Terms of Service — NeuralAI

01The agreement

By accessing or using the Services — including by sending a message to the live NEA1 chat — you agree to these Terms and to our Privacy Policy, which is incorporated into these Terms by reference. If you do not agree, do not use the Services.

If you use the Services on behalf of a company, organization, or other entity, you represent that you have authority to bind that entity to these Terms, and “you” refers to that entity. Additional terms may apply to specific offerings (for example, a separate written agreement for research partnerships or model access). If there is a conflict between these Terms and a signed written agreement between you and NeuralAI, the signed agreement controls for the subject matter it covers.

Nothing in these Terms limits any rights you may have under a license that applies to a specific downloadable artifact — in particular, the Apache License, Version 2.0, which governs the NEA1 model weights as described in Section 9.

02The services

NeuralAI is an independent artificial-intelligence research lab based in West Memphis, Arkansas, founded by DeAndrew Preston Harris. The “Services” covered by these Terms currently include:

The NeuralAI website (neuralai.wasmer.app), including the model specifications, benchmark results, roadmap, and company information published there;
The live NEA1 chat (hosted on a workers.dev address and embedded in the website), a public demonstration that lets visitors converse with the currently-serving NEA1 checkpoint, including multi-turn conversation and limited tool use;
The inquiry form on the website, through which you may contact the lab about research partnerships, model access, and general questions;
Links to open artifacts, including the NEA1 weights and model card on Hugging Face and the project repositories on GitHub. Your use of those platforms is also governed by their own terms.

NEA1 is a language model designed, tokenized, and trained from scratch by NeuralAI. The serving checkpoint, version, and published evaluation figures can change as new versions are gated and released; figures on the website describe the checkpoint they are labeled with and are not a promise of performance for any other version or for any particular use.

03Eligibility

You must be at least 13 years old to use the Services. If you are under 18 (or the age of majority where you live), you may use the Services only with the consent and supervision of a parent or legal guardian, who is responsible for your use. The Services are not directed to children under 13, and you may not use the Services if you are under 13.

You may not use the Services if you are barred from doing so under the laws of the United States or of the place where you access them, or if we have previously suspended or terminated your access for a violation of these Terms. You are responsible for making sure your use of the Services complies with the laws that apply to you, including export-control and sanctions laws: the Services and the model artifacts may not be used in or for the benefit of comprehensively sanctioned jurisdictions, or by persons on U.S. restricted-party lists, where applicable law prohibits it.

04Access and security

The public Services currently do not require an account, a password, or payment. If we introduce accounts or authenticated access in the future, you will be responsible for keeping your credentials confidential and for all activity under your account, and you agree to notify us promptly of any unauthorized use.

You agree not to attempt to gain unauthorized access to the Services, their underlying systems, or any non-public repository, key, token, configuration, or data store; not to probe, scan, or test the vulnerability of the Services except in good-faith security research that does not degrade the Services and that is reported to us through the inquiry form; and not to circumvent rate limits, quotas, access controls, or other technical measures we use to keep the public demo available and fair for everyone.

The chat demonstration runs on capacity-limited infrastructure. We may apply reasonable rate limits, message-length limits, and concurrency limits, and responses may be delayed while the inference backend wakes or scales. Limits may change without notice as we balance availability and cost.

05AI output and reliance

How output is produced. NEA1 generates text token by token based on statistical patterns learned in training. It does not “know” facts the way a person or a database does, it has no awareness of current events beyond what a tool call returns in the moment, and it can produce text that is inaccurate, fabricated (including invented citations, URLs, figures, and quotations), biased, offensive, or internally contradictory. Two identical prompts can produce different answers, and follow-up answers can contradict earlier ones.

Your responsibility. You are solely responsible for evaluating any output before you rely on, publish, or act on it, and for any decisions you make based on output. Use human judgment and authoritative sources for anything consequential — including health, safety, legal, financial, educational, employment, and engineering decisions.

Not professional advice. Output is provided for research and informational purposes only. It is not medical, legal, financial, tax, or other professional advice, and using the Services does not create any professional, fiduciary, or advisor relationship between you and NeuralAI.

