Last updated 9 November 2026
Terms of Service
These terms are the agreement between your organisation and Cerberius Pty Ltd (ABN 57 686 421 152). They cover the Cerberius Agents website at cerberius.com and the Cerberius Agents service.
Please read them carefully. In short: the agent acts only through the access you connect and the instructions your team gives it, within our safety rules. Every change it proposes is shown to a person to approve first. The one exception is Development's merge and deploy without asking, and only for a request where someone with Development access ticked it. Approving a change is your decision.
These terms place responsibility for approving replies, fixes and changes with you, and they limit our liability to the extent the law allows.
1.About these terms#
By accepting an invitation, signing in, connecting a system or otherwise using the service, you agree to these terms. If you don't agree, don't use the service.
If you use the service for an organisation, you confirm that you have authority to bind that organisation to these terms. In that case "you" means the organisation. You must be at least 18 years old.
The service is for businesses and other organisations. It is not offered to people for personal, domestic or household use.
If we have signed a separate written agreement with your organisation, that agreement applies where it differs from these terms. Additional terms shown in the app before you switch on a feature (such as the terms for changes) form part of these terms. Where they deal with that feature, they apply ahead of these terms.
2.Words we use#
- "We", "us" and "Cerberius" mean Cerberius Pty Ltd (ABN 57 686 421 152).
- "Service" means Cerberius Agents: the web app, its API, its AI agent, chatbots, setup assistants, the content writer, the chat widget, the phone app, the help centre and this website.
- "Members" means the people you, or we at your request, add to your organisation in the service, in any role.
- "Customer data" means everything you or your members put into the service or let it reach. That includes tickets, code, logs, database results, chatbot help content and conversations, and the drafts and records the service makes from them.
- "End users" means your own customers and other people whose information is in customer data, such as people who open tickets or chat with your chatbot.
- "AI output" means anything the service's AI produces: findings, draft replies, chatbot answers, proposed fixes and proposed changes.
- "Your systems" means the repositories, Kubernetes clusters, log stores, databases, ticket systems, Microsoft 365 and Google Workspace accounts, domains and DNS records, websites and other systems you connect to the service, including systems you look after for your own clients.
3.A new service, and beta features#
Cerberius Agents is a new service, and it changes often. That means:
- We may limit or pause sign-ups, and decide who we offer the service to.
- The service is provided "as is" and "as available", subject to section 18. It may have errors, gaps and interruptions.
- Features marked beta in the app or on our website, such as Development, are still being tried out. They are more likely to fail or change, and we may withdraw them.
- We may add, change, limit or withdraw any feature. We'll tell you first about changes that affect how you work, where we reasonably can. If we withdraw a feature that is a main part of the paid plan you're on, other than a beta feature, we'll give you at least 30 days' notice. You can then cancel, and we'll refund what you prepaid for the rest of the period.
- There is no service level agreement (SLA), uptime commitment or support response time, unless we have agreed one with you in writing.
- Plans and fees are described in section 14.
4.Accounts, invitations and members#
- Members sign in with their own Google or Apple account, or with a code sent to their email. An owner or admin of your organisation decides who is added, what role they have and which platforms they can see.
- Only add people who need access, and remove them promptly when they no longer do.
- Keep sign-in details, API keys and tokens secure. Don't share them, and don't let one person's account be used by others.
- Tell us promptly at [email protected] if you think someone has gained access they shouldn't have.
- Approving or acting in the phone app, or through your own tools with one of your API keys, counts the same as doing it in the web app.
Everything done through your organisation's accounts and API keys is treated as done by you. That includes the actions of your members, and of anyone using a member's account or one of your API keys, whether or not you authorised it, unless it happened because of our breach of these terms or our negligence.
5.AI output and your decisions#
The service uses AI models to investigate tickets, draft replies, answer chatbot questions and propose fixes and changes. AI output can be wrong, incomplete, out of date or misleading, even when it sounds confident. It is not professional advice of any kind.
What the agent does depends on what you control: the access you connect (keys, logins, accounts and domains) and the instructions your team gives it (your settings, your notes about your system, your organisation's rules and your members' messages to it). It always follows our safety rules, which your instructions can't override.
- You must review AI output before you rely on it, send it or act on it.
- You decide what is sent to your end users and what is changed in your systems. The service is a tool that helps you make those decisions. It doesn't make them for you.
