Acceptable Use Policy

This Acceptable Use Policy (the "Policy") sets out the activities that are prohibited on AhuraSense Cloud. It applies to every service we operate, including virtual machines and bare-metal servers, GPU and accelerator instances, block and object storage, networking and egress, managed databases, model hosting, and public or private inference endpoints. It also applies to our websites, consoles, APIs, and support channels.

The Policy is incorporated by reference into our Terms of Service and into any order form, enterprise agreement, or reseller agreement under which you obtain the services. Where a negotiated agreement addresses a matter covered here, the more specific and more restrictive requirement applies. Defined terms not explained in this Policy carry the meaning given to them in the Terms of Service.

We operate shared, multi-tenant infrastructure across facilities in India and serve customers contracting with our UK entity. Conduct that would be tolerable on dedicated, isolated hardware can still be prohibited here because of its effect on other tenants, on our IP reputation, on upstream transit providers, or on our obligations to data-centre operators and regulators. This Policy is intentionally broad; it describes categories of prohibited conduct rather than an exhaustive list, and we may treat conduct that is materially similar to a listed prohibition as a breach.

You are responsible for all activity that occurs under your account, whether carried out by you, your employees, your contractors, your end users, or anyone using credentials issued to you. This includes activity performed by automated agents, scheduled jobs, and third-party software you deploy, and it applies regardless of whether you were aware of the activity at the time.

If you resell capacity, host workloads for customers of your own, or expose an inference endpoint to the public, you must impose terms on your users that are at least as protective as this Policy, and you must have a practical means of investigating and stopping their abuse. "My user did it" is not a defence to a breach of this Policy, though your cooperation and speed of response are relevant to how we respond.

• Protect account credentials, API keys, SSH keys, and service tokens; rotate them on a defined schedule and revoke them immediately when staff or contractors leave.
• Maintain a monitored abuse or security contact for your organisation and keep it current in the console, so we can reach a human quickly.
• Secure what you deploy: patch operating systems and container images, avoid default or shared passwords, and do not expose administrative interfaces, model weight stores, or unauthenticated inference endpoints to the open internet.
• Have a lawful basis and any necessary rights, licences, or consents for the data, content, and model weights you upload, process, fine-tune on, or serve.

You must not use AhuraSense Cloud to conduct, facilitate, promote, enable, or conceal conduct that violates Applicable Law or this Policy.

A workload may be subject to more than one legal regime depending on the Customer, the infrastructure location, the individuals affected, and the nature of the activity. You are responsible for identifying and complying with laws applicable to your use of the Services.

Where legal obligations conflict or the legality of a proposed workload is uncertain, you should obtain appropriate legal advice before deploying it. Nothing in this Policy authorises activity merely because it may be lawful in one jurisdiction if it is prohibited by law applicable to the relevant Customer, workload, or infrastructure.

AhuraSense may restrict or decline Services where providing them would expose AhuraSense, its Affiliates, infrastructure providers, or customers to material legal or regulatory risk.

• Trafficking in controlled substances, weapons, endangered species, stolen goods, or human beings, and the operation of marketplaces or logistics for the same.
• Money laundering, terrorist financing, sanctions evasion, or the operation of unlicensed payment, remittance, or virtual-asset services.
• Gambling, lotteries, pharmaceutical sales, financial advice, or other regulated activity conducted without the licences your jurisdiction requires.
• Distribution of material that is unlawful to possess or transmit, or the use of our storage and egress capacity as a staging point for such material.

We may be required to restrict access from, or workloads associated with, embargoed territories and sanctioned parties. You must not use the services in breach of export control or sanctions law, including by re-exporting access to GPU capacity, model weights, or inference results to restricted parties.

You must not use the services to defraud, mislead, or extract value from anyone under false pretences. This covers advance-fee and investment fraud, fake storefronts and counterfeit goods, romance and confidence scams, invoice and business email compromise, ticket and refund fraud, and any scheme that depends on the target misunderstanding who they are dealing with.

It also covers fraud against us. You must not open multiple accounts to obtain repeated free trials or promotional credits, provide false identity or billing information, use stolen or unauthorised payment instruments, or use anonymising infrastructure to evade a prior suspension. Trial and credit-funded capacity exists so that customers can evaluate the platform, and we monitor for signup patterns that indicate systematic abuse of it.