Similarity to other content. Because models generalize from training data, output may resemble text produced for other users or text that exists elsewhere. NeuralAI makes no representation that output is unique, original, or free of third-party rights, and we do not claim ownership of output generated for you (see Section 9). You are responsible for ensuring your use of output does not infringe anyone’s rights.

Human identity. NEA1 is an AI system. You may not represent its output as the words of a human being, and if you publish or distribute output in a context where readers could reasonably mistake it for human-authored statements, you should disclose its AI origin.

06Tool use and fetched content

During a chat, NEA1 may decide to call a tool — for example, fetching the text of a public web page you name, reading a file from a restricted demonstration sandbox, or attempting a search. Tool behavior is limited and evolving: fetched content may be truncated, page structure and media are not preserved, some sites cannot be read at all, and some tools (including search) may be unavailable, in which case the model is trained to say so rather than invent a result — but it may not always succeed.

When you ask NEA1 to fetch or otherwise process content, you represent that you have the right to direct that access. You agree not to use tool features to: reach private, internal, credential-protected, or paywalled systems or content in circumvention of access controls; collect personal data about third parties; probe or attack any system; or retrieve content for a purpose that would itself violate these Terms or applicable law. Fetched pages are third-party content: NeuralAI does not endorse, verify, or adopt them, and the model’s summary of a page is not a representation that the summary is complete or correct.

Tool calls are executed by our servers on the model’s behalf. You understand that a request you make in chat can therefore cause our infrastructure to contact the address the model selects, and that the model may occasionally select an address different from the one you intended (for example, by altering a URL path). Review tool indicators shown in the interface and the resulting answer before relying on them.

07Your content

Ownership. As between you and NeuralAI, you retain ownership of the messages, text, files, and other material you submit to the Services (“Your Content”). These Terms do not transfer ownership of Your Content to us.

License to operate the Services. You grant NeuralAI a worldwide, non-exclusive, royalty-free license to host, store, transmit, process, display, and create derivative works (such as conversation context, logs, and evaluation samples) from Your Content solely as needed to operate, secure, debug, evaluate, and improve the Services, as further described in the Privacy Policy. This license ends when Your Content is deleted from our systems, subject to the retention practices described in the Privacy Policy.

Your responsibility. You are responsible for Your Content and represent that you have all rights needed to submit it and grant the license above, and that Your Content does not violate the law or anyone’s rights. Do not submit content you are not willing to have processed by an experimental system: in particular, do not submit passwords, private keys, authentication codes, payment-card numbers, government identifiers, precise financial or medical records, or personal information about other people (especially minors). We may remove content, decline to process it, or suspend access where we believe content violates these Terms or the law.

Feedback. If you send us ideas, suggestions, bug reports, or other feedback about the Services or the model, we may use it for any purpose without restriction or compensation to you, and you assign us any rights needed to do so if the license above is insufficient.

08Acceptable use

You agree not to use the Services, and not to help anyone else use them, to:

Violate any applicable law or regulation, or infringe, misappropriate, or violate any intellectual-property, privacy, publicity, or other right of any person;
Generate, store, or distribute content that exploits or sexualizes minors, or facilitate any sexual content involving minors in any form — this is an absolute prohibition with no exceptions;
Plan, facilitate, or carry out violence, weapons development (including biological or chemical weapons), cyberattacks on critical infrastructure, or other imminent physical harm;
Harass, threaten, stalk, defame, or impersonate any person, or generate deceptive content intended to be passed off as genuine statements by a real person;
Commit fraud, run scams, or create disinformation intended to manipulate an election, a market, or a public process;
Send spam, or use output to generate bulk unsolicited communications in violation of anti-spam laws;
Overload, degrade, or interfere with the Services — including automated querying that exceeds reasonable demo use, denial-of-service behavior, or attempts to exhaust the inference backend;
Extract or attempt to extract model weights, training data, system prompts, credentials, keys, or other non-public components of the Services, except for artifacts we intentionally publish under an open license;
Use the Services to develop, train, or improve a competing model through systematic harvesting of outputs by automated means, beyond ordinary individual use and evaluation;
Misrepresent your affiliation with NeuralAI, or use NeuralAI’s or NEA1’s names or logos in a way that implies endorsement, sponsorship, or partnership that does not exist.