- AI output for you may be similar to output for others. We don't promise that AI output is unique, or that it doesn't resemble someone else's material.
- Chatbots and setup assistants answer people directly, without a person checking each answer first. See section 8.
6.Replies, fixes and changes#
When a member approves a reply, a fix or a change, or when your settings send something automatically, that is your action and your decision. You are responsible for its content and its consequences.
Replies
A draft reply is sent to your ticket system only after a member approves it. The exception is automatic replies, which are off by default. If an owner or admin switches them on, you choose to send replies without a person reviewing each one. You do that at your own risk. The built-in checks that hold some replies back reduce risk. They don't remove it, and they won't catch every problem.
Code fixes
An approved fix becomes a pull request (or merge request) in your repository. We merge it only when one of your approvers asks us to. Your team is responsible for reviewing, testing and deciding whether to merge and deploy it, including asking us to merge it, and for everything that follows from doing so.
Changes to databases, Kubernetes, Microsoft 365, Google Workspace and DNS
The agent can propose a change to a database, a Kubernetes cluster, Microsoft 365, Google Workspace, or DNS records at GoDaddy. A change runs only after that kind of change has been made available to your organisation, switched on by your owner or admin, and approved by a member, exactly as shown. Before switching changes on, your owner or admin accepts the terms for changes shown in the app, for the whole organisation.
A member may edit a change before approving it. A change approved after editing is your decision, in the same way as one approved as the agent proposed it.
The service's safeguards reduce risk but don't make a change safe. They include trying Kubernetes changes first as a dry run; reading what a Microsoft 365, Google Workspace or DNS change touches before making it; running database changes in one transaction with a row limit; and never making some kinds of change, such as to admin roles, policies or name servers. They won't catch every problem.
You are responsible for having current, tested backups of your own systems and data, and for being able to restore them. We don't back up your systems.
Log watcher
If you switch on the log watcher, errors in your logs become tickets that the agent investigates. They are handled like any other ticket, under your settings.
The agent's own machine
Your owner or admin can give the agent its own machine on a platform. It is off until they switch it on. On it, the agent runs commands it chooses, such as looking up a domain's records or testing a mail server, without a person approving each one. If your owner or admin also lets it reach the internet, it can reach public websites and servers. The machine holds none of your keys.
With the internet on, a ticket written to trick the agent could get it to send content it has read, such as a list of users, to an outside address. By switching the internet on, you accept that risk. Switch it on only where the checks are worth it.
7.Development (beta)#
Development lets the members your owner or admin chooses ask the agent to change your product: its code, its website, its text or its pictures. The agent proposes the change and makes it in your repository, and the service opens a pull request. The service merges it when one of those members presses Merge or asks for it, or straight away if one of them ticked merge and deploy without asking for that request. Merging usually deploys the change to your live product.
Development is a beta feature, provided "as is". You use it entirely at your own risk, and we accept no responsibility for any change made, merged or deployed through it, or for what follows. In particular:
- the agent can't run, build or test the changes it makes. A change may not work, may break something else, or may not do what was asked;
- when someone with Development access ticks merge and deploy without asking for a request, that request's changes are merged and deployed without anyone reviewing them first. Ticking it is your decision;
- your owner or admin decides who has Development access, and everyone with it can merge and deploy. You are responsible for everything they do with it, and for reviewing, testing and, where needed, reversing changes;
- you are responsible for having the rights to anything you give the agent to put into your product, such as pictures, text and logos;
- you remain responsible for your product, your deployments, your backups and your users, as set out in section 6.
We may change Development, limit it or withdraw it at any time while it is in beta.
8.Chatbots, setup assistants and the content writer#
Chatbots
A chatbot answers your website or portal visitors directly and immediately, without a person checking each answer. You choose its help content, instructions and settings, and you're responsible for them and for its answers. The chat tells visitors that its answers come from an AI. You must not hide or remove that notice, and you must tell visitors how their conversations are handled.
Setup assistants
A setup assistant lets your own customers, signed in to your portal, ask for changes in their account in your system. It calls only the actions of your system that you switch on, using a short-lived key your portal's server gives it for that customer. A change runs when your customer presses Approve. No one at your organisation or ours checks it first.
- You decide which actions it may use, and you're responsible for your system, your API and its permission checks, and for what your customers ask for and approve.