Deceptive automation is treated as fraud where it is used at scale: mass creation of fake accounts, reviews, engagement, or ad impressions; synthetic traffic intended to distort analytics or ad spend; and credential-stuffing or scalping operations against third-party platforms.

You must not store, host, transmit, or serve material that infringes copyright, trade marks, patents, database rights, trade secrets, or rights of publicity. This includes pirated media libraries, cracked or unlicensed software, keygens and licence circumvention tooling, counterfeit branding, and mirrors or proxies whose purpose is to distribute infringing material.

For AI workloads specifically, you are responsible for having the rights to the training data, fine-tuning datasets, and model weights you bring to the platform, and for complying with the licence terms attached to any open-weight or commercially-licensed model you deploy. Deploying a model whose licence prohibits your intended use, or redistributing weights you are not licensed to redistribute, is a breach of this Policy as well as of the underlying licence.

We respond to properly substantiated infringement notices. Send them to [email protected] with enough detail to identify the work, the material complained of, and the location of that material on our infrastructure. We may remove or disable access to the material, forward the notice to the customer, and terminate accounts of repeat infringers.

You must not use the services to store, generate, or distribute content that incites or promotes violence, terrorism, or violent extremism; that constitutes hate speech targeting people on the basis of protected characteristics; that glorifies or provides instruction for self-harm, suicide, or eating disorders; or that provides operational instructions for weapons capable of mass casualties, including chemical, biological, radiological, nuclear, or explosive weapons.

We recognise that legitimate work touches difficult subject matter. Security research, trust-and-safety tooling, content moderation, academic study, journalism, and medical or public-health work frequently require handling material that would be prohibited if published or acted upon. Such work is permitted where it has a genuine protective or scholarly purpose, is appropriately access-controlled, and is not used to distribute the harmful material itself. If your workload sits near this line, tell us in advance rather than after an abuse report.

Content whose distribution is lawful may still be restricted on our network where it causes disproportionate operational harm — for example, material that reliably triggers upstream blocklisting, transit provider complaints, or law-enforcement action affecting shared address space.

AhuraSense Cloud has zero tolerance for child sexual abuse material (CSAM) and for any content or conduct that sexualises minors. This prohibition is absolute. It covers real imagery, recordings, and text; it covers computer-generated, drawn, and AI-synthesised depictions; and it covers models, LoRAs, embeddings, datasets, prompts, and pipelines created or tuned to produce such material. It also covers grooming, sextortion, the solicitation or trafficking of minors, and the operation of services whose purpose is to index or distribute this material.

Where we identify CSAM or related conduct on our infrastructure, we will terminate the account immediately and without prior notice. We will preserve relevant data and logs, and we will report the matter to law enforcement and to the appropriate child protection body — including the National Center for Missing & Exploited Children (NCMEC) or the equivalent reporting authority in the relevant jurisdiction, such as the Internet Watch Foundation in the United Kingdom and the National Cyber Crime Reporting Portal and law enforcement in India.

No grace period, remediation window, appeal, or refund applies to termination under this section. Nothing in the graduated enforcement described later in this Policy limits our ability to act instantly here. Attempts to evade detection — including encryption, obfuscation, or distribution across multiple accounts — are treated as aggravating conduct and may extend termination to affiliated accounts and organisations.

You must not use the services to harass, stalk, threaten, intimidate, or degrade any individual. Prohibited conduct includes coordinated brigading campaigns, doxxing or the publication of private information such as home addresses, identity numbers, or medical records, and the operation of platforms whose primary function is to enable targeted abuse.

• Non-consensual intimate imagery, including synthetically generated or altered imagery of identifiable people, and services that solicit or distribute it.
• Stalkerware, covert location tracking, and surveillance tooling deployed against individuals without their informed consent or a lawful basis.
• Scraping, aggregating, or enriching personal data to build profiles used for harassment, blackmail, or intimidation.
• Sexual services involving coercion or trafficking, and any content produced without the documented consent of the people depicted.

Where an individual reports that our infrastructure is being used to target them, we will act on credible reports quickly and may restrict the offending workload while we investigate, prioritising the safety of the person at risk.

You must not build, host, distribute, or operate malicious software or the infrastructure that supports it. This includes viruses, worms, trojans, rootkits, ransomware, wipers, keyloggers, info-stealers, and loaders, as well as command-and- control servers, exploit kits, malvertising redirect chains, and drop sites for exfiltrated data.