We may investigate suspected violations and cooperate with law enforcement where required or appropriate. Enforcement is at our discretion and may include warning, content removal, rate limiting, or suspension, with or without notice depending on severity.

09Intellectual property and open weights

The Services and site content. The website, its design, text, graphics, and software (other than openly licensed artifacts) are owned by NeuralAI or its licensors and are protected by intellectual-property laws. Except as expressly permitted, you may not copy, modify, distribute, or create derivative works from the site content, or frame or mirror the Services.

Open weights. NEA1’s weights are released under the Apache License, Version 2.0. That license — not these Terms — governs your use, reproduction, and distribution of the weights and other artifacts it covers, and if these Terms conflict with the Apache License as applied to those artifacts, the Apache License controls. The Apache License does not grant any trademark rights.

Names and marks. “NeuralAI,” “NEA1,” and the lab’s logos and visual identity are trademarks or trade dress of NeuralAI. You may refer to them truthfully (for example, to identify the model’s origin), but you may not use them in a way that suggests your product or content is made, endorsed, or certified by NeuralAI.

Output. As between you and NeuralAI, and to the extent permitted by law, you may use output generated in response to your prompts for lawful purposes, subject to these Terms and to the rights of any third parties. NeuralAI does not claim copyright ownership in such output. You acknowledge that output may not be protectable by intellectual-property law in some jurisdictions and that identical or similar output may be generated for others, to which you acquire no rights.

10Third-party services

The Services rely on and link to third-party platforms — including hosting and edge infrastructure (such as Wasmer and Cloudflare), inference compute providers (such as Modal), and distribution platforms for open artifacts (such as Hugging Face and GitHub). Those services are provided by their respective operators under their own terms and privacy policies, which we do not control. NeuralAI is not responsible for third-party services, content, availability, or practices, and a link or integration is not an endorsement. If a third-party platform suspends or changes its service, parts of our Services may change or become unavailable as a result.

11Availability and changes to the Services

The Services are research infrastructure offered free of charge, on an experimental basis. We may modify, suspend, replace, or discontinue any part of the Services at any time — including swapping the serving model checkpoint, changing decoding behavior, adding or removing tools, or retiring the public chat — without liability to you. The inference backend runs on capacity that scales to zero when idle, so first responses may be slow, and the Services may be unavailable during maintenance, provider outages, or capacity limits. Published benchmark figures, screenshots, and descriptions may lag the live system during transitions between model versions.

12Privacy

Our Privacy Policy explains what information we collect when you use the Services (including chat messages, inquiry-form submissions, and technical data), how we use and share it, how long we keep it, and the choices and rights available to you. By using the Services you acknowledge the practices described there. In short: we do not sell personal information, we do not run advertising trackers on these pages, and you should not send NEA1 sensitive personal information.

13Disclaimers

THE SERVICES AND ALL OUTPUT ARE PROVIDED “AS IS” AND “AS AVAILABLE,” WITHOUT WARRANTIES OF ANY KIND, WHETHER EXPRESS, IMPLIED, OR STATUTORY. TO THE MAXIMUM EXTENT PERMITTED BY LAW, NEURALAI DISCLAIMS ALL IMPLIED WARRANTIES, INCLUDING MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, TITLE, NON-INFRINGEMENT, ACCURACY, AND QUIET ENJOYMENT. WE DO NOT WARRANT THAT THE SERVICES WILL BE UNINTERRUPTED, SECURE, OR ERROR-FREE; THAT OUTPUT WILL BE ACCURATE, COMPLETE, OR SUITABLE FOR ANY PURPOSE; OR THAT DEFECTS WILL BE CORRECTED. SOME JURISDICTIONS DO NOT ALLOW THE EXCLUSION OF CERTAIN WARRANTIES, SO SOME OF THESE EXCLUSIONS MAY NOT APPLY TO YOU.