- Between you and us, a change your customer approves is treated as made on your instructions. We are not responsible for it, or for your arrangements with your customers.
- You must tell your customers they are using an AI assistant, and give them any terms and notices your own service needs.
Content writer
The content writer researches the public web and drafts articles for your website. Nothing is published until a member approves it. You are responsible for what you publish: checking that it is accurate, that you may use everything in it, and that it complies with the law, including laws on advertising and consumer protection. AI-written text may resemble other material, and may not be protected by copyright in the way human-written text is.
9.What you connect#
You decide what the service can reach. When you connect a system or add a key, you confirm that:
- you own the system or are authorised by its owner to connect it and to let the service use it as described. If you look after systems for your own clients, that includes your client's permission;
- you give each key and login only the access it needs. Read-only for investigation; write access only for the fixes and changes you have switched on, and only as much as they need;
- for Kubernetes, the logins you give can't read Secrets or change permissions, and are limited to the namespaces you want the service to see;
- for Microsoft 365 and Google Workspace, the app registration or service account you connect has only the permissions or scopes you are happy for the agent to read with and propose changes through;
- for GoDaddy, you list only domains the platform may use and that you are authorised to manage. One GoDaddy key reaches every domain in its account. The service keeps to the domains you list;
- for HaloPSA, each platform takes only the Halo clients whose tickets it should handle;
- you keep keys current, and rotate or remove them if they may have been exposed.
We are not responsible for what the service does with access you gave it, when it acts as these terms and the app describe. We are not responsible for the security, availability or configuration of your systems.
10.Acceptable use#
You must not, and must not let anyone else:
- use the service for anything unlawful, fraudulent, deceptive or harmful, or to help anyone else do so;
- upload or process content that is unlawful, infringes anyone's rights, or is malicious (such as malware), or send messages that people haven't agreed to receive;
- put into the service, or let it reach, personal information or other data that you don't have the right to process in this way;
- attack, probe, scan or load-test the service, or try to get into any part of it or any data that isn't yours, without our written permission;
- get around the service's limits or safeguards, including rate limits, cost limits, approval steps and the agent's read-only restrictions, or try to make the agent ignore its rules;
- copy, decompile or reverse engineer the service, except to the extent the law allows despite this restriction;
- resell, sublicense or provide the service to others, except to serve your own end users;
- use the service, or AI output, to build or train a competing product or service;
- use the service where its failure or an error in AI output could lead to death, personal injury, or serious physical, environmental or financial harm, such as in medical, emergency or safety-critical systems;
- use the service in breach of the terms of the AI provider or other third-party service it connects to;
- use the service if you, or anyone you act for, are the target of sanctions under Australian, United Nations, United States, United Kingdom or European Union law, or from a country or region subject to comprehensive sanctions.
11.Your data#
Your data stays yours. As between you and us, you own customer data, and you keep any rights you have in AI output made for you.
You give us a non-exclusive, worldwide, royalty-free licence to host, copy, process, transmit and display customer data, and to send it to the third-party services you choose, as needed to provide, support and secure the service and to meet our legal obligations. The licence ends when the data is deleted.
- We do not sell customer data.
- We do not use customer data to train AI models.
- Our staff access customer data only to provide, support or secure the service, to investigate a problem you raise or misuse of the service, or where the law requires it.
For personal information in customer data, you decide what is collected and why, and we handle it on your behalf and on your instructions. You are responsible for having a lawful basis to collect it and to have it processed by us and by the AI provider you choose; for giving your end users any notices and getting any consents the law requires, including telling them that an AI service reads and answers their messages; and for not putting in more personal or sensitive information than the work needs. How we handle personal information is described in our Privacy Policy.