Operating or renting botnet capacity is prohibited outright, as is participating in a botnet with instances under your control. If one of your instances is compromised and conscripted, we treat that as a security incident rather than as intentional abuse — but you are still required to remediate promptly, and we may isolate the instance at the network level in the meantime to protect other tenants and our address space.

Legitimate security work is permitted with appropriate controls. Malware analysis, reverse engineering, detonation sandboxes, and red-team tooling development are acceptable where samples are contained, egress is restricted so that samples cannot reach third parties, and the environment is isolated from other tenants. Tell us in advance if you intend to run this kind of workload so that our abuse team can distinguish it from a live compromise.

You must not host phishing pages, credential harvesting forms, fake login portals, or lookalike sites impersonating banks, government bodies, cloud providers, exchanges, or any other organisation. This extends to the supporting infrastructure: typosquatted domains resolving to our IP space, redirectors and link shorteners fronting phishing destinations, phishing kits stored in object storage, and mail infrastructure used to deliver the lure.

Also prohibited are credential stuffing and password spraying against third-party services, the operation of adversary-in-the-middle proxies designed to capture sessions or bypass multi-factor authentication, the trafficking of stolen credential dumps, and the use of GPU capacity to crack password hashes you are not authorised to test.

Phishing content is treated as severe abuse because of the immediate harm to victims and to the reputation of the addresses we share across tenants. It is one of the categories where we routinely act without prior notice, typically by disabling access to the specific content or instance first and contacting you immediately afterwards.

You must not access, or attempt to access, any system, network, account, or data without authorisation. This applies to third-party systems reached from our network, to other tenants on our infrastructure, and to AhuraSense systems themselves — including the hypervisor, the management network, the metadata service, the billing and console back-ends, and any accelerator or storage device beyond the boundary of the resources allocated to you.

• Attempting to escape a virtual machine or container, cross a tenancy boundary, or read memory, VRAM, or storage belonging to another customer.
• Probing or exploiting the hypervisor, GPU firmware, device passthrough paths, or shared storage fabric.
• Circumventing quotas, rate limits, metering, licensing checks, or authentication controls in our services.
• Using another customer's API keys, tokens, or session material, however obtained.

If you discover a vulnerability in an AhuraSense-operated system, report it promptly to [email protected] in accordance with our Vulnerability Disclosure Policy. Do not continue testing beyond what is reasonably necessary to establish the existence and impact of the vulnerability, and do not access, copy, modify, or retain another Customer's data.

Legal process, regulatory enquiries, and formal notices should instead be sent to [email protected].

You must not originate or participate in denial-of-service or distributed denial-of-service attacks from our network, and you must not use our capacity to coordinate, control, or amplify such attacks elsewhere. This includes volumetric floods, application-layer attacks, and slow-resource-exhaustion techniques.

You must not operate open resolvers, open relays, open proxies, or misconfigured services usable for reflection and amplification, including DNS, NTP, memcached, SSDP, CLDAP, and similar protocols. You must not spoof source addresses, forge packet headers, or otherwise disguise the origin of traffic leaving our network. You must not hijack, announce, or attempt to route address space you do not control.

Load testing and stress testing against your own systems are permitted, but must be arranged with us in advance where they will generate substantial egress or sustained packet rates, so that we can distinguish them from an attack and avoid automated mitigation being applied to your traffic. Testing against systems you do not own requires documented authorisation from the owner and remains subject to the section on scanning and exploitation.

Unauthorised port scanning, vulnerability scanning, service enumeration, brute-force authentication attempts, and exploitation of third-party systems from our network are prohibited. Internet-wide scanning campaigns are prohibited without our prior written approval, because they generate abuse complaints and blocklist entries that affect every tenant sharing the originating address space.

Authorised penetration testing is permitted where you can produce written authorisation from the owner of the target system, the scope is bounded, and the testing does not degrade shared infrastructure. We may ask to see that authorisation when we receive a complaint. Bug bounty participation is acceptable within the scope published by the programme owner.

Web scraping and crawling must respect the technical and legal boundaries set by the target: honour robots directives and rate limits, do not circumvent authentication or anti-bot controls, do not use residential proxy networks to disguise your origin, and do not collect personal data without a lawful basis. Aggressive crawling that functions as a denial-of-service attack will be treated as one.

You must not send unsolicited bulk email, SMS, chat, push, or voice messages from our infrastructure, and you must not use our capacity to build, host, or operate the tooling behind such campaigns. All commercial messaging you send must be based on verifiable opt-in consent, must accurately identify the sender, and must offer a functioning and promptly honoured unsubscribe mechanism.

• Purchased, scraped, harvested, or otherwise non-consensual recipient lists, and list-washing or list-validation services run against them.
• Forged headers, misleading subject lines, deceptive sender identities, and snowshoe patterns that spread volume across many addresses to evade filtering.
• Mass posting, comment spam, forum and wiki spam, and automated engagement against social platforms.
• Bulk outbound messaging generated by language models, which is subject to this section in exactly the same way as messaging written by a person.

Outbound SMTP may be restricted by default on new accounts and can be enabled after review. Because deliverability depends on the reputation of shared address space, we may cap send rates, require dedicated addresses, or require the use of a reputable third-party sending provider for high-volume workloads.

Our shared and burstable products are sized on the expectation of variable, bursty usage. You must not consume compute, GPU, memory, storage IOPS, or network capacity in a way that degrades service for other tenants, and you must not run sustained full-utilisation workloads on burstable instance types where a dedicated or reserved product is the appropriate fit. If you need sustained throughput, buy it — we will help you size it.

• Circumventing quotas, rate limits, or metering, including by fragmenting a workload across many accounts to stay under per-account thresholds.
• Using trial, free-tier, or promotional GPU credits for production workloads, sustained training runs, or resale of capacity.
• Operating open proxies, exit relays, or free VPN services that anonymise third-party traffic through our address space without prior written approval.
• Using storage or egress as a free content distribution network detached from any compute workload on the platform.

Where usage is abusive rather than merely heavy, we may throttle, cap egress, resize or reschedule the workload, or require migration to an appropriate product. We will normally contact you first and give you the option to move to suitable capacity.

Cryptocurrency mining, minting, and proof-of-work or proof-of-space validation are prohibited on trial accounts, free-tier resources, and any capacity funded by promotional or granted credits, without exception. They are also prohibited on shared, burstable, or spot-style compute and GPU resources unless we have given you prior written approval.

Mining is permitted only on dedicated or reserved capacity that we have approved in advance in writing for that purpose. If you want to mine on AhuraSense Cloud, contact us before you deploy: we will discuss the appropriate dedicated or reserved product, power and thermal constraints, and commercial terms. Approval is specific to the account, the capacity, and the workload described, and it is not transferable.

Unapproved mining is treated as resource abuse and, where it is funded by credits or free capacity, as fraud. It is one of the most common misuses of stolen credentials and stolen payment methods, so where we detect it on an unapproved footprint we may suspend the affected instances immediately, void the associated credits, and invoice for the capacity consumed at on-demand rates.

Blockchain work that is not proof-of-work mining — running validators or full nodes, indexing chain data, or developing and testing smart contracts — is generally acceptable on appropriately sized paid capacity, subject to the resource abuse section above.

The prohibitions in this Policy apply in full to model training, fine-tuning, evaluation, and inference on our GPU instances and hosted endpoints. Using a model as an intermediary does not change whether conduct is permitted: if you could not do it yourself on our infrastructure, you may not build or operate a model that does it for you.

• Training, fine-tuning, or serving models whose purpose is to generate prohibited content, including CSAM, non-consensual intimate imagery, malware, or weapons instructions.
• Building or operating models designed to defeat safety systems, content filters, CAPTCHAs, age verification, or fraud controls.
• Deliberately removing, bypassing, or degrading the safety mitigations of a model you deploy, where doing so is intended to enable prohibited outputs.
• Using model weights, datasets, or inference outputs in breach of the licence, contract, or platform terms under which you obtained them.

Safety research, red-teaming, and evaluation work that necessarily probes these boundaries is permitted where it is conducted in a contained environment, with restricted egress, for a genuine protective purpose, and without publishing the resulting harmful capability or content. Tell us in advance if you intend to run this kind of work.

You must not use our inference endpoints or hosted models to generate content that would be prohibited if you uploaded it directly. This includes sexual content involving minors, non-consensual sexual imagery of real people, targeted harassment and hate content, self-harm encouragement, and operational uplift for weapons capable of mass casualties or for serious cyberattacks.

