FREQUENTLY ASKED QUESTIONS FOR
THE SEARCH ONLY TERMS CONTRACT
DISCLAIMER
This FAQ is provided for general informational purposes only and is not legal advice. It is intended to give an overview of the Search Only Terms Contract: for Access and Use of Website (the “Contract”), but it does not replace or modify the Contract itself.
The Contract is the legally binding document governing access to and use of Website Content. In the event of any inconsistency between this FAQ and the Contract, the Contract shall prevail.
Application and enforcement of the Contract will depend on the specific circumstances of each case, including the nature of the Accessing Party, the method of access, and the relevant jurisdiction. Enforcement options, including litigation, may not be appropriate in all situations and should be considered on a case-by-case basis.
Publishers should seek independent legal advice where necessary, particularly in relation to implementation, enforcement strategy, and cross-border issues.
Nothing in this FAQ limits or waives any rights or remedies available to the Website Operator under the Contract or applicable law.
For the full terms, see the Contract.
Table of Contents
This contract governs how your website content may be accessed and used.
The Contract also establishes a default Access Fee of £500 per discrete “Product” accessed as a result of an Access Event, which is automatically incurred each time a Product is accessed outside the unwaived permitted licensed scope. This fee is conditionally waived only where the access falls strictly within permitted Licensed Access (such as Non-Commercial Use or Search Indexing) and all contractual conditions are met. Any failure to meet those conditions results in the Access Fee becoming immediately due.
It applies to all forms of access, from ordinary human browsing, to search engine crawling and indexing, and sets clear limits on how that content may be reused.
In particular, it permits search indexing only, while restricting scraping, bulk extraction, dataset creation, and any AI-related use unless expressly authorised.
It operates alongside any general website terms (such as terms of use or cookie policies), but is focused specifically on content access, indexing, and downstream reuse, including automated and AI-driven activity.
This contract exists to draw a clear boundary between what is permitted and what requires permission.
It allows standard search engine indexing so the website can be discovered, while making clear that activities such as AI training, dataset creation, bulk scraping, or other datadriven reuse are not included in that permission.
Historically, these uses have often been treated as overlapping or implicit. This contract removes that ambiguity by expressly separating ordinary search indexing from AI and data-extraction activities and treating the latter as controlled uses that require explicit authorisation.
This contract applies to any person, entity, or system that accesses or interacts with the website or its content, regardless of the method or technology used.
It covers access by:
It applies equally whether access is:
The intention is that there are no gaps or grey areas. If a person, company, program, or system in any way accesses, retrieves, processes, or interacts with the website or its content, that access falls within the scope of this contract and is governed by its terms.
Think of it as working in layers — both legal and technical — at the same time:
The Contract is made visible on the website, for example through integration via a link into website terms of service (sometimes called Terms and Conditions, or Website Terms etc.) or as an additional legal footer link or an on-page notice.
Please see “How do I apply this Contract” question below for how to apply the Contract on your Website.
The Contract clearly states that anyone who accesses or uses the website is subject to its rules.
Acceptance does not require a tick box or signature. Instead, agreement is formed through conduct, simply by accessing, viewing, crawling, indexing, or otherwise interacting with the website or its content.
This ensures that:
Alongside the legal contract, the website uses technical, machine-readable links to the Contract to communicate permissions and restrictions to automated systems.
These may include:
These signals tell automated systems — such as crawlers, bots, and AI systems — the location of the Contract which defines what they are allowed to do and what they are not allowed to do when accessing the website. As many websites use the same location for the Contract it is machine-readable and can be readily identified across many websites.
The legal terms and the technical instructions are intentionally designed to say the same thing.
For example:
Because the legal rules and technical signals are aligned, there is no ambiguity about permitted use. This significantly strengthens enforcement, as a user or system has been put on notice in more than one way.
This Contract does not replace any existing website policies. Instead, it operates alongside them, each addressing a different aspect of how the website is run and used.
In the same way as:
this contract focuses specifically on how website content may be accessed, indexed, and reused, particularly by automated systems and AI-driven technologies.
How to implement the Search Only Terms Contract: for Access and Use of Website (the “Contract”) (Publisher instructions)
Please follow the steps below to correctly implement the Contract on your website.
1. Use the official MOW URL (no self-hosting required)
You must not create or host your own version of the Contract as the Contract is hosted and managed exclusively by MOW.
Each version of the Contract is assigned a unique, fixed URL (https://m4ow.uk/socw/2.txt), which is stewarded by MOW.