When we handle personal information for you
For personal information in customer data, we act on your behalf (in the language of some laws, as your "processor" or "service provider"). We will:
- handle it only to provide the service to you, on your instructions. Your instructions are these terms, your settings and what your members ask for in the service. We'll tell you if we think an instruction breaks the law;
- not sell or share it, not combine it with personal information from other sources except as the service needs, and not use it for our own purposes, such as marketing or training AI models;
- make sure the people who can reach it are bound to keep it confidential;
- protect it as described in section 12;
- use only the service providers named in our Privacy Policy, on terms that protect it at least as well as these. We'll update that list before we add one, and email your organisation's owner about a new provider that will receive customer data. If you object, you can cancel, and we'll refund what you prepaid for the rest of the period. We remain responsible for our providers;
- help you, as far as reasonably possible given what we can see, to answer requests from people about their information, and to meet your duties on security, breach notification and privacy impact assessments;
- tell you without undue delay after we become aware of a breach affecting it, with what we know. You decide whether to tell the people affected and the regulator, as the one who controls that information, unless we agree otherwise. We'll help;
- tell you if we can no longer meet these commitments. You may then stop us using the information and fix any misuse;
- delete it when you delete it in the service, or when this agreement ends, as described in section 17;
- give you the information you reasonably need to show that these commitments are met. If that isn't enough, you may audit our compliance, at your cost, on 30 days' notice, at most once a year, in a way that protects other customers' data and our security.
Customer data is stored in Australia. If you are in the European Union, the United Kingdom or Switzerland and the law requires an agreed safeguard for personal information you send us, such as the European Commission's standard contractual clauses, contact us and we'll put it in place with you.
12.Confidentiality and security#
Each of us may receive information from the other that is not public. The one receiving it will use it only for this agreement, protect it with reasonable care, and not disclose it except to staff, contractors and advisers who need it and are bound to keep it confidential, or where the law requires. This doesn't apply to information that is or becomes public through no fault of the receiver, or that the receiver already had or developed independently.
We take reasonable technical and organisational measures to protect customer data. These include encrypting stored keys, tokens and passwords, hiding values that look like secrets before tool output and logs are stored, keeping each organisation's data separate, and limiting what the agent can reach. They are described on our Security page.
No system is completely secure. We don't guarantee that the service or customer data will be free from unauthorised access, loss or corruption. We don't hold any security certification. If we become aware of a breach that affects your data, we'll tell you as soon as reasonably practicable and act as described in our Privacy Policy.
13.Third-party services and AI providers#
The service works with services we don't control. They include AI providers (Anthropic for Claude, and OpenAI if you choose it for a chatbot), GitHub and GitLab, HaloPSA, Microsoft, Google (for sign-in and Google Workspace), Apple sign-in, GoDaddy, Slack and Telegram, Expo (for phone notifications), the payment processors that take your payments, and your own ticket system and other systems.
- Your use of each one is governed by your agreement with its provider.
- We are not responsible for what they do or fail to do, including outages, changes to their features or prices, how they handle data, or the content their models produce.
- If a third party changes or withdraws something the service depends on, we may have to change or stop a feature.
AI keys are yours. You add your own AI provider keys, and AI usage runs under your account with that provider. You pay the provider's charges, whatever caused them. The service's cost and time limits help you control usage, but they are estimates and safeguards, not a guarantee of what you'll be charged. If we agree in writing to provide a key for you, the terms we agree for that apply.
14.Plans, fees and taxes#
- Our plans and their prices are on our pricing page, and the plans form shows the price, the period and what's included before you confirm. Paid plans are charged in advance for each period, a month or a year. Anything else we agree with you in writing, such as an Enterprise quote, applies instead.
- The Free plan has no charge. We may change its limits or stop offering it, by giving you at least 30 days' notice.
- Free trials. A paid plan may start with a free trial, shown in the plans form with its length (currently 30 days). We ask for a payment method at the start. Unless you cancel before the trial ends, the plan carries on and you're charged for it from the end of the trial. Cancelling during the trial costs nothing.
- Prices are in Australian dollars if you sign up from Australia, and in US dollars otherwise. Your currency stays the same after you sign up.
- Each plan sets how many platforms, people and chatbots your organisation can have. Adding more than your plan allows is refused until you change plan. Development, the content writer and setup assistants come only with the plans that include them.
- You can change plan in the app. The plans form shows what you'll pay, and when the change applies, before you confirm.
- Plans and add-ons renew automatically at the end of each period until you cancel them. An owner or admin can cancel at any time under Your account › Plan & billing. What you've paid for stays on until the end of its period, and isn't renewed. Deleting your organisation cancels its plan in the same way.
- Each ticket the agent starts on counts towards your plan's monthly allowance. On a paid plan, tickets over the allowance are charged at the price on the pricing page, on your next invoice. On the Free plan, the agent stops starting new tickets once the allowance is used.
- AI usage is charged to you by your AI provider, through your own key. It isn't part of our fees.
- We may change our plans and prices. We'll give you at least 30 days' notice before a new price applies to your plan, and you can cancel before then.
- Payments are taken by our billing service and its payment processors, such as Square, Stripe and PayPal. Card details go to the payment processor; we don't store them.
- We don't charge GST at the moment, so the price shown is what we charge. If that changes, we'll tell you at least 30 days before GST is added, and you can cancel before then. You pay any taxes your own country applies to what you buy from us, other than taxes on our income. Your invoices are in the app.
- Except where the law requires or we agree otherwise, fees already paid are not refundable, including for the rest of a period after you cancel.
- If a bill stays unpaid and your billing account is locked for it, the agent stops starting work until it's paid. We may suspend the rest of the service if it is still unpaid at least 7 days after we tell you.
15.Intellectual property and feedback#
We and our licensors own the service, its software, design, documentation and the agent's instructions, and all intellectual property in them. These terms give you a limited, non-exclusive, non-transferable right to use the service while you have access. They don't transfer any ownership to you.
If you give us feedback or suggestions, we may use them for any purpose without paying you or owing you anything. You don't have to give us feedback.
We won't name your organisation as a customer publicly without your permission.
16.Availability and support#
We don't promise that the service will be available at any particular time, uninterrupted or free of errors. We may carry out maintenance, which can cause interruptions. The service also depends on AI providers and other third parties, which can fail or change independently of us.
We provide support by email at [email protected], on a reasonable-efforts basis. We don't promise a response time unless we have agreed one in writing.
17.Suspension and termination#
By you
You can stop using the service at any time. An owner can delete the organisation in the app, under Your account › Team, which cancels its plan and removes its platforms and everything in them. You can also ask us at [email protected]. An owner can delete a single platform too, which removes its tickets, history, settings and chatbots.
By us
We may end this agreement for any reason by giving you at least 30 days' notice, for example if we stop offering the service. We'll then refund what you prepaid for the period after it ends.
We may suspend or end your access, or any member's access, immediately, with notice as soon as practical, if:
- you or a member seriously breach these terms, or breach them and don't fix it within 14 days of our notice;
- we reasonably believe your use creates a security risk, could harm the service, us, other customers or third parties, or could expose us to liability;
- we are required to by law, or by a third party the service depends on;
- fees are still unpaid at least 7 days after we told you they were overdue.
We may also switch off a feature, such as automatic replies, fixes or changes, for your organisation if we reasonably believe it is causing or could cause harm.
What happens after
- Your right to use the service ends, and fees that are owed become payable.
- Requests that are waiting to run, such as approved changes, may not run.
- When an owner deletes the organisation in the app, its customer data is deleted straight away. Ask us for a copy of anything you want to keep before you do.
- If your access ends another way, you can ask us for a copy of your tickets, replies and findings, in a common machine-readable format, for 30 days after it ends. We then delete your customer data, or sooner if you ask.
- We don't charge for leaving, or for a copy of your data.
- We don't delete old backups yet, so copies stay in the backups taken before your data was deleted. Backups are used only to restore the service after a failure. We also keep what the law requires us to. See our Privacy Policy for how long we keep data.
- Sections that by their nature should continue, including 11, 12, 15 and 18 to 26, continue after this agreement ends.
18.Your rights under Australian Consumer Law#
The Australian Consumer Law (Schedule 2 to the Competition and Consumer Act 2010 (Cth)) and other laws may give you guarantees and rights that can't be excluded, restricted or modified. Nothing in these terms excludes, restricts or modifies those rights.
Where the law lets us limit our liability for failing to meet such a guarantee, our liability is limited, at our choice, to supplying the services again or paying the cost of having them supplied again.
19.Disclaimers#
Subject to section 18, and to the extent the law allows:
- the service and AI output are provided "as is" and "as available", and we exclude all warranties, conditions and guarantees not expressly set out in these terms, whether express or implied, including those about quality, fitness for a particular purpose, accuracy and non-infringement;
- we don't warrant that AI output will be correct, complete or suitable, that a fix will work, or that a change will have the effect intended;
- we don't warrant that the service will meet your requirements or work with all of your systems.