If you operate an endpoint that is reachable by people other than your own staff, you are responsible for the outputs it produces. We expect you to apply input and output filtering proportionate to the risk of your application, to retain sufficient logs to investigate abuse reports, to rate limit anonymous access, and to give your users a way to report harmful outputs to you.

You must not present model output as verified fact where doing so would foreseeably cause harm, and you must not use generated content to fabricate evidence, records, credentials, or official communications.

You must not use our services to generate synthetic media that depicts a real, identifiable person without their consent in a way that is deceptive or harmful. This includes face and voice cloning used for fraud, synthetic imagery or audio of politicians, officials, or journalists presented as authentic, fabricated statements attributed to real people, and synthetic identity documents or biometric samples intended to defeat verification systems.

Impersonating an organisation is equally prohibited: generating communications, brand assets, or support channels that purport to come from a company, public body, or regulator you do not represent. Using generated audio or video to bypass voice authentication, liveness checks, or know-your-customer processes is treated as fraud under this Policy.

Consensual and clearly labelled synthetic media is permitted — including entertainment, satire that a reasonable viewer would recognise as such, accessibility tooling, dubbing and localisation, and corporate avatars used with the subject's permission. Where you generate synthetic likenesses at scale, we expect you to obtain and retain evidence of consent, to disclose the synthetic nature of the output to viewers, and to apply provenance signalling such as watermarking or content credentials where it is practical to do so.

Scale is an aggravating factor. Conduct that might be a minor nuisance performed once by a person becomes serious abuse when a model or agent performs it thousands of times per hour, and our enforcement reflects that. Automated systems you deploy must operate within the same boundaries as your staff.

• Bulk generation of spam, review manipulation, astroturfing, or synthetic engagement across social, commerce, or review platforms.
• Agentic systems that create accounts, complete transactions, or interact with third-party services in breach of those services' terms, or that circumvent bot detection and rate limits.
• Automated scraping pipelines that ignore access controls or that assemble personal data at scale without a lawful basis.
• Autonomous agents with unbounded outbound network access and no human oversight, where a malfunction or prompt injection could cause harm to third parties.

If you run agents that act on the open internet, you must be able to identify the agent, attribute its actions, throttle it, and stop it quickly. We may require you to demonstrate those controls where an agentic workload generates abuse complaints.

Some applications are permitted but carry heightened obligations because a wrong output materially affects someone's rights, safety, livelihood, or liberty. If you deploy models on our infrastructure for the uses below, you must maintain meaningful human oversight, document the limitations of your system, test for accuracy and bias across affected groups, and provide affected people with a route to challenge or appeal an outcome.

• Medical and health: diagnosis, triage, treatment recommendation, mental health support, or drug interaction advice — requiring qualified clinical review and appropriate regulatory clearance before clinical use.
• Legal: advice, document generation, or case assessment presented to a person as a substitute for a qualified practitioner.
• Financial: credit scoring, lending, insurance underwriting, and investment advice, where explainability and anti-discrimination obligations apply.
• Employment and education: automated screening, ranking, monitoring, discipline, or admissions decisions affecting candidates and students.
• Biometric identification: facial recognition, gait or voice identification, and emotion inference — which must have a lawful basis and must not be used for indiscriminate mass surveillance or for inferring protected characteristics.
• Critical infrastructure: control or safety functions in energy, water, transport, telecommunications, or industrial systems, which require fail-safe design and human-in-the-loop control.
• Law enforcement and justice: predictive policing, risk assessment, evidence evaluation, and immigration or border decisions, which must not be operated autonomously against individuals.

Fully automated decision-making in these domains, with no human able to review or reverse the outcome, is prohibited. You remain responsible for meeting the sectoral and data protection requirements that apply to you; we can support your compliance programme with documentation about the infrastructure we operate, but we do not certify your application, and nothing in this Policy should be read as a claim that your use case is compliant.

We investigate suspected breaches of this Policy on the basis of abuse reports, complaints from third parties and upstream providers, blocklist and IP reputation signals, network telemetry such as traffic patterns and flow data, and billing or signup signals that indicate fraud. We do not routinely inspect the content of customer workloads, and we access customer data only where it is necessary to investigate a specific report, to comply with a legal obligation, or to protect the platform and its users.

