mirror of
https://github.com/privacyguides/privacyguides.org.git
synced 2025-07-24 04:11:05 +00:00
style: Move AI terms from glossary to recommendation page (#2974)
Signed-off-by: Jonah Aragon <jonah@privacyguides.org> Signed-off-by: fria <fria@privacyguides.org>
This commit is contained in:
redoomed1
@@ -11,7 +11,7 @@ cover: ai-chatbots.webp
|
||||
- [:material-account-cash: Surveillance Capitalism](basics/common-threats.md#surveillance-as-a-business-model){ .pg-brown }
|
||||
- [:material-close-outline: Censorship](basics/common-threats.md#avoiding-censorship){ .pg-blue-gray }
|
||||
|
||||
Since the release of ChatGPT in 2022, interactions with Large Language Models (LLMs) have become increasingly common. LLMs can help us write better, understand unfamiliar subjects, or answer a wide range of questions. They can statistically predict the next word based on a vast amount of data scraped from the web.
|
||||
The use of **AI chat**, also known as Large Language Models (LLMs), has become increasingly common since the release of ChatGPT in 2022. LLMs can help us write better, understand unfamiliar subjects, or answer a wide range of questions. They work by statistically predicting the next word in their responses based on a vast amount of data scraped from the web.
|
||||
|
||||
## Privacy Concerns About LLMs
|
||||
|
||||
@@ -41,7 +41,7 @@ To run AI locally, you need both an AI model and an AI client.
|
||||
|
||||
### Choosing a Model
|
||||
|
||||
There are many permissively licensed models available to download. [Hugging Face](https://huggingface.co/models) is a platform that lets you browse, research, and download models in common formats like [GGUF](https://huggingface.co/docs/hub/en/gguf). Companies that provide good open-weights models include big names like Mistral, Meta, Microsoft, and Google. However, there are also many community models and 'fine-tunes' available. As mentioned above, quantized models offer the best balance between model quality and performance for those using consumer-grade hardware.
|
||||
There are many permissively licensed models available to download. [Hugging Face](https://huggingface.co/models) is a platform that lets you browse, research, and download models in common formats like [GGUF](https://huggingface.co/docs/hub/en/gguf). Companies that provide good open-weights models include big names like Mistral, Meta, Microsoft, and Google. However, there are also many community models and [fine-tuned](https://en.wikipedia.org/wiki/Fine-tuning_(deep_learning)) models available. As mentioned above, quantized models offer the best balance between model quality and performance for those using consumer-grade hardware.
|
||||
|
||||
To help you choose a model that fits your needs, you can look at leaderboards and benchmarks. The most widely-used leaderboard is the community-driven [LM Arena](https://lmarena.ai). Additionally, the [OpenLLM Leaderboard](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) focuses on the performance of open-weights models on common benchmarks like [MMLU-Pro](https://arxiv.org/abs/2406.01574). There are also specialized benchmarks which measure factors like [emotional intelligence](https://eqbench.com), ["uncensored general intelligence"](https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard), and [many others](https://www.nebuly.com/blog/llm-leaderboards).
|
||||
|
||||
@@ -62,7 +62,7 @@ To help you choose a model that fits your needs, you can look at leaderboards an
|
||||
|
||||
{align=right}
|
||||
|
||||
Kobold.cpp is an AI client that runs locally on your Windows, Mac, or Linux computer. It's an excellent choice if you are looking for heavy customization and tweaking, such as for role-playing purposes.
|
||||
**Kobold.cpp** is an AI client that runs locally on your Windows, Mac, or Linux computer. It's an excellent choice if you are looking for heavy customization and tweaking, such as for role-playing purposes.
|
||||
|
||||
In addition to supporting a large range of text models, Kobold.cpp also supports image generators such as [Stable Diffusion](https://stability.ai/stable-image) and automatic speech recognition tools such as [Whisper](https://github.com/ggerganov/whisper.cpp).
|
||||
|
||||
@@ -82,7 +82,7 @@ In addition to supporting a large range of text models, Kobold.cpp also supports
|
||||
|
||||
</div>
|
||||
|
||||
<div class="admonition note" markdown>
|
||||
<div class="admonition info" markdown>
|
||||
<p class="admonition-title">Compatibility Issues</p>
|
||||
|
||||
Kobold.cpp might not run on computers without AVX/AVX2 support.