20.Limitation of liability#
Subject to section 18, and to the extent the law allows:
- Indirect loss. Neither of us is liable to the other for any indirect, incidental, special or consequential loss, or for any loss of profit, revenue, business, opportunity, goodwill or reputation, or for loss or corruption of data, or business interruption, however caused. This doesn't apply to fees you owe us, to your indemnity in section 21, or to a breach of section 10.
- Your decisions. We are not liable for loss arising from: a reply, fix or change that you approved, as proposed or after a member edited it, or that your settings sent automatically, including changes to databases, Kubernetes, Microsoft 365, Google Workspace and DNS records; a pull request you merged or deployed, or asked us to merge; any change made, merged or deployed through Development (section 7); anything the agent's own machine did once you switched it on, including what it sent over the internet if you allowed that; your reliance on AI output; the access you gave the service to your systems, and the instructions your team gave the agent; the acts or omissions of any third-party service; or the lack of a working backup of your systems.
- Cap. The total liability of each of us to the other, for all claims arising out of or in connection with the service or these terms, added together, is limited to the greater of the fees you paid us for the service in the 12 months before the event that gave rise to the claim, or A$500. This cap doesn't apply to fees you owe us, to your indemnity in section 21, or to a breach of section 10.
- Contribution. The liability of each of us is reduced to the extent that the other, or the other's people or systems, caused or contributed to the loss.
These limits apply whatever the legal basis of the claim, including contract, negligence and other torts, or statute. They don't limit liability that can't be limited by law, such as for fraud.
21.Your indemnity#
You indemnify us, and our directors, staff and contractors, against all claims, losses, damages, costs (including reasonable legal costs) and liabilities arising from:
- customer data, including a claim that it infringes someone's rights or that you had no right to provide it;
- your instructions, settings and approvals, including replies sent, fixes merged, changes run (as proposed or after a member edited them), automatic replies you switched on, what the agent's own machine did once you switched it on, and changes made, merged or deployed through Development;
- your use of the service in breach of these terms or of any law;
- claims by your end users, your members or other third parties about your products, your services or how you dealt with them, including through replies or chatbot answers sent through the service.
Your liability under this indemnity is reduced to the extent the loss was caused by our breach of these terms or our negligence.
22.Privacy#
Our Privacy Policy explains how we handle personal information, including personal information in customer data, and forms part of these terms.
23.Changes to these terms#
We may change these terms. We'll publish the new version on this page with a new "Last updated" date.
If a change materially reduces your rights or increases your obligations, we'll give you at least 30 days' notice by email to your organisation's owner or in the app, unless the change is needed sooner for legal, security or safety reasons. If you don't accept a change, you can stop using the service before it takes effect, and we'll refund any fees you prepaid for the period after that date. If you keep using the service after a change takes effect, you accept it.
Other changes, such as corrections and clarifications, take effect when published.
24.Events outside our control#
Neither of us is liable for a delay or failure caused by something beyond reasonable control. Examples are an outage at an AI provider, hosting provider or other service we depend on; a failure of the internet or of power; a cyber attack; a natural disaster, pandemic or industrial action; or an act of government. This doesn't excuse paying money that is owed.
25.General#
- Assignment. You may not transfer this agreement without our written consent. We may transfer it to a related company, or as part of a sale or reorganisation of our business, by telling you.
- Subcontractors. We may use subcontractors and service providers to provide the service. We remain responsible for our obligations under these terms.
- Entire agreement. These terms, the Privacy Policy, additional terms you accept in the app, and any written agreement with us, are the whole agreement between us about the service. They replace any earlier statements or summaries.
- Severability. If any part of these terms can't be enforced, it is read down or removed to the extent needed, and the rest still applies.
- No waiver. Not enforcing a right, or delaying, doesn't mean we give it up.
- Relationship. We are independent contractors. Nothing in these terms makes us partners, employees, agents or a joint venture.
26.Governing law and disputes#
These terms are governed by the laws of Victoria, Australia. Each of us submits to the non-exclusive jurisdiction of the courts of Victoria and the courts that hear appeals from them.
Before starting court proceedings (except for urgent relief), tell the other in writing about the dispute and try in good faith to resolve it by discussion for at least 30 days.
27.Contact and notices#
Cerberius Pty Ltd (ABN 57 686 421 152)
[email protected]
We send notices to the email address of your organisation's owner, or through the app. Send notices to us by email to the address above. A notice by email is received when it is sent, unless the sender gets a message that it wasn't delivered.