English — with a second definition for what it means when it appears in a sales demo.
Nothing matching that here yet — tell me the term and it'll be in next week's update.
Software that uses an AI model to decide for itself which steps to take towards a goal — reading, choosing tools and acting in sequence — rather than following a fixed script.
When a vendor says their platform is “agentic”, ask what the agent is actually allowed to decide. Much of what is sold as agentic is a fixed workflow with one AI step in the middle — which is often exactly what you want, but is not autonomy and should not be priced like it. Where genuine autonomy exists, your first questions are about control: what can it do without a human approving, and where is the log of what it did?
Marketing that overstates or invents a product's AI capabilities — rebadging old rules-based features, or claiming AI where little exists.
Regulators now treat this as a live enforcement area — the SEC has charged investment advisers over false AI claims, and the FCA’s Consumer Duty work points the same direction for firms repeating vendor claims to clients. The practical defence is simple: for every AI claim in a pitch, ask what the model actually does, on what data, with what failure rate. Vendors with real capability answer specifically; AI-washers answer with adjectives.
The doorway software offers so other software can use it — how your CRM, inbox and AI tools connect to each other without a human copying data between screens.
“Has an API” is a genuinely important procurement checkbox, because it determines whether a tool can join your automations or becomes an island. The follow-ups matter though: is the API included in your plan or a pricier tier, does it cover the data you actually need, and is it rate-limited into uselessness? “API access on the Enterprise plan only” is a common quiet upsell.
Automation follows rules you define in advance and is perfectly repeatable; AI interprets and generates, handling ambiguity at the cost of occasional confident errors. Most good systems combine both.
The distinction that prices and de-risks projects. Rules-based automation is cheap, testable and auditable — if a task can be written as rules, automate it that way and skip the AI premium. Use AI where judgement, language or messy input genuinely feature, and wrap it in review. A vendor who cannot tell you which parts of their product are rules and which are AI has told you something anyway.
The amount of text an AI model can consider at once — its working memory for a single task, measured in tokens.
When a vendor says the tool “reads your whole client file”, the context window is the fine print. If the file is bigger than the window, something is being left out or summarised on the way in, and the tool’s answer only reflects what made it through. A fair question in a demo: what happens when the document set exceeds the window — is anything silently dropped?
A marketing label for an AI assistant embedded in software you already use, suggesting and drafting alongside a human who stays in charge.
Originally a Microsoft product name, now a generic label — which is the point to be careful about: “copilot” tells you the interaction style, not the capability. Two products both called copilots can differ completely in what they access, what they retain, and whether output is checked. Look past the label to the specifics: what data does it see, where does it go, what is logged?
A way of turning text into lists of numbers that capture meaning, so a system can find passages that are about the same thing even when they use different words; a vector database stores and searches those numbers.
This is the machinery behind “search that understands meaning” and most RAG systems. Two things worth knowing in a procurement conversation: embeddings of your documents are still your client data and belong inside the same data-protection perimeter, and “semantic search” quality varies enormously — insist on a trial against your own documents, not the vendor’s demo set.
Further training an existing AI model on your own examples so it picks up your formats, tone or domain patterns — changing the model itself rather than just what you feed it.
Vendors sometimes say “trained on your data” when they mean the much lighter (and usually better) options: retrieval or careful prompting. True fine-tuning is expensive, needs curated examples, and locks you to a model version. If a vendor claims it, ask what specifically was fine-tuned, on how many examples, and — critically — whether your firm’s data is used to train models that serve other customers.
The handful of very large, general-purpose AI models — built by companies like Anthropic, OpenAI and Google — that almost all AI products are constructed on top of.
“Frontier” simply means one of the current best. The practical significance for your firm is concentration risk: if the same provider sits behind your transcription tool, your report writer and your CRM’s AI features, one upstream outage, price change or model update touches all three at once. Your vendor due diligence should ask what model each tool runs on — and what the vendor does when it changes.
Techniques that tie an AI's answers to identifiable sources — your documents, cited passages, linked references — so claims can be checked rather than taken on trust.
“Every answer is grounded with citations” is a strong claim worth testing, because citation display and citation accuracy are different things: models can cite a real document for a claim it does not support. In a demo, click the citations and read them. A tool whose citations check out is materially safer for regulated work than one that merely looks referenced.
Technical controls placed around an AI system to stop it doing unwanted things — leaking data, giving advice it shouldn't, acting outside its remit.
“Enterprise-grade guardrails” is doing heavy lifting in many pitches. Guardrails are real and worth having, but they are probabilistic filters, not guarantees — determined misuse and unlucky phrasing get through. Ask the vendor to name the specific guardrails relevant to your risk (client data leaving the tenant, advice-like output, actions taken without approval) and how failures are detected and reported to you.
When an AI model confidently produces information that is false — invented citations, fabricated figures, plausible-sounding claims with no source.
Every LLM does this; the question is only how often and with what safeguards. Treat any vendor who says their product “doesn’t hallucinate” as having failed a basic honesty test — the honest version is “we reduce and catch hallucinations, like this”. For regulated work, the practical control is a human verification gate on anything citable before it leaves the firm.
A system design where a person reviews or approves the AI's output before it takes effect — the control regulators most consistently expect for client-facing work.
The phrase to probe in any compliance conversation, because there is a wide gap between a human genuinely reviewing output and a human clicking an approve button at speed. The ICO has signalled attention to rubber-stamping, and a nominal check may not count as meaningful human involvement. Design reviews so the reviewer can realistically catch AI-specific failures — and can show they did.
