Acceptable Use Policy

This is the Acceptable Use Policy of Vernius Limited (“Vernius”, “we”, “us”, “our”), a company registered in England & Wales under company number 17114305. It sets the rules for using the services we run and the AI features we provide.

It governs use of:

• Hosted Services (each a Hosted Service). Any solution we host or operate for a Client, and the use of that solution by the Client and its end-users.
• AI features. The on-site assistant and any other AI features we make available, whether on this website or within a Hosted Service.

This policy supplements, and is in addition to, our Terms of Use and, where an engagement exists, our Terms of Business. If there is any conflict between this policy and the Terms of Business, the Terms of Business govern the commercial relationship. The specifics of a Hosted Service, including any limits and permitted uses, are as set out in the applicable Order.

This policy binds:

• the Client (“you”, “your” in the context of an engagement);
• the Client’s authorised users;
• end-users of any Hosted Service; and
• visitors using the on-site assistant, for whom “you” means the website visitor or user.

By accessing or using the services, or by allowing anyone else to do so, you agree to this policy. The Client is responsible for making its authorised users and end-users aware of this policy and for ensuring they comply with it.

This policy does not create any right enforceable under the Contracts (Rights of Third Parties) Act 1999 by anyone who is not a party to it. End-users and visitors are bound through the Client, and the rights to enforce it rest with Vernius and the Client.

You must not upload, store, generate, transmit or make available through the services any content that is, or that includes, the following. This list gives examples and is not exhaustive.

• Unlawful material. Anything that is unlawful or that breaches any applicable law or regulation.
• Infringing material. Anything that infringes intellectual property rights (including copyright, trade marks or database rights) or that misappropriates trade secrets or confidential information.
• Privacy breaches.Anything that breaches privacy or data protection rights, or that discloses another person’s personal data without a lawful basis or the authority to do so.
• Malicious code. Malware, viruses, worms, trojans, ransomware, or any other malicious or harmful code.
• Child sexual abuse material. This is prohibited absolutely, including any AI-generated or computer-generated depiction of a minor, and so is any content that sexualises or endangers children.
• Abusive content. Anything that is harassing, bullying, threatening, defamatory, obscene or grossly offensive.
• Hateful or violent content. Anything that incites or promotes violence, terrorism, extremism, hatred or discrimination against any person or group, including on the grounds of any characteristic protected by law.
• Deceptive content. Anything that is fraudulent or deceptive, or that is designed to impersonate any person or entity.

You must not use the services, or attempt to use them, for any of the following. This list gives examples and is not exhaustive.

• Gaining or attempting to gain unauthorised access to any system, account, network or data.
• Probing, scanning or testing the vulnerability of, or breaching, any security or authentication measure. Unauthorised access and related conduct are also offences under the Computer Misuse Act 1990.
• Interfering with, disrupting, overloading or degrading the service or any user’s use of it.
• Any denial-of-service or distributed denial-of-service activity.
• Sending spam, unsolicited bulk or commercial messages, or phishing, or otherwise breaching the Privacy and Electronic Communications Regulations 2003.
• Crypto-mining, or any other activity that consumes disproportionate or excessive resources.
• Scraping or harvesting data, imposing excessive automated load, or using bots in any way that is not expressly permitted.
• Circumventing, disabling or attempting to defeat any usage limit, rate limit, access control or authentication measure.
• Using the service to breach any law, or to infringe or misappropriate the rights of any third party.
• Reselling, sublicensing or providing the service to any third party except as permitted by the applicable Order.

The on-site assistant and any AI features are provided on the following terms, which are in addition to the rest of this policy.

• Lawful use. Do not use the assistant or any AI feature to generate, promote or facilitate unlawful, harmful, deceptive, fraudulent or infringing content, including deepfakes, impersonation or synthetic media intended to mislead.
• Prompt injection and jailbreaking. Do not attempt to manipulate, bypass or defeat the safeguards, system instructions or guardrails of the assistant or the underlying models, whether by prompt injection, jailbreaking or any other means.
• Model integrity. Do not attempt to reverse engineer, decompile, extract training data or system prompts from, or expose the internals, weights or parameters of, or otherwise probe, the underlying models.
• No competing use. Do not use outputs to build, train or improve a competing AI model, or to scrape or distil the model.
• No automated abuse. Do not use the assistant for bulk querying, as a free API proxy, or to circumvent rate limits.
• Human review for high-risk decisions. Do not rely on AI outputs for high-risk, safety-critical, legal, medical, financial, insurance, employment or similar decisions without qualified human review.

Accuracy. AI outputs may be inaccurate, incomplete, biased or misleading. You are responsible for checking them before relying on or acting on them. Outputs are not professional advice.

Provider policies.You must comply with the policies of the underlying model providers, including Anthropic’s usage policies at anthropic.com/legal/aup. A breach of those policies is also a breach of this policy.

Data flow.When you use the assistant, your messages are sent to Anthropic via Vercel’s AI gateway to generate a reply. API inputs are not used to train Anthropic’s models. Our Privacy Policy explains how that data is handled.

The Client is responsible for all use of the services by its authorised users and end-users, and for all content and data it or they submit (“Client Materials”, as defined in the Terms of Business). Acts or omissions of the Client’s users are treated as the Client’s own for the purposes of this policy.

The Client must ensure that it holds all rights, consents and lawful bases needed for the content and data it submits and processes through the services.

Where the Client acts as controller and Vernius as processor, our Data Processing Agreement governs the processing of personal data. Nothing in this policy reduces those obligations.

We do not routinely or actively monitor content or usage, and nothing in this policy implies a duty to pre-screen content.

We do, however, reserve the right to:

• investigate suspected breaches, and to access, review, preserve or disclose content and logs where reasonably necessary to do so, to comply with law, or to protect the service, us or others;
• remove, disable or restrict access to content, and suspend or restrict access, immediately where use is unlawful, poses a security risk, or risks harm to the service, to us or to any third party, and otherwise on reasonable notice.

Suspension or restriction to protect the service does not relieve the Client of its payment obligations under the applicable Order.

Report suspected breaches, illegal content or security issues to contact@vernius.co.uk. Please include enough detail to let us identify and locate the content or activity.

We will review reports and take the action we consider appropriate.

Where you breach this policy, we may take one or more of the following steps:

• issue a warning;
• remove or disable the content in question;
• suspend or restrict access; and
• terminate access or the engagement.

For an engagement, a breach of this policy may be treated as a breach of the Terms of Business. A serious or repeated breach may be a material breach entitling us to suspend or terminate. Where the breach is capable of remedy, this is subject to the 14-day cure period for material breach set out in the Terms of Business.

We reserve the right to cooperate with law enforcement and regulatory authorities, and to disclose information where required by law or to investigate suspected unlawful activity.

The Client remains liable for its own breaches and those of its users. The indemnity and liability position, including any caps, is set out in the Terms of Business and is not restated here. These rights are without prejudice to, and in addition to, any other rights or remedies we may have.

Nothing in this policy excludes or limits any liability that cannot lawfully be excluded or limited, including liability for death or personal injury caused by negligence or for fraud.

We may update this policy from time to time. The current version is the one published on this website, and the “last updated” date is shown at the top of this page.

Continued use of the services after a change takes effect constitutes acceptance of the updated policy. Material changes affecting an engagement will be handled in line with the Terms of Business.

Questions about this policy can be sent to contact@vernius.co.uk.

For formal notices, our registered identity is:

Vernius Limited, company number 17114305, registered in England & Wales. Registered office: