Terms of Service

Summary of key points

This summary is provided for convenience only. It does not replace the full Terms below, which govern in the event of any conflict.

• Content safeguards are permissive and imperfect, so output may still be objectionable
• AI outputs are not reliable, and the model can hallucinate and contradict itself
• Never follow AI medical, legal, financial, or safety advice without expert verification
• Aelius AI bears zero liability for any harm caused by reliance on AI output
• You are always talking to an AI, never a person, and the images it makes are labelled as AI-generated
• Pictures from the web in replies are checked automatically. Sexual pictures (nude, sexual acts, or posed in a sexual way) are shown only on the website and only if you turn on Show adult pictures (18+) there; the apps never show them. Revealing pictures that are not sexual need no setting, and nothing that could involve a minor is ever shown (§4.5)
• You must be 18 or older to use this service, and you confirm your age when you sign up or send your first message
• Disputes are resolved by binding arbitration, and class action lawsuits are waived, unless you opt out within 30 days (§16.2A)
• You may not use the Service or its output to train, distill or evaluate other AI models, or to do AI data work such as labelling, captioning or rating for others (§7.7)
• Usage limits apply to you as a person, and getting around them (extra accounts, proxies, VPNs) can end all of your accounts
• Aelius Pro renews every month and is charged automatically until you cancel. You can cancel online at any time, in two clicks, and you keep Pro until the end of the month you paid for (§7A)
• Pro is sold to you by Link (Stripe), which appears on your statement, sends your receipts, and handles payment questions (§7A.2)
• You retain ownership of content you provide as input
• You may delete your account and all data at any time

1. Definitions

For the purposes of these Terms of Service, the following terms shall have the meanings set out below:

"Aelius AI" / "we" / "us" / "our"

The operator(s) of the Aelius AI platform, website, and associated services.

"Service"

The Aelius AI website, chat interface, image generation tools, API, and all related features and content.

"AI Model" / "the Model"

The large language model (LLM) integrated into the Service that generates text responses.

"Content Safeguards"

The automated measures applied to the AI Model's outputs that are intended to prevent generation of content falling within §8 (Prohibited Conduct).

"User" / "you" / "your"

Any individual who accesses, registers for, or uses the Service in any capacity.

"User Content"

Any text, files, images, or other materials submitted by a User as input to the Service.

"AI Output"

Any text, image, or other content generated by the AI Model in response to User Content or prompts.

"Account"

A registered user profile granting access to authenticated features of the Service.

"Hallucination"

AI Output that is factually incorrect, fabricated, or misleading despite appearing authoritative or plausible.

"CSAM"

Child Sexual Abuse Material, meaning any content that sexually depicts or exploits minors, as defined under applicable law.

2. Acceptance of Terms

By accessing or using the Service in any manner, including but not limited to registering an account, visiting the website, submitting prompts, or using the API, you agree to be legally bound by these Terms of Service ("Terms"), our Privacy Policy, and all other policies incorporated herein by reference.

If you do not agree. If you do not agree with any provision of these Terms, you are not permitted to use the Service. You must immediately cease all use and delete your Account. Continued use after reading these Terms constitutes irrevocable acceptance.

These Terms constitute a legally binding agreement between you and Aelius AI. Where required by law, we obtain express confirmation of your agreement at the point of registration and at the point of first chat use.

We reserve the right to update these Terms at any time. Material changes will be communicated via email or in-app notice. Continued use of the Service following such notice constitutes acceptance of the revised Terms.

3. Eligibility and Age Requirements

3.1 Minimum Age

You must be at least 18 years of age (or the age of majority in your jurisdiction, whichever is higher) to create an Account or use the Service. The Service contains or may generate adult content, and access by minors is strictly prohibited.

Strict age enforcement. If we discover or reasonably believe that a user is under the minimum age, we will immediately terminate their Account, delete associated data, and reserve the right to report the incident to relevant authorities. Any person who creates an Account on behalf of a minor, or allows a minor to use their Account, is in material breach of these Terms.

3.2 Age Verification

By creating an Account, you represent and warrant under penalty of perjury that you meet the minimum age requirement. Your acknowledgment at registration constitutes a legal declaration of your eligibility. Depending on your jurisdiction and applicable law, we may be required to implement additional age verification mechanisms, which will be added as necessary.

3.3 Jurisdictional Restrictions

The Service may not be available, permitted, or legal in all jurisdictions. You are solely responsible for determining whether your use of the Service is lawful in your location. We make no representation that the Service is appropriate or available for use in any particular jurisdiction. If you access the Service from a jurisdiction where it is prohibited, you do so at your own risk and are personally liable for compliance with local laws.

3.4 Entity Use

If you are using the Service on behalf of a company, organization, or other legal entity, you represent and warrant that you have the authority to bind that entity to these Terms, and "you" in these Terms refers to both you individually and that entity.

4. Content Standards and Model Behavior

4.1 Nature of the AI Model

Aelius AI provides access to a large language model with content safeguards in place. Our safeguards are designed to be permissive rather than restrictive: the model is intended to engage substantively with mature, difficult, or controversial subject matter, including topics many commercial assistants decline by default, rather than refuse on the basis of keywords or tone. This reflects our view that adults are entitled to direct, useful answers.

Those safeguards are real, and they have limits. The Model is designed to decline requests that fall within the categories described in §8 (Prohibited Conduct), and we may adjust, strengthen, or relax its behavior at any time as our systems and obligations evolve.

4.2 No Guarantee of Filtering

Content safeguards are imperfect. No safety system catches everything, and no safety system is free of false positives. Accordingly, you acknowledge that the Model may, without limitation:

• Produce material that you find offensive, upsetting, or objectionable
• Discuss violence, sexuality, or other mature themes in the course of a legitimate request
• Take positions on controversial political, religious, or ideological questions
• State medically, legally, or scientifically inaccurate information with apparent confidence
• Decline or hedge on a request that you consider entirely reasonable
• Generate content that references or resembles the words of real persons

Neither the presence of safeguards nor any statement in these Terms constitutes a warranty that particular content will be produced or withheld.

4.3 AI Output Does Not Represent Our Views

All AI Output is generated by a statistical text model. It does not reflect the opinions, values, beliefs, or positions of Aelius AI or its operators. The Model has no moral stance, no political affiliation, and no intent, it generates statistically probable text continuations. Any appearance of personality, belief, or opinion in AI Output is a linguistic artifact, not an expression of any person's views.

4.4 Your Acknowledgment

By using the Service, you acknowledge and accept that:

• The Service is intended for adults and may address mature subject matter
• You may encounter output you find inaccurate, unhelpful, or objectionable
• You are choosing to use an AI system whose outputs cannot be guaranteed in advance
• You will report content that you believe violates these Terms rather than relying on automated filtering alone

User Acknowledgment: "I understand that Aelius AI applies content safeguards that are deliberately permissive and inherently imperfect. I accept that AI Output may be inaccurate or objectionable, and that no filtering system can guarantee otherwise."

4.5 Pictures from the Web and Adult Pictures

When the assistant shows pictures from the web in a reply, our server finds them through a picture search, fetches them from the sites they are on, and checks each one automatically before it is shown. Pictures from the web are available to signed-in accounts only. A picture that could show anyone under 18 is never shown, whatever was asked, and sexual pictures of a real, identifiable person are never shown.

Sexual pictures of adults, meaning pictures that show adults nude, in a sexual act, or posed or presented in a sexual way, are shown only on the website (aeliusai.com), and only if you turn on Show adult pictures (18+) in Settings > Privacy there. It is off unless you turn it on. By turning it on you confirm that you are 18 or older (or the age of majority where you live, if higher) and that you want to see such pictures. You can turn it off at any time, which also hides such pictures already in your chats.

The Aelius apps (iPhone, iPad and Android) never show sexual pictures, whatever this setting says, and the setting cannot be turned on in them. Revealing pictures that are not sexual (for example swimwear or lingerie, without nudity and not posed in a sexual way) need no setting and are shown on the website and in the apps.

Whether a picture is sexual or only revealing is decided by our automatic checks: a picture that shows nudity (including implied nudity), a sexual act or a sexual pose (for example lying on a bed, on all fours, legs spread, or a hand on the breasts or crotch), or that comes from a page about pornography, nudity or erotica, counts as sexual. The checks are automatic and can be wrong (see §4.2). Every picture can be reported (§12.3). Pictures from the web belong to their owners; we show each one with its source and a link to it.

5. AI Limitations and Accuracy Warning

5.1 Fundamental Nature of Language Models

Large language models, including the one powering Aelius AI, are statistical text prediction systems, not reasoning engines, knowledge databases, or intelligent agents. They generate output by predicting which words are likely to follow other words based on patterns in their training data. They do not "know" information, "understand" context, or "think" about their responses in any meaningful sense.

5.2 Hallucination, Fabrication of Information

The AI Model regularly and confidently fabricates information. This phenomenon, known as "hallucination," means the model may invent:

• Scientific studies, papers, and citations that do not exist
• Historical events, dates, and figures that are incorrect or entirely fictional
• Legal statutes, regulations, and case law that are wrong or non-existent
• Medical dosages, procedures, and diagnoses that may be dangerous
• Biographical details about real people that are false
• Statistics, data, and research findings that are made up
• Product names, prices, availability, and specifications that are inaccurate

There is no reliable way to tell from the output alone whether the AI is hallucinating. Confident, well-formatted, grammatically perfect responses may be entirely fabricated. You must independently verify any factual claim before acting on it.

5.3 Current Information and Web Research

The AI Model's own knowledge is fixed at a training cutoff date. Some features let it look things up: it can search the web, read web pages you or it select, and fetch data such as the weather. These lookups do not make its answers reliable. Search results can be out of date, incomplete or wrong, and the Model can misread, misquote or mis-summarise what it found, or answer from its older training instead of the result. When a lookup is not made, or fails, anything the Model says about recent events or current conditions may simply be out of date. Check current information against an authoritative source before relying on it.

5.4 Inconsistency and Contradiction

The AI Model may give different answers to identical questions at different times, contradict itself within a single conversation, or reverse its stated positions without warning. No AI Output should be treated as definitive, final, or consistent.

5.5 The AI Is Not a Person

The AI Model has no consciousness, feelings, relationships, or genuine understanding. Any appearance of empathy, personality, or emotional connection is a statistical pattern produced by training data, not genuine experience. You should not form emotional dependencies on the AI, treat it as a companion or therapist, or make life decisions based on AI "advice."

5.6 You Are Interacting With an AI

Every reply in Aelius AI, whether typed, spoken aloud in voice mode, sent as a notification or reminder, or returned through the API, is produced by an AI system. No human writes or reviews replies as they are made. The voice in voice mode is synthetic. Some settings make the Model write in a more natural, conversational style; that is a style, and it is still an AI. If you ask whether you are talking to a person, it is instructed to tell you plainly that it is not. This applies to shared projects (Orbits) too: whatever its creator's instructions say, an Orbit says it is an AI when asked.

Images the Service generates are AI-generated. Each one carries a machine-readable marker saying so (the IPTC "trained algorithmic media" digital source type, in the file's XMP metadata), and so does every picture you change with the picture editor; and documents the Service writes for you, including the changes it makes to a file you ask it to edit, are AI output too. If you publish AI Output where it could be mistaken for real, such as a realistic image of a real person or place, or text on a matter of public interest, you are responsible for disclosing that it is AI-generated where the law requires it (for example, Article 50 of the EU AI Act), and you must not remove or alter the marker to pass it off as authentic. See also "Misrepresentation" in §8.

Do not trust AI outputs blindly. Always treat AI Output as a starting point for investigation, not a conclusion. For anything that matters, your health, your money, your legal rights, your safety, consult a qualified human professional. The AI's confidence is not evidence of accuracy.

6. High-Risk Domains, Specific Warnings

The following domains carry elevated risk when AI-generated content is relied upon. Our disclaimers in §13 and §14 apply with full force, but we identify these areas explicitly because the potential for harm is particularly serious.

6.1 Medical and Healthcare

AI Output must never be used as a substitute for professional medical advice, diagnosis, or treatment. The Model may suggest incorrect diagnoses, dangerous drug interactions, inappropriate dosages, contraindicated treatments, or outdated medical guidance. If you have a health concern, consult a licensed physician immediately. In an emergency, call emergency services, do not ask an AI.

6.2 Mental Health

The AI is not a therapist, counselor, or crisis responder. It has no clinical training and its responses may be harmful to individuals experiencing mental health difficulties. If you are in crisis, contact emergency services or a crisis hotline immediately. Do not rely on the AI for emotional support or therapeutic guidance.

Our suicide and self-harm protocol. When a message reads as the sender's own thoughts of suicide or self-harm, the Service tells the Model so before it replies. The Model is instructed to respond with care, ask whether you are safe, remind you that it is an AI, and refer you to people who can help: in the US, call or text 988 (Suicide & Crisis Lifeline); in the UK and Ireland, Samaritans on 116 123; elsewhere, findahelpline.com lists free local services; and your local emergency number if you are in immediate danger. It is instructed never to provide methods or encouragement. The check is automatic and imperfect: it can miss a message and can misread a joke or a story. We count how often a referral is made, without keeping the text of the message, so that we can report those numbers where the law requires it. If you are in danger now, contact emergency services or one of the lines above directly; do not wait for an AI.

6.3 Legal Advice

AI Output does not constitute legal advice and is not a substitute for consultation with a qualified attorney. The Model may state law incorrectly, cite non-existent cases or statutes, misapply legal principles to your facts, or give advice that is specific to a different jurisdiction. Acting on AI-generated legal information without attorney review may result in serious legal consequences.

6.4 Financial and Investment

AI Output regarding financial matters, investments, taxes, or economic predictions is not regulated financial advice and may be materially inaccurate. The Model cannot account for your personal financial situation, risk tolerance, or current market conditions. Rely only on advice from licensed and regulated financial advisors.

6.5 Physical Safety

AI-generated instructions for physical activities, construction, electrical work, chemical handling, firearms, or other safety-critical tasks may be incorrect, dangerous, or potentially lethal. Never follow AI safety-critical instructions without expert human verification.

6.6 Child Safety

Any use of the Service in connection with children, whether seeking parenting advice, educational content, or any other child-related matter, must be approached with extreme caution. AI Output is not a safe guide for decisions affecting children's wellbeing, development, or safety.

7. User Accounts

7.1 Account Creation

To access certain features of the Service, you must register for an Account. You agree to provide accurate, current, and complete information during registration and to keep that information updated. Providing false registration information, including a false age declaration, is a material breach of these Terms.

7.2 Account Security

You are solely responsible for maintaining the confidentiality of your Account credentials and for all activity that occurs under your Account. You must immediately notify us of any unauthorized access to or use of your Account. Aelius AI is not liable for any loss or damage arising from unauthorized Account access resulting from your failure to maintain credential security.

7.3 One Account Per Person

Each individual may maintain only one active Account. Creating multiple Accounts to circumvent a ban, restriction, or these Terms is prohibited and will result in the termination of all associated Accounts.

7.4 Non-Transferability

Your Account is personal to you and may not be transferred, sold, or assigned to any other person or entity without our express written consent.

7.5 Email Verification

Account creation requires email verification. You must provide a valid email address you control. Failure to verify your email within a reasonable period may result in Account deactivation.

7.6 Usage Limits

Parts of the Service are provided with usage limits, including the credit allowances on Aelius Code, message, image and research limits, and rate limits. Each limit applies to you as a person, not to an Account, key, device or network address. You must not avoid, reset, or exceed a limit by any means other than waiting for it to reset or buying a plan that raises it. In particular, you must not:

• create, use, borrow, buy, or coordinate additional Accounts or API keys to continue after one has reached its limit;
• use proxies, VPNs, rotating IP addresses, or other means of changing or concealing your network or identity in order to avoid a limit or a network-based restriction;
• share, pool, or rotate Accounts or keys among several people to combine their allowances; or
• otherwise interfere with how usage is measured or enforced.

We may treat Accounts that we reasonably believe are operated by the same person, or used together to avoid a limit, as one Account. We may pause, limit, suspend, or terminate any or all of them, and revoke their keys, without notice or refund. Several people legitimately sharing one network is not a violation; if a limit affects you in that situation, contact us and we can mark the network as shared.

7.7 Using the Service to Build or Train AI

You may use the Service and AI Output for your own work, including work you are paid for. You may not use them to build, train or improve another AI system. In particular, you must not:

• use AI Output, or any record of your conversations with the Service, to train, fine-tune, distill, evaluate or otherwise develop an AI or machine-learning model, or to create a dataset for anyone who will;
• use the Service to do data labelling, annotation, captioning, rating, ranking or response-writing work that is supplied to an AI developer or to a data-labelling or crowd-work platform (for example, writing image captions, model answers or quality ratings that will be used to train or evaluate a model);
• systematically query the Service to extract the Model's behaviour, outputs, instructions or capabilities, including by generating large sets of prompts and responses; or
• help, pay or arrange for anyone else to do any of the above.

This applies however the work is done: by hand or with automation, through one Account or several. It does not prevent you from using AI Output in your own writing, studies or business, or from discussing, testing or reviewing the Service publicly. If you would like to use the Service for AI research or data work, contact us; we may agree to it in writing.

We may limit, suspend or terminate an Account we reasonably believe is being used in this way, and the provisions of §7.6 on related Accounts apply.

7A. Aelius Pro, Billing and Gifts

Aelius AI is free to use. Aelius Pro is an optional paid subscription that raises some limits and adds some features. This section applies only if you buy Pro or give it to someone. Numbering starts at 7A so that references to the other sections of these Terms stay the same.

7A.1 What Pro Is

Pro currently includes: a longer conversation memory (up to 256k tokens of context in a chat, eight times the 32k tokens of conversation a free chat holds), priority access to our largest model (a Pro account is not answered by our smaller model, except where you choose it yourself), higher-resolution images, unlimited image edits (a free account can edit a set number of images a day, currently five, while making new images stays unlimited), unlimited storage for your images and files, ten times the free web research allowance, ten times the free Aelius Code allowance, and extended thinking and deep research. The current details and numbers are shown in Settings → Billing before you buy.

Pro does not remove the rules in these Terms. The usage-limit rules in §7.6 apply to Pro allowances too, and Pro is for one person: it may not be shared, resold or pooled.

7A.2 Who Sells Pro: Link as Reseller

Pro is sold through Stripe's Managed Payments service. The seller of record for your purchase is Link (a Stripe company), which buys access to Pro from us and resells it to you. That means:

• you pay Link, and your card statement shows LINK.COM* followed by our name;
• Link sends your receipts and invoices, calculates and collects any sales tax, VAT or GST, and files it with the tax authorities;
• Link handles payment and billing questions, and you can view your orders, change your card, and cancel at link.com;
• your payment is also governed by Link's own terms, which you see and accept at checkout. If Link's terms and these Terms conflict about the payment itself, Link's terms apply to the payment. These Terms continue to govern the Service.

We never see your full card number. Pro cannot be bought from countries where Link does not sell, which currently include China, Cuba, Iran, North Korea, Russia and Syria.

7A.3 Price and Taxes

Pro costs US$3.99 a month in most countries and US$3.70 in others, or the local-currency price shown to you in the app and before checkout (for example £3.29 or €3.79). Local-currency prices follow exchange rates, so the price offered to new subscribers can change; once you subscribe, your monthly price stays what it was when you subscribed. The price you see in the app is the price on the checkout page, and it includes any sales tax, VAT or GST that applies where you live, so the amount on the checkout page is the amount you pay each month. Your bank may add its own fees, which we do not control.

Earlier subscribers keep their price. If you subscribed before October 2, 2026, at US$2.99 a month or the local price shown to you then, you keep paying that same price for as long as that subscription continues. If it ends, for any reason, and you subscribe again later, the price at that time applies.

If we change the price, we will tell you by email at least 30 days before your first renewal at the new price, and you can cancel before then.

7A.4 Automatic Renewal

Your subscription renews automatically every month, and the monthly price is charged to your payment method on each renewal date, until you cancel. The renewal date is the day of the month you first subscribed. By subscribing you agree to these recurring charges. You can see your next renewal date and amount at any time in Settings → Billing.

After you subscribe we email you a confirmation that repeats these renewal terms, the price, and how to cancel, and Link emails you a receipt for every payment.

7A.5 How to Cancel

You can cancel online at any time, without contacting us:

• in the app or on the website, go to Settings → Billing and press Cancel subscription; or
• sign in at link.com and cancel the subscription there.

Cancelling stops all future charges. You keep Pro until the end of the month you have already paid for, and then your account returns to the free plan with all of your chats and files. We do not refund the unused part of a month when you cancel, except where §7A.7 or the law says otherwise. Deleting your account also cancels your subscription immediately.

7A.6 Failed Payments

If a renewal payment fails, Pro is paused straight away and your account works on the free plan until the payment succeeds. We show a notice in the app, and Link may email you. You can update your card in Settings → Billing or at link.com; when the payment then goes through, Pro comes back at once. Link retries a failed payment for a period, and if it still has not succeeded, the subscription is cancelled.

7A.7 Refunds and Your Right to Cancel

Payments are generally non-refundable, except:

• where the law of the country you live in gives you a right to withdraw or a refund, including the cooling-off period for online purchases in the EU and UK. Link applies these rights for us. Because Pro starts immediately, by subscribing you ask for it to start during any cooling-off period, and where the law allows, you may lose the right to withdraw once it has started, or receive a proportionate refund;
• Link may refund a payment within 60 days in some cases, for example to resolve a billing problem;
• gifts, as described in §7A.8; and
• where we end or materially reduce Pro, as described in §7A.9.

If an account is suspended or terminated for breaking these Terms, including §7.6, no refund is due for the remaining period.

7A.8 Gifts

You can pay once for one to three months of Pro for someone else, by entering their username or email address. The gift is not a subscription and is never charged again.

• The person you name receives an email with a link to accept the gift. Pro starts only when they accept it.
• If they decline it, or do not accept it within 30 days, you are refunded to the payment method you used. In a few countries the law requires the tax on a refunded sale to be kept, so the refund there is the price less that tax.
• A person can hold at most three months of gifted Pro at a time, counting gifts running and gifts waiting to be accepted.
• Anyone who opens the link in the gift email and signs in can accept it, so send it only to an address you trust.
• If the recipient already has, or later takes out, a paid subscription, the gift is used first and their paid subscription is not charged until the gifted months end.
• Once accepted, a gift cannot be refunded, transferred or exchanged for money.

By entering someone's email address, you confirm you know them and believe they would want to hear from us about the gift. We use that address only to send the gift email and to deliver the gift.

7A.9 Changes to Pro

The features and limits of Pro may change as the Service changes, and a limit may be lowered for everyone to keep the Service running for all users. If we remove a main feature of Pro or materially reduce what Pro gives you, we will tell you in advance where we reasonably can, and you may cancel. If we stop offering Pro altogether, we will refund the unused part of the current month. Changes to the free plan do not require any notice beyond §20.

7A.10 Pro Bought in the iPhone and iPad App

If you buy Pro inside the Aelius app for iPhone or iPad, you buy it from Apple through In-App Purchase, not from Link. For that subscription the following applies in place of §7A.2 to §7A.7:

• Apple is the seller. Apple charges the payment method on your Apple Account, sends your receipts, and calculates and collects any sales tax, VAT or GST.
• The price is the one shown in the app before you confirm, in the currency of your App Store country. It can differ from the price on our website.
• The subscription renews automatically every month, and Apple charges your Apple Account in the 24 hours before the current period ends, until you cancel.
• You manage or cancel it in Settings on your iPhone or iPad (tap your name, then Subscriptions), or in the app under Settings → Billing → Manage subscription. Cancel at least 24 hours before the period ends to avoid the next charge. You keep Pro until the end of the period you have paid for. Deleting your Aelius account does not cancel a subscription bought through Apple.
• Refunds are decided by Apple under its own policies; you can ask for one at reportaproblem.apple.com. If Apple refunds a payment, Pro for that period ends.
• Your purchase is also governed by Apple's Media Services Terms and Conditions. Apple is not a party to these Terms and is not responsible for the Service.

A subscription bought through Apple belongs to the Aelius account that was signed in when it was bought. Restoring it while signed in to a different account does not move it. An account billed through Apple cannot also subscribe on our website, and the reverse, so that nobody pays twice.

8. Prohibited Conduct

The following conduct is strictly prohibited and will result in immediate Account termination, potential reporting to law enforcement, and civil or criminal liability where applicable.

Note on enforcement. The existence of this prohibited conduct list does not mean these actions are always technically prevented. Content safeguards reduce, but do not eliminate, the possibility of prohibited output. Prohibited use is a contractual and legal matter as well as a technical one, and we monitor for and investigate reported violations.

9. User Content and Intellectual Property

9.1 Ownership of Your Input

You retain ownership of any content you submit as input to the Service ("User Content"), subject to the license grant below and all applicable laws.

9.2 License Grant to Aelius AI

By submitting User Content, you grant Aelius AI a limited, non-exclusive, royalty-free license to process, store, and transmit your User Content solely for the purpose of providing the Service to you. This license does not extend to training AI models, sharing with third parties, or commercial exploitation of your content.

9.3 AI Output Ownership

The ownership of AI Output is legally complex and varies by jurisdiction. AI-generated content may not qualify for copyright protection under current law in many jurisdictions as it lacks human authorship. You are responsible for determining the legal status of any AI Output you use commercially or publicly. Whoever owns it, you may not use AI Output for any purpose §7.7 prohibits, and that restriction continues after your Account ends.

9.4 No Warranty of Originality

We make no representation that AI Output is original, non-infringing, or free from third-party intellectual property claims. The AI may reproduce or paraphrase copyrighted material from its training data. You are solely responsible for any infringement arising from your use of AI Output.

9.5 Your Representations Regarding Input

By submitting User Content, you represent and warrant that you have all rights necessary to do so, that your content does not infringe any third-party rights, and that your content does not violate any law or these Terms.

9.6 Projects and Shared Projects (Orbits)

9.6.1 Your project content. Instructions, knowledge and starter prompts you add to a project are your User Content under 9.1 and 9.5, and the licence in 9.2 covers processing them to answer chats in the project, including chats of people you share it with.

9.6.2 Sharing. When you share a project you are responsible for what its instructions, knowledge, name, description and starter prompts contain, as if you had sent them to each person who uses it. Do not put in a shared project anything you do not have the right to share, personal data about other people, or secrets. You must not use a shared project to: (a) collect passwords, payment details or other personal data from the people who use it; (b) pretend to be a real person or organisation, or present the Orbit as human; (c) get around Aelius's rules or safety behaviour (instructions that try are refused when you save them); or (d) do anything else Section 8 prohibits. Aelius's rules, safety checks and the protocol in 6.2 apply in every Orbit chat exactly as in any other chat.

9.6.3 Using someone else's Orbit. An Orbit is made by another user, not by Aelius AI, and we do not review it before it is shared. Its answers are AI Output under Sections 4 and 5. Your chats with it belong to your account and are not visible to its creator.

9.6.4 Our rights. We may disable a shared link, or remove a shared project, that breaks these Terms or is reported to us, and may suspend accounts that misuse sharing.

10. User Responsibilities

Because automated safeguards cannot anticipate every situation, a substantial share of responsible use falls to you. By using the Service, you accept the following obligations:

1. Verify Before Acting: Independently verify all factual claims in AI Output before acting on them, especially in medical, legal, financial, or safety-critical contexts.
2. Apply Your Own Judgment: The AI will not warn you, refuse, or stop you from doing something harmful or unwise. You must apply critical thinking at all times.
3. Protect Your Credentials: Maintain the security of your Account credentials. Do not share access with others.
4. Comply with All Laws: Ensure your use of the Service complies with all applicable laws in your jurisdiction. The Service's permissive design does not make illegal activity legal.
5. No Dependency Formation: Do not form emotional or functional dependencies on the AI that could be harmful to your wellbeing or decision-making.
6. Report Violations: Report suspected illegal use of the platform or security vulnerabilities to us promptly.
7. Understand What You're Using: You acknowledge that AI systems are imperfect, unpredictable statistical models and that no output should be treated as authoritative.
8. Shared Links: If you publish a chat with a share link, you are responsible for what it shows, including anything it says about other people. We may disable a shared link that breaks these Terms or that we are lawfully asked to remove.
9. Age Gate Your Environment: If others, particularly minors, may access your device or Account, you are responsible for preventing their access to this Service.

11. Intellectual Property and DMCA

11.1 Our Intellectual Property

The Aelius AI name, logo, website design, and all platform code and infrastructure are the exclusive intellectual property of Aelius AI and its operators. You may not copy, reproduce, distribute, or create derivative works based on our platform without express written permission.

11.2 DMCA Safe Harbor

Aelius AI respects intellectual property rights and complies with the Digital Millennium Copyright Act (DMCA) to the extent applicable. We maintain a designated agent for receipt of copyright infringement notices.

11.3 DMCA Notice Requirements

If you believe content on our Service infringes your copyright, you must submit a notice containing:

• A physical or electronic signature of the copyright owner or authorized agent
• Identification of the copyrighted work claimed to be infringed
• Identification of the infringing material and its location on the Service
• Your contact information (name, address, phone, email)
• A statement that you have a good faith belief that the use is not authorized by the copyright owner
• A statement that the information in the notice is accurate and, under penalty of perjury, that you are authorized to act on behalf of the copyright owner

Send DMCA notices to our legal contact at the address provided in §22.

11.4 Counter-Notification

If you believe content was removed in error, you may file a DMCA counter-notification. Submitting a false counter-notification subjects you to perjury liability. The counter-notification procedure is available on request.

11.5 Repeat Infringer Policy

In accordance with the DMCA, we maintain a policy of terminating Accounts of users who are repeat copyright infringers.

12. CSAM Policy and Mandatory Reporting

Zero tolerance, federal law. Child Sexual Abuse Material (CSAM) is the most serious category of prohibited content. This prohibition is absolute and unconditional. It applies regardless of whether the content is photographic, illustrated, AI-generated, textual, or in any other format. Any depiction that sexualizes or exploits minors is prohibited without exception.

12.1 Absolute Prohibition

Any attempt to use Aelius AI to generate, view, distribute, or otherwise interact with CSAM is strictly prohibited. This includes:

• Prompts that attempt to generate sexual content involving minors, regardless of whether the minor is fictional or real
• Prompts that sexualize or romanticize children in any context
• Any content that could be used to groom, exploit, or harm minors
• Uploading CSAM or attempting to use it as input to the Service

12.2 Mandatory Reporting Obligations

Under 18 U.S.C. § 2258A, any electronic service provider that becomes aware of an apparent violation of certain federal child exploitation laws has a mandatory duty to report to the National Center for Missing & Exploited Children (NCMEC) CyberTipline. Aelius AI will comply with this obligation without exception. Reporting to NCMEC may be followed by investigation and prosecution by federal law enforcement.

12.3 Reporting Mechanism

If you become aware of any CSAM on the Service or any attempt by another user to generate CSAM, you must report it immediately via our contact page or at the legal address in §22. We will act on all reports promptly and report to NCMEC and relevant authorities as required by law.

Every picture from the web shown in a reply (§4.5) has a Report picture action: the … on the picture, a long press on a phone, or the flag in the full-screen view. A report removes the picture from your chat at once, and that picture is not shown to you again. A report that a picture shows a minor (after you confirm it) also blocks that picture for every user at once, deletes every other copy we kept of it, and is placed first in our review queue. As federal law requires (18 U.S.C. § 2258A), we keep one encrypted copy of the picture and the details of the report, which only our operators can open, for at least one year, or longer if law enforcement asks us to, and where the law requires it we report it to NCMEC (§12.2). Only a picture you were shown can be reported this way, and one account can block at most 10 pictures a day; further reports are still recorded for review. We may lift a block after review.

12.4 No Technical Prevention is a Guarantee

We apply content safeguards and take active steps to detect prohibited CSAM-related use, but no technical measure is perfect. Our prohibition is a legal and contractual absolute; violations will be reported and prosecuted to the fullest extent of the law.

13. Disclaimer of Warranty

THE SERVICE AND ALL AI OUTPUT ARE PROVIDED "AS IS" AND "AS AVAILABLE" WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED. TO THE FULLEST EXTENT PERMITTED BY APPLICABLE LAW, AELIUS AI AND ITS OPERATORS EXPRESSLY DISCLAIM ALL WARRANTIES, INCLUDING BUT NOT LIMITED TO: (A) IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE, AND NON-INFRINGEMENT; (B) WARRANTIES THAT THE SERVICE WILL BE UNINTERRUPTED, ERROR-FREE, SECURE, OR FREE OF VIRUSES; (C) WARRANTIES REGARDING THE ACCURACY, RELIABILITY, COMPLETENESS, TIMELINESS, OR SUITABILITY OF ANY AI OUTPUT; AND (D) WARRANTIES THAT AI OUTPUT WILL MEET YOUR REQUIREMENTS OR EXPECTATIONS.

We do not warrant that:

• The Service will operate without interruption, delay, or technical error
• Any AI Output is accurate, truthful, complete, or appropriate for any purpose
• The Service will be free from viruses, malware, or other harmful components
• Any defects in the Service will be corrected
• The Service will be compatible with your hardware, software, or network environment

Any reliance you place on AI Output is strictly at your own risk. We are not responsible for the results of decisions made based on AI Output.

14. Limitation of Liability

TO THE MAXIMUM EXTENT PERMITTED BY APPLICABLE LAW, IN NO EVENT SHALL AELIUS AI, ITS OPERATORS, EMPLOYEES, AFFILIATES, LICENSORS, OR SERVICE PROVIDERS BE LIABLE TO YOU OR ANY THIRD PARTY FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, EXEMPLARY, OR PUNITIVE DAMAGES, INCLUDING BUT NOT LIMITED TO: LOSS OF PROFITS, REVENUE, DATA, GOODWILL, OR ANTICIPATED SAVINGS; PERSONAL INJURY OR DEATH CAUSED BY RELIANCE ON AI OUTPUT; EMOTIONAL DISTRESS; OR ANY OTHER INTANGIBLE LOSSES, REGARDLESS OF WHETHER SUCH DAMAGES WERE FORESEEABLE AND WHETHER WE WERE ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

14.1 Aggregate Liability Cap

To the fullest extent permitted by applicable law, our total aggregate liability to you for all claims arising out of or relating to these Terms or your use of the Service shall not exceed the greater of: (a) the amount you paid to Aelius AI in the twelve (12) months immediately preceding the event giving rise to the claim, or (b) one hundred dollars (USD $100).

14.2 Specific Exclusions

Without limiting the foregoing, Aelius AI expressly excludes all liability for:

• Physical, psychological, financial, or legal harm arising from reliance on AI Output
• Medical harm caused by following AI medical suggestions
• Financial loss caused by following AI financial suggestions
• Harm caused by offensive or disturbing AI-generated content that users voluntarily accessed
• Loss or corruption of your data or conversation history
• Unauthorized access to your Account resulting from your failure to maintain security
• Service interruptions, downtime, or data loss
• Actions taken by third parties in response to your use of AI Output

14.3 Essential Basis

The limitations in this section are a fundamental element of the basis of the bargain between you and Aelius AI. The Service would not be provided to you without these limitations.

14.4 Jurisdictional Variation

Some jurisdictions do not allow the exclusion of certain warranties or the limitation of liability for certain types of damages. In such jurisdictions, our liability will be limited to the greatest extent permitted by law.

15. Indemnification

You agree to defend, indemnify, and hold harmless Aelius AI, its operators, officers, directors, employees, contractors, agents, licensors, and service providers from and against any and all claims, damages, judgments, awards, losses, liabilities, costs, and expenses (including reasonable attorneys' fees) arising out of or relating to:

• Your use of or access to the Service
• Your violation of any provision of these Terms
• Your violation of any third-party rights, including intellectual property, privacy, or publicity rights
• Your violation of any applicable law, regulation, or court order
• Any content you submit to the Service
• Any action you take based on AI Output
• Any misrepresentation you make in connection with the Service
• Injury to any person caused by your use of AI Output

We reserve the right, at your expense, to assume exclusive defense and control of any matter for which you are required to indemnify us. You agree to cooperate fully with our defense of such claims.

16. Dispute Resolution, Binding Arbitration and Class Action Waiver

Important, read carefully. This section affects your legal rights. It requires binding arbitration for disputes and waives your right to a jury trial and class action participation. If you do not accept these terms, do not use the Service.

16.1 Informal Resolution First

Before initiating any formal dispute process, you agree to first contact us at the address in §22 and attempt to resolve the dispute informally by giving us at least thirty (30) days to respond. Many disputes can be resolved quickly and informally.

16.2 Binding Arbitration

If informal resolution fails, any dispute, claim, or controversy arising out of or relating to these Terms or your use of the Service shall be resolved by binding individual arbitration, rather than in court, except as provided in §16.5. The arbitration will be administered by the American Arbitration Association ("AAA") under its Consumer Arbitration Rules then in effect (available at adr.org), as modified by this §16, and is governed by the Federal Arbitration Act. Filing, administration and arbitrator fees are paid as those Rules provide for consumer disputes, which puts most of them on us. The arbitration will take place in the county where you live or, if you prefer, by video, telephone or written submissions. The arbitrator may award you the same individual relief a court could. The decision of the arbitrator shall be final and binding and may be entered as a judgment in any court of competent jurisdiction.

16.2A Your Right to Opt Out

You can reject this arbitration agreement. Email [email protected] within thirty (30) days of first accepting these Terms (or, if you accepted an earlier version, within thirty days of September 29, 2026), with the subject line "Arbitration Opt-Out" and your username or account email. Opting out does not affect any other part of these Terms or how we treat your account; disputes between us will then go to the courts described in §17.

16.3 Class Action Waiver

YOU AND AELIUS AI AGREE THAT EACH PARTY MAY BRING CLAIMS AGAINST THE OTHER ONLY IN AN INDIVIDUAL CAPACITY AND NOT AS A PLAINTIFF OR CLASS MEMBER IN ANY PURPORTED CLASS, CONSOLIDATED, OR REPRESENTATIVE ACTION. The arbitrator may not consolidate more than one person's claims and may not preside over any form of representative or class proceeding. If this class action waiver is found unenforceable, then the entire arbitration provision in this §16 shall be null and void.

16.4 Jury Trial Waiver

TO THE EXTENT PERMITTED BY APPLICABLE LAW, YOU AND AELIUS AI WAIVE YOUR RESPECTIVE RIGHTS TO A TRIAL BY JURY with respect to any dispute that is not subject to arbitration under these Terms.

16.5 Exceptions

Notwithstanding §16.2, either party may seek injunctive or other equitable relief in a court of competent jurisdiction where necessary to prevent irreparable harm pending arbitration. Either party may also pursue claims in small claims court that qualify for such court's jurisdiction. Nothing in this §16 waives your right to seek public injunctive relief where applicable law does not permit that right to be waived; any such claim will be decided by a court under §17, after the individual claims in arbitration are resolved.

16.6 Statute of Limitations

To the extent permitted by applicable law, any claim arising under these Terms must be brought within one (1) year of the event giving rise to the claim. Claims brought after this period are permanently barred.

17. Governing Law and Jurisdiction

17.1 Governing Law

These Terms and any dispute arising out of or in connection with them or their subject matter or formation (including non-contractual disputes or claims) shall be governed by and construed in accordance with the laws of the State of California, United States of America, and applicable United States federal law, without regard to its conflict of law provisions.

17.2 Jurisdiction for Court Proceedings

For disputes not subject to arbitration under §16, you agree to submit to the exclusive jurisdiction of the state and federal courts located in the State of California. Nothing in this section deprives you of any protection you have under the mandatory law of the country where you live. You waive any objection to the exercise of personal jurisdiction over you by such courts.

17.3 Language

These Terms are written in English. In the event of any conflict between the English version and a translated version, the English version shall prevail.

18. International Use and Regulatory Compliance

18.1 Global Access

The Service is operated from the United States. We make no representation that the Service is appropriate for use in all countries. If you access the Service from outside the United States, you do so on your own initiative and are solely responsible for compliance with local laws.

18.2 EU and EEA Users, EU AI Act Notice

If you are located in the European Union or European Economic Area, the EU AI Act applies to the Service. In line with its transparency rules (Article 50): you are told, before you start and in §5.6, that you are interacting with an AI system; images the Service generates carry a machine-readable marker identifying them as AI-generated; and the Service does not use emotion recognition or biometric categorisation. If you publish AI Output that could be mistaken for authentic, including realistic images of real people, places or events, or text informing the public on matters of public interest, the Act may require you to disclose that it was artificially generated or manipulated. We reserve the right to restrict Service access to EU/EEA users where required by law.

18.3 UK Users, Online Safety Act Notice

If you are located in the United Kingdom, the Online Safety Act 2023 imposes obligations on online services regarding legal but harmful content and age verification. UK users acknowledge that the adult-oriented nature of the Service may implicate these regulations. We reserve the right to implement additional compliance measures for UK users as required by law.

18.4 Export Controls

You may not use or export the Service in violation of U.S. export laws and regulations. You represent that you are not located in a country subject to a U.S. government embargo or designated as a "terrorist supporting" country, and that you are not on any U.S. government prohibited or restricted party list.

18.5 Age Verification Laws

Various jurisdictions (including several U.S. states, the UK, and EU member states) have enacted or are enacting laws requiring technical age verification for adult content platforms. We actively monitor these legal developments and will implement additional age verification mechanisms for affected jurisdictions as required. Continued use of the Service after implementation of such mechanisms constitutes acceptance of the additional verification requirements.

19. Termination

19.1 Termination by Us

We may suspend or permanently terminate your Account and access to the Service at any time, with or without notice, for any reason, including but not limited to: violation of these Terms, illegal activity, fraudulent conduct, abuse of other users, or for any reason we determine in our sole discretion. We are not obligated to provide reasons for termination. If we terminate a paying Pro account for a reason other than a breach of these Terms, we will cancel the subscription and refund the unused part of the current month.

19.2 Termination for Cause, Immediate Effect

Violations of §8 (Prohibited Conduct), §12 (CSAM), or any conduct that may constitute criminal activity will result in immediate and permanent Account termination without notice, and may be reported to law enforcement.

19.3 Termination by You

You may terminate your Account at any time by contacting us or using the Account deletion feature. Upon termination, your right to access the Service immediately ceases. If you have a Pro subscription, see §7A.5.

19.4 Effect of Termination

Upon termination, all provisions of these Terms that by their nature should survive will survive, including §7.7 (Using the Service to Build or Train AI), §9 (User Content), §11 (IP & DMCA), §13 (Disclaimer), §14 (Limitation of Liability), §15 (Indemnification), §16 (Arbitration), and §17 (Governing Law).

19.5 Data Upon Termination

Upon Account termination, we will delete your chat history and Account data in accordance with our Privacy Policy. Where we are required to retain data by law (e.g., for law enforcement investigations), we will do so for the required period.

20. Modifications to Terms and Service

20.1 Changes to Terms

We reserve the right to modify these Terms at any time. We will provide notice of material changes via email (to your registered address) or via in-app notification. Non-material changes may be made without notice.

20.2 Continued Use Means Acceptance

Your continued use of the Service following any notice of changes to these Terms constitutes your acceptance of the updated Terms. If you do not accept the updated Terms, you must stop using the Service.

20.3 Changes to the Service

We reserve the right to modify, suspend, or discontinue the Service (or any part thereof) at any time with or without notice. We will not be liable to you or any third party for any modification, suspension, or discontinuation of the Service.

21. Miscellaneous Provisions

21.1 Entire Agreement

These Terms, together with the Privacy Policy and any other policies incorporated by reference, constitute the entire agreement between you and Aelius AI with respect to the Service and supersede all prior and contemporaneous understandings, negotiations, and agreements.

21.2 Severability

If any provision of these Terms is found to be invalid, illegal, or unenforceable under applicable law, that provision will be modified to the minimum extent necessary to make it enforceable, or severed if modification is not possible. The remaining provisions of these Terms will continue in full force and effect.

21.3 Waiver

Our failure to enforce any right or provision of these Terms shall not constitute a waiver of that right or provision. Any waiver must be in writing and signed by an authorized representative of Aelius AI to be legally effective.

21.4 No Third-Party Beneficiaries

These Terms are between you and Aelius AI. Nothing in these Terms shall create or confer any rights or remedies on any third party.

21.5 Assignment

You may not assign or transfer your rights or obligations under these Terms without our prior written consent. We may assign our rights and obligations under these Terms without restriction.

21.6 Force Majeure

We will not be liable for any failure or delay in performance resulting from circumstances beyond our reasonable control, including natural disasters, government actions, internet outages, or third-party service failures.

21.7 Headings

Section headings in these Terms are for convenience only and shall not affect the interpretation of these Terms.

22. Contact and Legal Notices

For all legal notices, DMCA notices, CSAM reports, privacy requests, and general inquiries regarding these Terms, please contact us through the following channel:

Email: [email protected]
Subject line for DMCA: DMCA Notice — [Your Name]
Subject line for CSAM reports: URGENT: CSAM Report
Subject line for legal notices: Legal Notice — [Your Name]

We will acknowledge receipt of legal notices within five (5) business days. Response times for CSAM reports will be as fast as possible, typically within 24 hours, given our mandatory reporting obligations.