Where an investigation requires your input, we will normally contact your registered abuse or technical contact with a description of the issue and a deadline for response. Deadlines reflect severity: hours for active phishing, malware distribution, or an ongoing attack; several business days for lower-risk matters such as configuration issues or contested content claims.

Cooperating fully and quickly is the single biggest factor in how a matter resolves. Failing to respond, providing inaccurate information, or repeatedly recurring abuse after remediation will escalate our response.

If you believe AhuraSense Cloud infrastructure is being used in breach of this Policy, report it to [email protected]. Reports are reviewed by our abuse team, and reports concerning child safety, active phishing, and ongoing attacks are prioritised.

A useful report includes the IP address, hostname, or endpoint URL involved; accurate timestamps with the time zone; relevant log excerpts or raw message headers; a description of the harm; and contact details we can use for follow-up. Reports without enough detail to identify the responsible resource are difficult to action.

• Legal notices, infringement claims, and requests from law enforcement should go to [email protected].
• Please do not include copies of illegal material in your report — describe its location instead, and report child sexual abuse material directly to law enforcement or the relevant national hotline as well as to us.
• We treat reporter identities as confidential where we can, but we may need to share details of a report with the customer to allow them to remediate it.

Where a breach is limited to specific material or a specific resource, we prefer the narrowest effective remedy. That may mean disabling access to a file or bucket, null-routing a single IP address, disabling an inference endpoint or API key, stopping one instance, or removing a DNS record — rather than acting against your whole account.

Where we can, we will ask you to remove or remediate the material yourself within a stated window, because you are better placed to do so without disrupting your other workloads. Where the harm is active and severe, we may remove or disable access first and notify you immediately afterwards.

We will tell you what was actioned and why, and we will restore access if a subsequent review shows the material did not breach this Policy. We do not maintain copies of removed customer content beyond what we are required to retain for legal or evidential purposes.

Our response to a breach is normally graduated, and we escalate only as far as the circumstances require. The typical sequence is: a notice describing the issue with a remediation window; a limitation such as rate limiting, egress caps, or blocking a specific port or protocol; suspension of the affected resources or of the account; and finally termination.

We reserve the right to skip these steps and act immediately, without prior notice, where the conduct causes or threatens severe harm. That includes child safety matters, active phishing or malware distribution, attacks originating from our network, compromised instances, conduct exposing us or other customers to legal liability, and payment fraud. In those cases we act first and notify you as soon as practicable.

In assessing severity we consider the harm caused, whether the conduct was intentional, its scale and duration, your history on the platform, the speed and quality of your response, and whether you have taken credible steps to prevent recurrence. Repeat breaches of the same kind, or evasion of an enforcement action, will usually result in termination.

Fees remain payable for the period up to suspension or termination, and suspension for breach does not entitle you to a refund or service credit. If you believe an enforcement action was mistaken, you may ask us to review it by writing to [email protected]; no appeal is available for terminations under the child safety section.

We comply with valid legal process from competent authorities in the jurisdictions in which we operate, including India and the United Kingdom. We assess each request for validity, scope, and proportionality, and we push back on requests that are overbroad or that lack a proper legal basis.

Where a request concerns your account, we will notify you so that you can seek to challenge it, unless we are legally prohibited from doing so or notification would create a risk to life or to an investigation. We may preserve data pending a valid request, and we may disclose information voluntarily where we believe in good faith that it is necessary to prevent imminent harm to a person.

Certain categories — child sexual abuse material above all — are reported proactively to the relevant authorities and child protection organisations without waiting for a request, and without notice to the account holder.

We may update this Policy to reflect changes in law, in the threat landscape, or in the services we offer. Material changes will be posted on this page with a revised "Last updated" date and, where the change materially restricts a use you rely on, we will make reasonable efforts to notify account administrators in advance. Continued use of the services after a change takes effect constitutes acceptance of the updated Policy.

This Policy should be read alongside our Terms of Service and Privacy Policy. If you are unsure whether a planned workload is permitted — particularly for security research, red-teaming, high-volume scanning, bulk messaging, mining on dedicated capacity, or a high-risk AI application — ask us before you deploy. We would much rather scope it with you in advance than discover it through an abuse report.

• Abuse reports and enforcement questions: [email protected]
• Legal notices, infringement claims, and law enforcement requests: [email protected]