|
||||
@@ -97,7 +97,7 @@ Kobold.cpp allows you to modify parameters such as the AI model temperature and
|
||||
|
||||
{align=right}
|
||||
|
||||
Ollama is a command-line AI assistant that is available on macOS, Linux, and Windows. Ollama is a great choice if you're looking for an AI client that's easy-to-use, widely compatible, and fast due to its use of inference and other techniques. It also doesn't involve any manual setup.
|
||||
**Ollama** is a command-line AI assistant that is available on macOS, Linux, and Windows. Ollama is a great choice if you're looking for an AI client that's easy-to-use, widely compatible, and fast due to its use of inference and other techniques. It also doesn't involve any manual setup.
|
||||
|
||||
In addition to supporting a wide range of text models, Ollama also supports [LLaVA](https://github.com/haotian-liu/LLaVA) models and has experimental support for Meta's [Llama vision capabilities](https://huggingface.co/blog/llama32#what-is-llama-32-vision).
|
||||
|
||||
@@ -123,9 +123,9 @@ Ollama simplifies the process of setting up a local AI chat by downloading the A
|
||||
|
||||
<div class="admonition recommendation" markdown>
|
||||
|
||||
{align=right}
|
||||
{align=right}
|
||||
|
||||
Llamafile is a lightweight single-file executable that allows users to run LLMs locally on their own computers without any setup involved. It is [backed by Mozilla](https://hacks.mozilla.org/2023/11/introducing-llamafile) and available on Linux, macOS, and Windows.
|
||||
**Llamafile** is a lightweight, single-file executable that allows users to run LLMs locally on their own computers without any setup involved. It is [backed by Mozilla](https://hacks.mozilla.org/2023/11/introducing-llamafile) and available on Linux, macOS, and Windows.
|
||||
|
||||
Llamafile also supports LLaVA. However, it doesn't support speech recognition or image generation.
|
||||
|
||||
@@ -137,7 +137,9 @@ Llamafile also supports LLaVA. However, it doesn't support speech recognition or
|
||||
<details class="downloads" markdown>
|
||||
<summary>Downloads</summary>
|
||||
|
||||
- [:fontawesome-solid-desktop: Desktop](https://github.com/Mozilla-Ocho/llamafile#quickstart)
|
||||
- [:fontawesome-brands-windows: Windows](https://github.com/Mozilla-Ocho/llamafile#quickstart)
|
||||
- [:simple-apple: macOS](https://github.com/Mozilla-Ocho/llamafile#quickstart)
|
||||
- [:simple-linux: Linux](https://github.com/Mozilla-Ocho/llamafile#quickstart)
|
||||
|
||||
</details>
|
||||
|
||||
@@ -170,11 +172,11 @@ Please note we are not affiliated with any of the projects we recommend. In addi
|
||||
|
||||
### Minimum Requirements
|
||||
|
||||
- Must be open-source.
|
||||
- Must be open source.
|
||||
- Must not transmit personal data, including chat data.
|
||||
- Must be multi-platform.
|
||||
- Must not require a GPU.
|
||||
- Must support GPU-powered fast inference.
|
||||
- Must support GPU-powered, fast inference.
|
||||
- Must not require an internet connection.
|
||||
|
||||
### Best-Case
|
||||
@@ -185,4 +187,11 @@ Our best-case criteria represent what we *would* like to see from the perfect pr
|
||||
- Should have a built-in model downloader option.
|
||||
- The user should be able to modify the LLM parameters, such as its system prompt or temperature.
|
||||
|
||||
*[LLaVA]: Large Language and Vision Assistant (multimodal AI model)
|
||||
*[LLM]: Large Language Model (AI model such as ChatGPT)
|
||||
*[LLMs]: Large Language Models (AI models such as ChatGPT)
|
||||
*[open-weights models]: AI models that anyone can download and use, but the underlying training data and/or algorithms for them are proprietary.
|
||||
*[system prompt]: The general instructions given by a human to guide how an AI chat should operate.
|
||||
*[temperature]: A parameter used in AI models to control the level of randomness and creativity in the generated text.
