1. Summary
Lexradar is an AI-assisted regulatory intelligence and workflow tool. We summarise regulatory changes, score their likely impact, and help you assign compliance tasks. We are not a law firm, a compliance consultancy, a regulator, or a source of legal, tax, accounting, or professional advice. AI Output may be inaccurate, incomplete, or out of date. You are responsible for verifying anything you rely on against the original source and for getting professional advice when it matters.
2. What Lexradar does
The Service is designed to help compliance teams:
- discover regulatory changes from selected UK, EU, US, and APAC sources, faster than manual monitoring allows;
- understand a change in plain English, with the obligations and effective dates extracted;
- assess likely business impact using a transparent scoring model;
- assign tasks, attach evidence, and produce an audit trail for internal reviews and external assurance work.
3. What Lexradar does not do
To be unambiguous, Lexradar does not:
- provide legal advice or act as your lawyer;
- provide tax, accounting, audit, or actuarial advice;
- guarantee that the regulatory feed is complete or that every change has been captured;
- guarantee that any AI Output is accurate, current, or applicable to your situation;
- create an attorney–client, solicitor–client, or any other professional relationship with you;
- replace the role of your in-house compliance team, external counsel, or appointed compliance officer.
4. No legal or professional advice
AI Output, change summaries, impact scores, and the answers produced by the Lexradar assistant are informational only. They are general guidance about a regulatory change and are not tailored to your organisation, your products, your customers, your contracts, or your risk appetite.
Nothing in the Service, in any AI Output, or in any communication from a Lexradar employee constitutes legal, tax, accounting, regulatory, or other professional advice. You should not act, or refrain from acting, on the basis of AI Output alone, and you should not present AI Output as the formal position of TITADE Ltd or of any regulator.
Where a compliance decision has legal consequences — for example, a notification to a regulator, a change to a regulated product, or a disclosure to investors — you must obtain advice from a qualified professional who is aware of your specific circumstances.
5. Accuracy of AI Output
Lexradar's AI is built on large language models that generate text statistically. The same input can produce subtly different outputs on different runs. Even when an AI Output is internally consistent and confidently worded, it may still be:
- factually wrong about a provision, definition, or effective date;
- based on an out-of-date version of a source document;
- missing a nuance that a human expert would have caught;
- over-claiming certainty in an area where the underlying regulation is ambiguous.
Every AI Output that could lead to a compliance decision should be checked against the linked primary source before you rely on it. Where the primary source is paywalled or unavailable, you should treat the Output as a starting point for further research rather than an answer.
6. Hallucinations and omissions
Large language models are known to hallucinate — that is, to produce text that looks plausible but is not supported by the source. This is an industry-wide limitation of generative AI, not a bug in Lexradar. We apply a number of mitigations:
- the assistant is grounded in the documents and changes you can already see in your workspace;
- we cite the source document and the relevant section whenever we make a claim about a regulation;
- we surface confidence indicators for impact scores and obligation extraction.
These mitigations reduce, but do not eliminate, the risk of a hallucination or an omission. If a claim looks too clean, too broad, or too confident, treat it as a prompt to verify, not a final answer.
7. Source coverage and timeliness
The regulatory feed is built from a curated list of public sources (regulators, official journals, and trade publications). The full list of sources is published on our Sources page and is updated as we add or remove coverage. The list is also the operative description of what we cover: if a regulator is not on the list, Lexradar does not monitor it.
The delay between a change being published by a source and appearing in your feed depends on the source's publication format, our crawl frequency, and any downstream AI processing. In normal operation new items appear within minutes to a few hours; in some cases — for example, if a source publishes a notice behind a JavaScript-heavy page — the delay can be longer.
We do not warrant that the feed is exhaustive, that all changes will be captured, or that items will appear within a specific timeframe.
8. Human judgement remains essential
Lexradar is built to make human judgement faster, not to replace it. The people on your team remain responsible for:
- interpreting AI Output in the context of your business;
- deciding which changes are in scope for your organisation;
- approving the wording of any communication to a regulator;
- maintaining the policies, controls, and training that turn a change into actual compliance.
9. We do not train on your content
Your Customer Content and your AI conversations are neverused to train any AI model. Our configuration of the Anthropic API is set to zero-retention and zero-training; prompts and responses are discarded after the inference call. This is contractually guaranteed in our Data Processing Agreement with Anthropic and is the most important boundary on AI Output that we can offer you.
See our Privacy Policy for the full statement of how we handle AI-assisted processing.
10. Your responsibility
By using the Service, you agree that you will:
- verify any AI Output against the original source before acting on it;
- consult a qualified professional for any decision with legal or regulatory consequences;
- not present AI Output as the formal position of any regulator, law firm, or professional adviser;
- not use AI Output as the sole basis for a notification, filing, or disclosure to a regulator.
11. No warranty
To the maximum extent permitted by law, the Service and all AI Output are provided as is and as available, without warranty of any kind, express or implied, including any warranty of accuracy, completeness, fitness for a particular purpose, or non-infringement. Nothing in this disclaimer is intended to exclude liability that cannot lawfully be excluded.
For the full allocation of risk between you and TITADE Ltd, see the Terms of Service.
12. Contact
If you have questions about this disclaimer or the limits of AI Output, email Admin@lexradar.co.