Running an already-trained AI model to get an answer — as opposed to training, which is building the model in the first place. Every query your staff make is inference.
Inference is where the ongoing cost lives: every use of an AI feature costs the vendor compute, which is why AI add-ons carry per-seat or usage pricing and why “unlimited” plans have fair-use clauses buried in them. When a vendor’s AI feature is suspiciously cheap, the interesting question is what corners the inference is cutting — a smaller model, heavier caching, or your data subsidising something.
The core AI technology behind tools like ChatGPT and Claude: a model trained on vast amounts of text that predicts likely continuations, which makes it remarkably good at reading, summarising and drafting.
Almost every AI feature you are shown in 2026 is an LLM with product design around it. The vendor rarely built the model — they buy access from one of a handful of providers. That matters practically: ask whose model sits underneath, what happens to your data on its way there, and what happens to the product when the underlying model changes.
An open standard that lets AI assistants connect to other systems — files, CRMs, databases — through one common plug rather than custom integrations for each.
Increasingly name-dropped in 2026 demos as “supports MCP”. The genuine benefit is less lock-in: tools speaking a common protocol are easier to rewire when you change vendors. The governance angle is access: each MCP connection is a door from an AI system into a business system, so somebody in your firm should be able to list which doors exist and what each one can reach.
An AI model that works with more than text — reading images and documents, hearing audio, sometimes producing them too.
The practically useful version for a service firm is usually document vision: reading scanned PDFs, handwritten forms and screenshots without separate OCR software. In a demo, test it with your worst real documents — a skewed scan, a fax-quality statement — not the vendor’s clean samples. That is where multimodal claims earn or lose their keep.
Tools that let non-programmers build workflows and automations through visual interfaces — Zapier, Make, Power Automate — with low-code adding small amounts of scripting.
Genuinely useful, and the backbone of sensible first automation projects. The caution is sprawl: no-code makes it easy for automations to accumulate with no owner, no documentation and no offboarding when their creator leaves. Treat no-code automations as real systems — inventoried, owned and reviewed — or you are quietly building shadow infrastructure with a friendly interface.
The instructions given to an AI model; prompt engineering is writing and refining those instructions so the model produces reliable, correctly-formatted output.
Much of a polished AI product’s value lives in its prompts — which also means one badly-written prompt template can misfire systematically across every client it touches, unlike a human error which is usually one-off. Ask vendors how prompt changes are tested and rolled out; ask internally who owns the prompts your firm relies on and where they are versioned.
A technique where the AI first looks up relevant passages from your own documents, then writes its answer from what it found — grounding responses in your material rather than the model's general training.
“We use RAG on your knowledge base” is one of the most common vendor sentences of the past two years. It is a genuinely useful pattern — but its quality depends almost entirely on the retrieval step, which is the part demos hide. Ask to see it fail: pose a question whose answer is not in your documents and watch whether the tool says so or improvises.
A newer class of AI model that works through problems step by step before answering — slower and more expensive per query, but markedly better at multi-step analysis.
“Now with reasoning” is the current premium tier in many products. The honest trade-off: reasoning models are better at genuinely hard tasks (reconciliations, multi-document analysis, edge-case logic) and unnecessary for routine drafting and summarising. If a vendor charges extra for it, ask which of your actual tasks need it — paying reasoning prices for email drafting is buying a lorry to deliver letters.
Staff using AI tools the firm hasn't approved or doesn't know about — usually personal accounts on consumer tools, often with real client data.
Research suggests over 80% of workers use unapproved AI tools, and senior people are the heaviest users — a ban does not stop the use, it stops the reporting. Shadow AI is why an AI register and an approved-tools list are the first governance controls worth building: you cannot review, contract for, or defend processing you do not know is happening.
The standing instructions a product gives its AI model before your text arrives — defining its role, rules and tone. Users don't see it, but it shapes every answer.
When an AI tool behaves oddly — refusing reasonable requests, adopting a strange persona, ignoring your instructions — the system prompt is often why. For firms building their own AI workflows, the system prompt is where your compliance rules belong (“never state performance figures without a source”, “always flag uncertainty”), which also makes it a document compliance should have sight of.
A setting controlling how predictable an AI model's output is: low temperature gives consistent, conservative answers; high temperature gives varied, more creative ones.
Mostly invisible in finished products, but worth knowing when someone technical says a tool is “non-deterministic” — meaning the same question can produce different answers on different days. For regulated outputs you generally want low temperature and consistency; if a vendor’s tool gives materially different suitability wording on identical inputs, that is a question to raise, and temperature is the vocabulary to raise it in.
The unit AI models read and write text in — roughly three-quarters of a word in English — and the unit AI usage is metered and billed in.
Tokens appear in pricing conversations. “Token limits” and “per-token pricing” are how vendors pass on their own model costs, and generous-sounding allowances can be smaller than they look once documents are involved — a fifty-page PDF is tens of thousands of tokens before anyone asks a question. If a quote is usage-based, ask for the assumed monthly token volume in plain terms: how many documents, how many queries.
Zero-shot: an AI model doing a task from instructions alone, with no examples. Few-shot: giving it a handful of examples in the prompt to copy the pattern of.
Useful mainly as a reality check on capability claims. “Works out of the box on your documents” is a zero-shot claim, and quality usually jumps when a system is shown a few of your firm’s real examples instead. In evaluation, test tools on your formats with your examples — a few-shot setup that fits your templates beats a zero-shot demo that fits the vendor’s.
Missing a term you've been pitched? Tell me and it will be in next week's update. For how these terms play out in practice, the AI governance series is the companion reading.
Send a message straight to my inbox — I'll reply personally.