|
||||
|
||||
[^1]: A file checksum is a type of anti-tampering fingerprint. A developer usually provides a checksum in a text file that can be downloaded separately, or on the download page itself. Verifying that the checksum of the file you downloaded matches the one provided by the developer helps ensure that the file is genuine and wasn't tampered with in transit. You can use commands like `sha256sum` on Linux and macOS, or `certutil -hashfile file SHA256` on Windows to generate the downloaded file's checksum.
|
||||
|
||||
@@ -19,7 +19,7 @@ Even if you use OpenPGP, it does not support [forward secrecy](https://en.wikipe
|
||||
|
||||
## What is the Web Key Directory standard?
|
||||
|
||||
The Web Key Directory (WKD) standard allows email clients to discover the OpenPGP key for other mailboxes, even those hosted on a different provider. Email clients which support WKD will ask the recipient's server for a key based on the email address' domain name. For example, if you emailed `jonah@privacyguides.org`, your email client would ask `privacyguides.org` for Jonah's OpenPGP key, and if `privacyguides.org` has a key for that account, your message would be automatically encrypted.
|
||||
The [Web Key Directory (WKD)](https://wiki.gnupg.org/WKD) standard allows email clients to discover the OpenPGP key for other mailboxes, even those hosted on a different provider. Email clients which support WKD will ask the recipient's server for a key based on the email address' domain name. For example, if you emailed `jonah@privacyguides.org`, your email client would ask `privacyguides.org` for Jonah's OpenPGP key, and if `privacyguides.org` has a key for that account, your message would be automatically encrypted.
|
||||
|
||||
In addition to the [email clients we recommend](../email-clients.md) which support WKD, some webmail providers also support WKD. Whether *your own* key is published to WKD for others to use depends on your domain configuration. If you use an [email provider](../email.md#openpgp-compatible-services) which supports WKD, such as Proton Mail or Mailbox.org, they can publish your OpenPGP key on their domain for you.
|
||||
|
||||
|
||||
@@ -31,7 +31,7 @@ In addition to (or instead of) an email provider recommended here, you may wish
|
||||
|
||||
## OpenPGP Compatible Services
|
||||
|
||||
These providers natively support OpenPGP encryption/decryption and the [Web Key Directory standard](basics/email-security.md#what-is-the-web-key-directory-standard), allowing for provider-agnostic E2EE emails. For example, a Proton Mail user could send an E2EE message to a Mailbox.org user, or you could receive OpenPGP-encrypted notifications from internet services which support it.
|
||||
These providers natively support OpenPGP encryption/decryption and the [Web Key Directory (WKD) standard](basics/email-security.md#what-is-the-web-key-directory-standard), allowing for provider-agnostic E2EE emails. For example, a Proton Mail user could send an E2EE message to a Mailbox.org user, or you could receive OpenPGP-encrypted notifications from internet services which support it.
|
||||
|
||||
<div class="grid cards" markdown>
|
||||
|
||||
@@ -104,7 +104,7 @@ Certain information stored in [Proton Contacts](https://proton.me/support/proton
|
||||
|
||||
#### :material-check:{ .pg-green } Email Encryption
|
||||
|
||||
Proton Mail has [integrated OpenPGP encryption](https://proton.me/support/how-to-use-pgp) in their webmail. Emails to other Proton Mail accounts are encrypted automatically, and encryption to non-Proton Mail addresses with an OpenPGP key can be enabled easily in your account settings. Proton also supports automatic external key discovery with [Web Key Directory (WKD)](https://wiki.gnupg.org/WKD). This means that emails sent to other providers which use WKD will be automatically encrypted with OpenPGP as well, without the need to manually exchange public PGP keys with your contacts. They also allow you to [encrypt messages to non-Proton Mail addresses without OpenPGP](https://proton.me/support/password-protected-emails), without the need for them to sign up for a Proton Mail account.