Once published the content at a given URL must never be changed. If any amendment is required, a new version will be created and assigned a new URL (e.g. /2.txt, /3.txt, etc.). Updates versions will be published on the MOW website (https://movementforanopenweb.com/).
2. Update robots.txt file
Next, add the following lines to the website’s robots.txt file (this file is usually located at: yourdomain.com/robots.txt):
# License https://m4ow.uk/socw/2.txt
User-agent: *
tdl: https://m4ow.uk/socw/2.txt
Allow: /
This step ensures all crawlers are explicitly informed that access is governed by the Contract.
3. Link in your Terms & Conditions
You should include a reference to the Search Only Terms Contract: for Access and Use of Website into your terms governing use of your website (for example, your website Terms and Conditions or Legal Notice).
To validly incorporate the Contract into your website terms and conditions, you should insert the following wording below into your website terms to formally incorporate the contract terms.
(A) Mandatory wording – adoption of the Contract
Insert the following text in the relevant place in your website terms and conditions
[Clause 1 – insert relevant clause number in line with your website terms and conditions] These [terms of service* use the name of the legal terms you offer on our website] refers and incorporates the following additional terms, which also apply to your use of our [site- use the same term you use on your terms for your website]:
[Clause 1.1] Our Search Only Terms Contract: for Access and Use of Website (available at the version-controlled URL published by MOW, currently being the following link: https://m4ow.uk/socw/2.txt) governs crawling, indexing and other automated or programmatic access to the Website and its content.
[Clause 1.2] If there is an inconsistency between any of the provisions of these [Terms and Conditions] and the provisions of [Search Only Terms Contract: for Access and Use of Website as linked above], the provisions of the Search Only Terms Contract: for Access and Use of Website, the version hosted at the referenced MOW URL shall prevail.
(B) Optional wording – override of cost of breach (Clause 14 only)
(Insert only if you wish to apply a different contractual sum for small claims for breach of contract per [Clause 14] Binding Search Only Contract for Access and Use of Website)*
Optional insert – Clause 14 (Cost of Breach) Override
Clause 14 of the Contract (relating to fixed sums, minimum amounts or deemed charges for unauthorised or unlicensed access) continues to expressly apply for the purposes of this [Website] except for only the sum to be claimed for breach of the terms by Unlicensed Access by an Unlicenced User which is to be replaced by the amount [£X being [amount in text] GBP] per instance of unauthorised or unlicensed access, as set out in the incorporated Binding Search Only Contract for Access and Use of Website.
For the avoidance of doubt, all other provisions of the Contract continue to apply without modification.
4. Record the change for evidence (important)
To create a record that this Contract was active at a specific point in time, you should archive your robots.txt file after updating it.
A simple way to do this is using the Internet Archive (Wayback Machine):
This provides timestamped proof that the instruction existed at that date.
Answer: MOW is a not-for-profit that needs to make content available for AI training, so AI responses are informed by alternative positions and arguments from those presented by AI vendors and Big Tech. As such the Contract is not compatible with the purpose of the MOW website.
Answer:
Short answer:
Yes. The Contract still applies.
Longer explanation:
When someone visits or uses the website after being clearly told there is a Contract governing access, the law generally treats that as acceptance. This approach is widely used across the web. You do not need a pop up or tick box for the rules to apply. What matters is that the rules are clearly stated and the user chooses to go ahead anyway.
LEGAL NOTE – why wording matters
The phrases “by accessing or using” and “governing access and use” are critical.
These link the Contract to conduct, not express consent.
Avoid softening this to “we ask users to comply” or “this sets expectations” — that risks undermining the contractual footing.
Isn’t this just trying to contract around copyright limits (e.g. fair dealing or fair use)?
Answer: The Contract doesn’t replace copyright law. Instead, it sets conditions for using the website that sit alongside the regulatory copyright system. Just as venues can impose rules even where the law allows entry, a website can limit how its content is used as a condition of access.
LEGAL NOTE
Keep language that says the Contract “sits alongside statutory rights”.
Do not say “overrides” or “takes precedence over” copyright law — that invites challenge.
The word “contractual” is important here.
Answer: No. The Contract clearly distinguishes between permitted “Index Access” for search functionality. Search engines are allowed to look at pages so they can point visitors to the original site. What they are not allowed to do is reuse the content for AI training, summaries, or datasets. Permission is given for one narrow purpose only.
Why this may come up: The distinction between search indexing and AI use is increasingly blurred (particularly with large platforms), so this is a likely point of confusion.
LEGAL NOTE
The distinction between “indexing” and “AI training / datasets” is foundational.
The phrase “purposelimited” is important — don’t replace it with vague wording like “generally allowed”.
Answer: It builds on them for search. Free indexing is preserved as a baseline. The Contract just clarifies that AI and reuse are separate activities requiring permission. The Contract preserves standard search indexing as a baseline, while making clear that other forms of reuse—particularly AI-related uses—are separate and require express permission which the Website Operator can negotiate with the AI company separately to the Contract.
Answer:
This Contract may be incorporated into other contractual or technical frameworks in several ways, depending on the system used by the Website Operator. These may include:
Where incorporation occurs, it is intended that this Contract applies specifically to licensed access for indexing and permitted crawling activities, as defined herein.
LEGAL NOTE:
This preserves flexibility. The key is that incorporation is multi-channel (technical + contractual + notice-based) and not dependent on a single mechanism.
Answer:
Where this Contract is incorporated into another agreement or set of terms, it is intended to govern the specific subject matter of Website Content access, indexing, crawling, and reuse for AI or data-related purposes.
In the event of any conflict:
LEGAL NOTE:
This avoids overreach. The Website Operator is not claiming universal supremacy over all contracts, only subject-matter primacy for indexing/AI/data use. That is much more defensible.
Answer:
The Website may operate different layers of terms depending on the type of user or access method. For example:
LEGAL NOTE:
This addresses a common concern that applying different terms could create inconsistency or confusion. In reality, the Contract applies a single legal framework, but distinguishes between different types of access and use (e.g. Non-Commercial Use, Search Indexing, and prohibited AI/data uses).
The distinction is use-based, meaning the same rules apply but different conditions are triggered depending on how the Website Content is accessed. This avoids overreach, removes ambiguity, and strengthens enforceability by ensuring the Contract is clear.
Answer:
This Contract does not replace or override copyright law. Instead, it sets out additional contractual conditions for access and use of Website Content.
This means:
LEGAL NOTE:
This preserves the critical distinction: contractual licence + statutory copyright coexistence. Do not blur them.
Answer: Content can carry hidden identifiers or patterns that show up later inside AI outputs or datasets. If those indicators appear, and there’s no permission on record, Unlicensed Access has likely happened.
Some Accessing Parties might include links to the original source of results they display. Such links would prove Unlicensed Access has happened.
Further some Accessing Parties may identify themselves using IP addresses and User-Agent strings which can be recorded in log files to prove that access took place.
The Contract also allows the website owner to ask the organisation involved to explain how the content was collected and used.
Why this may come up: Enforcement against AI systems is often seen as evidentially difficult, so this is likely to be a key practical concern.
LEGAL NOTE
References to watermarks, provenance signals, data labels, and disclosure obligations are doing heavy lifting.
The right to demand explanations and records is essential — do not remove or water it down.
Avoid framing this as “we might ask nicely” — it is a condition of access.
Answer: Only if it’s misconfigured, which is why alignment matters, if used without nuance it could. The Contract and the technical signals need to say the same thing. The intention is not to block normal search results but to clearly block AI training and bulk data use no matter how it occurred. When those are matched properly, search visibility is preserved.
Google do not currently list terms and conditions as a factor in search engine results listing. If they were to modify their search index list to penalise websites that incorporate this Contract then that could possibly be an abusive act under antitrust laws.
LEGAL NOTE
Keep language stating that technical controls operate alongside legal terms.
This avoids arguments that robots.txt is the only relevant signal.
Answer: Evidence can be built by demonstrating patterns, not perfection through keyword tracking, systematic sampling of AI outputs, and screen captures showing replication of content. Courts and regulators don’t expect every instance to be captured. What matters is showing that misuse is happening repeatedly or systematically. For example, through repeated AI outputs that closely resemble the Website Content.
LEGAL NOTE
The idea of “pattern evidence” is important.
Avoid language suggesting every instance must be proven — that raises the bar unnecessarily.
Answer: The Contract applies to the content wherever it ends up. If content is passed on without permission and then used for AI, that downstream use can still be challenged. This also highlights the need for early enforcement to discourage liability avoidance structures.
Any AI operator found to be using unlicensed sources is likely to suffer brand embarrassment as well as significant financial costs. This acts as a commercial incentive to cease such practices.
LEGAL NOTE
Language about “downstream use” and “derivative datasets” is legally significant.
Avoid implying only the first scraper is responsible.
Answer: Monitoring is expected to combine mix of internal checks and specialist tools. Publishers don’t need complex infrastructure. Simple monitoring, combined with third party services where needed, is usually sufficient. External services, for which there is a growing market for third-party tools that specialise in detecting and evidencing AI-related misuse can be used.
LEGAL NOTE
Avoid language that implies surveillance obligations.
Emphasise proportionality.
Answer: Yes, especially together. Tokenisation and provenance systems could allow content to be traced across reuse contexts. No single system is perfect, but combined technical markers make it much easier to detect content reuse and support enforcement where problems arise.
LEGAL NOTE
Keep wording that says these tools support evidence, not replace legal rights.
Answer: The Data Labels standards reduce ambiguity for machines. They provide a foundation for machine-readable permissions and traceability via immutable URLs. Clear, machine readable links tell automated systems what contract applies to a request. By adopting a small number of common contracts – such as this Contract - these become machine readable in practice. This makes compliance easier and excuses by Accessing Parties for non-compliance impossible to justify. Standardisation and widespread adoption is key to ensuring that these signals are recognised and respected across platforms.
LEGAL NOTE
Keep “machine-readable legal restrictions” language — it connects technical signals to legal intent.
Answer: No. The Contract does not restrict accessibility or assistive technologies. Tools such as screen readers, texttospeech systems, and other accessibility agents used to help people with disabilities access the website are permitted. The Contract targets bulk reuse and secondary exploitation of content, not user-facing non-commercial access (section 4.1.1).
Why this may come up:
Accessibility advocates and regulators are rightly concerned about anything that
could chill or impair access for people with disabilities. This question helps
make clear that the Contract targets misuse, not accessibility.
LEGAL NOTE
Keep explicit language that accessibility and assistive technologies are
permitted. Avoid broad bans on “automated access” without carveouts that risks
unintended consequences. Framing accessibility tools as user-facing access
mechanisms, not secondary reuse systems, is important.
Can websites covered by this Contract be used for academic and research purposes?
Answer: Limited non-commercial access (including reading and research) is permitted. However, activities such as scraping, dataset creation, or AI training — even for research — require separate permission.
LEGAL NOTE
Avoid blanket exemptions for “research” or “academic use” as these can undermine
enforceability. Keep the requirement for a separate agreement clear and
explicit. Do not suggest that all academic use is prohibited; the issue is
permission, not category of user.
Answer:
Technical signals such as robots.txt, HTTP headers, and other machine-readable directives are used to communicate access permissions to automated systems.
These signals operate alongside this Contract and are intended to ensure that:
Where these signals and this Contract are used together, they should be interpreted consistently as part of a single access control framework.
LEGAL NOTE:
This avoids the argument that robots.txt is “only technical” or “only advisory”. You frame it as part of a unified legal-technical system, not separate layers in conflict.
Answer:
No. There is no requirement to modify your website’s HTML in order to implement the Contract.
The primary method of implementation remains through machine-readable signals such as the robots.txt file, which identifies the applicable Terms Document Locator (TDL).
However, publishers may choose to additionally include references to the Contract within the HTML of their webpages. For example, where multiple licences are used at the following URLs:
https://m4ow.uk/socw/2.txt
https://example.com/other/tos.txt
These can also be included in the HTML <head> element via the terms-of-service link relation, as shown below:
<head>
<link rel="terms-of-service" href="https://m4ow.uk/socw/2.txt"/>
<link rel="terms-of-service" href="https://example.com/other/tos.txt"/>
</head>
This approach is optional and does not replace the need to correctly implement the robots.txt configuration, but may provide an additional layer of visibility and consistency across systems.
LEGAL NOTE
Keep the distinction clear: HTML inclusion is supplementary, not required.
Avoid implying that HTML implementation alone is sufficient — robots.txt and machine-readable signals remain primary.
Answer:
Only if Common Crawl has entered into a binding agreement ensuring that Website Content is not used or shared for any restricted purposes under this Contract. In practice, inclusion in Common Crawl would typically constitute Dataset creation and downstream distribution, which is prohibited unless expressly licensed.
Answer:
No, unless the Internet Archive (or equivalent service) has entered into an express written agreement with the Website Operator permitting such activity. In the absence of such permission, this is not a permitted use. This is because such activity involves the systematic copying, storage, and potential redistribution of Website Content, which falls outside the limited licence granted under Clause 4 (Licence to Access Website) and is prohibited under Section 7 (Prohibition on AI Scraping and Data Mining) and Section 8 (Database Rights Protection), and is further restricted by the machine-readable controls described in Section 10.
Answer:
There is no standard price or marketplace for valuing content. This contract does not seek to apply a market value, the Contract simply makes clear that use beyond search indexing is not free by default. The Contract also establishes an Access Fee of £500 per Access Event as defined as any instance of Access to a Product on the Website (e.g. an article, video image or photographic work, etc.), which applies where access falls outside the permitted licensed scope.
LEGAL NOTE
Preserve the statement that use requires permission first.
Do not suggest payment is automatic or mandatory in all cases.
Answer:
Not necessarily. The baseline model allows free use for search indexing, while reserving other uses (such as AI training) for permission or licensing. Search engines can continue to index content without payment. Payment only applies where companies want to use the content for other purposes, such as AI training.
LEGAL NOTE
Keep the baseline vs additional use framing.
Avoid wording that suggests a “paywall for search”.
Answer:
Even if a company is overseas, courts can order intermediaries — like internet providers or platforms — to stop access or block misuse. These tools are specifically designed for situations where bad actors are hard to reach directly.
Why this may come up: This reflects a practical enforcement concern, audiences will want to understand whether the framework is effective against overseas/anonymous actors.
LEGAL NOTE
Terms like “website blocking orders”, “intermediaries”, and “injunctive relief” should stay in supporting material even if not front and centre.
Avoid saying “we can always enforce anywhere” — stick to credible, recognised mechanisms.
Answer:
While technically possible, this is usually inefficient. Straightforward claims may begin in lower courts for speed and cost efficiency, but complex or high-value disputes may move to higher courts. While small claims are available, bringing dozens or hundreds of them is inefficient. A better approach is to group evidence and use a handful of cases to show widespread misuse. A mixed strategy is expected, with at least one strong case used to establish precedent. The Contract allows aggregation of multiple Unlicensed Access Events into a single claim, and the Website Operator may choose the most efficient strategy, including issuing aggregated claims or multiple proceedings where appropriate.
LEGAL NOTE
Keep references to aggregation and representative cases.
This preserves flexibility without committing to one enforcement path.
Answer:
It depends on the scale and seriousness. Smaller disputes can be handled quickly and cheaply via the UK Small Claims courts and Money Claim Online. Bigger or more complex cases may need higher courts. Often, one strong case sets the tone and resolves many others without further litigation.
LEGAL NOTE
Avoid locking yourself into a single forum in publicfacing wording.
Retain discretion language such as “may” rather than “will”.
How do I make a claim?
The Contract provides a structured framework for bringing claims as a debt after issuing the invoice, including the ability to recover sums due through England and Wales small claims track via Money Claim Online. We have prepared a “Guide to Small Claims via Money Claims Online”, which will guide you through the practical steps such as issuing invoices, sending a Letter Before Claim, and using the streamlined procedure envisioned by using the small claims track where appropriate. The approach is designed to enable efficient and proportionate enforcement, including by publishers without extensive litigation resources. (For more detail, please see the Small Claims Guidance.)
Answer:
Yes. Establishing a clear early enforcement position helps remove uncertainty. Once expectations are clear, many other actors adjust their behaviour without needing further action.
LEGAL NOTE
Retain emphasis on deterrence through clarity, not punishment.
Answer:
Often yes. Acting together reduces costs, avoids duplication, and increases leverage especially against large platforms, and increases pressure on non-compliant actors.
LEGAL NOTE
Avoid wording that implies compulsory collective action.
Answer:
Regulatory routes complement litigation. Regulators are useful for systemic issues, while courts remain the fastest way to stop specific misuse. However, litigation remains an important and immediate enforcement mechanism.
LEGAL NOTE
Retain litigation as an immediate option.
Answer:
There is some risk, but large-scale withdrawal of indexing is considered unlikely due to commercial, regulatory, and competitive pressures. Search engines need high quality content. Widespread adoption of this Contract, or similar policies, reduces the incentive to retaliate.
Answer:
Primarily to coordinate policy documents, not control. Its role would be to support standards, tools, and consistency, not to take content ownership or run lawsuits. The model is intended to remain decentralised, with publishers retaining full control. Publishers may wish to band together via trade bodies for efficiency of licensing beyond the permitted uses of this Contract or for litigation.
LEGAL NOTE
Maintain decentralisation language to avoid competition or control concerns.
Answer:
Full standardisation is unlikely in the short term. Different markets will move at different speeds, but shared principles are likely to emerge.