We’re founded on trust, and our purpose is giving you the right result, and getting you there — as fast as we can. We use use a range of techniques to protect your privacy.
Your personal data in Epi is yours, end of story. We respect your intellectual property. You have strong privacy choices in the use of our services.
This privacy feature is fundamental and world-leading. You can choose to help improve or not, and as we believe that you should determine your own privacy choices, the default is “no thanks”.
If you choose to help improve, you improve the service for everyone, and you build trust on the Internet. It will reveal websites that are missing in Epi so we can launch your favourite sites, and we can further discover bad links and phishing so we can better answer your link checks. When helping to improve, a sample of selected queries are recorded with differential privacy in our telemetry system.
If you choose not to help improve, we record the time at which queries occur but we do not record queries. This is in contrast to major search engines, which record and store all your queries, tightly bound to your identity. Also, although most major privacy-focused search engines do not record your searches against your identity, they will store all queries made in their search engine. Queries can be sensitive and can identify people — we don’t record them unless you choose to help improve.
When you enable this feature, all websites launched by Epi will open in your browser’s private browsing mode.
We employ a variety of layers in our processes to ensure you are in control of your Epi data. We consider customer privacy at every stage of development and in every area of the business.
Privacy at the heart. We take privacy seriously throughout our business. Your data is not in our business model. We can’t personally identify you, we don’t share your data, and we don’t track you.
Customer isolation. We secure data in customer-specific containers, and where required, in dedicated customer environments, to keep customer data separate and isolated.
Customer-key and end-to-end encryption. We protect application data with strong cryptographic techniques. Information is secured with keys specific to the customer, and where possible, with customer end-to-end encryption.
Epi-key encryption. We additionally secure Epi data with keys that we manage to protect data at rest on our servers with server-side encryption, rather than solely relying on cloud provider encryption.
Infrastructure security. We use infrastructure providers that provide strong security-of-the-cloud operations, like Amazon Web Services. We follow best cloud practices and industry-standard principles in our deployment to provide strong security-in-the-cloud operations.
Transport encryption. All data is further protected by another layer of industry-standard security in transit with HTTPS encryption.
Secure code principles. We employ secure code principles in all our development. We design, write and review code for its security and privacy implications. We evaluate and monitor supply chain dependencies for their actions. We deploy in secure environments and utilise tools like Content Security Policy (CSP) to protect clients and client data.
Principle of least privilege. We believe in the principle of minimal access. Our staff only have access to what they need, for when they need it, and no more. Our identity and access management (IAM) infrastructure policies enforce this principle. Senior engineers, the CTO and the CEO are even limited in their capabilities in our cloud platform.
Provider security. We evaluate prospective providers and select those with a reputation for privacy and security of customer data. We use strong authentication with all providers, including multi-factor authentication. We do not use third-party analytics or external telemetry providers to limit third-party tracking originating from your queries and data.
Better services shouldn’t come at the expense of your privacy.
Differential privacy. When we collect telemetry, this technique allows us to see general patterns — without specifics that could be traced back to you.
Data minimisation. If we don’t need it, we don’t collect it.