|
||||
Proton Mail has [integrated OpenPGP encryption](https://proton.me/support/how-to-use-pgp) in their webmail. Emails to other Proton Mail accounts are encrypted automatically, and encryption to non-Proton Mail addresses with an OpenPGP key can be enabled easily in your account settings. Proton also supports automatic external key discovery with WKD. This means that emails sent to other providers which use WKD will be automatically encrypted with OpenPGP as well, without the need to manually exchange public PGP keys with your contacts. They also allow you to [encrypt messages to non-Proton Mail addresses without OpenPGP](https://proton.me/support/password-protected-emails), without the need for them to sign up for a Proton Mail account.
|
||||
|
||||
Proton Mail also publishes the public keys of Proton accounts via HTTP from their WKD. This allows people who don't use Proton Mail to find the OpenPGP keys of Proton Mail accounts easily, for cross-provider E2EE. This only applies to email addresses ending in one of Proton's own domains, like @proton.me. If you use a custom domain, you must [configure WKD](./basics/email-security.md#what-is-the-web-key-directory-standard) separately.
|
||||
|
||||
@@ -161,7 +161,7 @@ However, [Open-Exchange](https://en.wikipedia.org/wiki/Open-Xchange), the softwa
|
||||
|
||||
Mailbox.org has [integrated encryption](https://kb.mailbox.org/en/private/e-mail-article/send-encrypted-e-mails-with-guard) in their webmail, which simplifies sending messages to people with public OpenPGP keys. They also allow [remote recipients to decrypt an email](https://kb.mailbox.org/en/private/e-mail-article/my-recipient-does-not-use-pgp) on Mailbox.org's servers. This feature is useful when the remote recipient does not have OpenPGP and cannot decrypt a copy of the email in their own mailbox.
|
||||
|
||||
Mailbox.org also supports the discovery of public keys via HTTP from their [Web Key Directory (WKD)](https://wiki.gnupg.org/WKD). This allows people outside of Mailbox.org to find the OpenPGP keys of Mailbox.org accounts easily, for cross-provider E2EE. This only applies to email addresses ending in one of Mailbox.org's own domains, like @mailbox.org. If you use a custom domain, you must [configure WKD](./basics/email-security.md#what-is-the-web-key-directory-standard) separately.
|
||||
Mailbox.org also supports the discovery of public keys via HTTP from their WKD. This allows people outside of Mailbox.org to find the OpenPGP keys of Mailbox.org accounts easily, for cross-provider E2EE. This only applies to email addresses ending in one of Mailbox.org's own domains, like @mailbox.org. If you use a custom domain, you must [configure WKD](./basics/email-security.md#what-is-the-web-key-directory-standard) separately.
|
||||
|
||||
#### :material-information-outline:{ .pg-blue } Account Termination
|
||||
|
||||
@@ -320,7 +320,7 @@ We regard these features as important in order to provide a safe and optimal ser
|
||||
|
||||
- Encrypts all account data (Contacts, Calendars, etc.) at rest with zero-access encryption.
|
||||
- Integrated webmail E2EE/PGP encryption provided as a convenience.
|
||||
- Support for [WKD](https://wiki.gnupg.org/WKD) to allow improved discovery of public OpenPGP keys via HTTP.
|
||||
- Support for WKD to allow improved discovery of public OpenPGP keys via HTTP.
|
||||
GnuPG users can get a key by typing: `gpg --locate-key example_user@example.com`
|
||||
- Support for a temporary mailbox for external users. This is useful when you want to send an encrypted email, without sending an actual copy to your recipient. These emails usually have a limited lifespan and then are automatically deleted. They also don't require the recipient to configure any cryptography like OpenPGP.
|
||||
- Availability of the email provider's services via an [onion service](https://en.wikipedia.org/wiki/.onion).
|
||||
|
||||
@@ -362,7 +362,7 @@ We [recommend](dns.md#recommended-providers) a number of encrypted DNS servers b
|
||||
<div class="grid cards" markdown>
|
||||
|
||||
- { .twemoji loading=lazy } [Kobold.cpp](ai-chat.md#koboldcpp)
|
||||
- { .twemoji loading=lazy } [Llamafile](ai-chat.md#llamafile)
|
||||
- { .twemoji loading=lazy } [Llamafile](ai-chat.md#llamafile)
|
||||
- { .twemoji loading=lazy } [Ollama (CLI)](ai-chat.md#ollama-cli)
|
||||
|
||||
</div>
|
||||
|
||||
Reference in New Issue
Block a user