14Limitation of liability

TO THE MAXIMUM EXTENT PERMITTED BY LAW, NEURALAI AND ITS FOUNDER, PERSONNEL, AND SUPPLIERS WILL NOT BE LIABLE FOR ANY INDIRECT, INCIDENTAL, SPECIAL, CONSEQUENTIAL, EXEMPLARY, OR PUNITIVE DAMAGES, OR FOR ANY LOSS OF PROFITS, REVENUE, DATA, GOODWILL, OR OTHER INTANGIBLE LOSSES, ARISING OUT OF OR RELATING TO THE SERVICES OR OUTPUT — EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. OUR AGGREGATE LIABILITY FOR ALL CLAIMS ARISING OUT OF OR RELATING TO THE SERVICES WILL NOT EXCEED THE GREATER OF (A) THE AMOUNTS YOU PAID NEURALAI FOR THE SERVICES IN THE TWELVE MONTHS BEFORE THE CLAIM, OR (B) ONE HUNDRED U.S. DOLLARS ($100). THESE LIMITS APPLY REGARDLESS OF THE THEORY OF LIABILITY AND EVEN IF A REMEDY FAILS OF ITS ESSENTIAL PURPOSE. NOTHING IN THESE TERMS LIMITS LIABILITY THAT CANNOT BE LIMITED UNDER APPLICABLE LAW.

15Indemnification

To the extent permitted by law, you agree to indemnify, defend, and hold harmless NeuralAI and its founder, personnel, and suppliers from and against any claims, damages, losses, liabilities, and reasonable expenses (including attorneys’ fees) arising out of or relating to: (a) Your Content; (b) your use of the Services or of any output in violation of these Terms or applicable law; or (c) your violation of any right of a third party. We may participate in the defense of any claim at our own expense, and you agree to cooperate with our reasonable requests.

16Suspension and termination

You may stop using the Services at any time. We may suspend or terminate your access to all or part of the Services at any time, with or without notice, if we believe you have violated these Terms, created risk or legal exposure for us or others, or if continued operation becomes impractical for security, capacity, or cost reasons. Sections that by their nature should survive termination — including Sections 5–9, 12–15, 17, and 19 — will survive.

17Governing law and disputes

These Terms and any dispute arising out of or relating to them or the Services are governed by the laws of the State of Arkansas, without regard to its conflict-of-laws rules, except where applicable law requires otherwise.

Talk to us first. Before filing any claim, you agree to contact us through the website inquiry form, describe the issue in reasonable detail, and allow thirty (30) days for an informal resolution attempt. Most misunderstandings around a research demo can be resolved that way. If we cannot resolve the dispute informally, either party may bring an action exclusively in the state or federal courts located in Arkansas, and both parties consent to personal jurisdiction and venue there. Nothing in this section prevents either party from seeking injunctive or other equitable relief in any competent court, or from bringing an individual claim in small-claims court where eligible.

18Changes to these Terms

We may update these Terms from time to time as the Services evolve. When we do, we will post the revised Terms on this page and update the “Last updated” date above. For material changes, we will make reasonable efforts to draw attention to the update on the website. Revised Terms take effect when posted unless we state a later effective date. Your continued use of the Services after revised Terms take effect constitutes acceptance of them; if you do not agree to a revision, stop using the Services.

19General provisions

Entire agreement. These Terms, together with the Privacy Policy and any applicable open-source licenses, are the entire agreement between you and NeuralAI regarding the Services and supersede prior or contemporaneous understandings on the same subject.

Severability. If any provision of these Terms is held unenforceable, that provision will be modified to the minimum extent necessary to make it enforceable, or if that is not possible, severed, and the remaining provisions will continue in full effect.

No waiver. Our failure to enforce any provision is not a waiver of that provision or of our right to enforce it later.

Assignment. You may not assign or transfer these Terms or your rights under them without our prior written consent. We may assign these Terms in connection with a reorganization, acquisition, or transfer of the lab’s assets, with notice posted on the website.

Notices. We may provide notices by posting to the website or, where you have given us contact details through the inquiry form, by email to that address. Headings are for convenience only. “Including” means “including without limitation.”

20Contact

Questions about these Terms, requests under the dispute-resolution step in Section 17, and security reports can be sent through the inquiry form on the NeuralAI homepage — choose wording in your message that makes the subject clear (for example, “Terms question”). NeuralAI is based in West Memphis, Arkansas, United States.