Where to report
We accept all reports at [email protected] in English, Russian or Arabic. Use the subject line to say what your message is about:
- Vulnerabilities in the website or products: subject “Vulnerability report”. Details: Vulnerability disclosure policy
- Abuse: subject “Abuse report”. Details: Abuse reports
- Requests about your personal data: subject “Personal data”. Details: Privacy policy
- Difficulties with reading or navigation: subject “Accessibility”. Details: Accessibility
Security and abuse reports are handled first.
Abuse reports
Tell us if you notice abuse involving Plasma EdTech, Plasmida.ai or Plasma [Learn]:
- spam or phishing in our name or sent from our domains;
- malicious files or links on our websites or in our products;
- illegal, abusive or harmful content posted by users of our products;
- use of our products for fraud, harassment or to infringe other people’s rights;
- copyright or trademark infringement;
- websites or accounts impersonating Plasma EdTech.
Email [email protected] with the subject “Abuse report” and include links, screenshots, the date and time with time zone and, for emails, the full message headers. For an infringement complaint, identify the work or trademark concerned and how you hold the rights to it.
We acknowledge complaints within 2 business days and prioritise phishing, malware and threats to people’s safety. After review, we may remove content, restrict or suspend an account and, where the law requires it, share information with the authorities. We will tell you the outcome unless doing so would breach other people’s rights or the law.
We do not reveal who made a complaint unless the law requires it. Plasma EdTech never asks for passwords or payment details by email or messenger.
How the website is protected
- The website is served over HTTPS only; HTTP requests are redirected to a secure connection.
- Browsers receive strict security headers: a Content Security Policy, a ban on embedding the site in other pages, and HSTS.
- Fonts are hosted on our own server and the Google map loads only on request, so pages do not contact third-party services when they open.
- The website has no accounts, payments, comments or forms, so it does not collect passwords or payment details.
- The website sets no cookies of its own and uses no analytics or advertising pixels.
- External links open with noopener and noreferrer, so other websites cannot access the page that opened them.
- Only staff who need it for their work can access correspondence and hosting logs.
AI transparency
Plasma EdTech products use language models and are still in development. By the time each product launches, its documentation will state:
- which AI models and providers are used, and for which features;
- in which country learners’ data is processed and stored;
- whether learners’ data is used to train models. Our principle is not to use learning data or conversations for this without the learner’s explicit consent;
- where people review decisions. Our principle is that, at the learner’s request, a teacher or staff member reviews any assessment that affects their studies or career;
- how to recognise AI output. Our principle is to label answers and materials generated by AI in the interface.
This website does not use AI to process visitors’ data.
The principles and limits that govern our use of AI are set out in “Responsible use of AI”.
Service providers
The website relies on the following providers:
- Cloudflare: DNS, page delivery through its CDN and filtering of malicious requests. Cloudflare processes visitors’ IP addresses, browser details and request data; Cloudflare Access protects staff access to the admin;
- Google Maps: the map in the Contact section, loaded only on request;
- our email provider: correspondence sent to [email protected];
- our hosting provider: hosting the website and its technical logs.
The privacy policy and the cookies document explain what data these providers receive.
Products
The same rules for security testing and reporting apply to the website and our products. Each product’s documentation explains how it handles learning data.