all messages for Emacs-related lists mirrored at yhetil.org
 help / color / mirror / code / Atom feed
From: Debanjum Singh Solanky <debanjum@khoj.dev>
To: rms@gnu.org
Cc: Ihor Radchenko <yantar92@posteo.net>,
	emacs-tangents@gnu.org, jporterbugs@gmail.com, ahyatt@gmail.com,
	team@khoj.dev
Subject: Re: Collaborative training of Libre LLMs (was: Is ChatGTP SaaSS? (was: [NonGNU ELPA] New package: llm))
Date: Sat, 9 Sep 2023 19:18:44 -0700	[thread overview]
Message-ID: <CAM_WPJStUsCM0jSyPEBjnjFwg-aV=fL7JHkSEGFT=_Xir_qRGA@mail.gmail.com> (raw)
In-Reply-To: <E1qf8Dj-0000ay-02@fencepost.gnu.org>

[-- Attachment #1: Type: text/plain, Size: 1995 bytes --]

>   > However, if the "patching" technology can only serve a single "patch" +
>   > main model, there is a problem. Improving libre neural networks will
>   > become difficult, unless people utilize collaborative server to
>   > continuously improve a model.
>
>   > Such collaborative server, similar to ChatGPT, will combine "editing"
>   > (training) and "consulting" together. And, unlike Wikipedia, these
>   > activities are hard to separate.
>
> If the users in this "community" can't move their work outside of a
> private "collaborative server", they are in effect prisoners of that
> server.  Whoever keeps them stuck there will have power, and that will
> tempt per to mistreat them with it.
>

Versus traditional software, AI systems rely critically on the usage
data generated to improve the original model. Using copyleft licensed
models maybe enough to prevent a server owner from being able
to train a better closed model? This would prevent them from holding
users hostage on their server.



>   > This raises a moral question about practical ways to improve libre
>   > neural networks without falling into SaaSS practices.
>
> From the example above, I conclude it is crucial that people who use a
> particular platform to modify and run the model have the feasible
> freedom of copying their modified versions off that platform and onto
> any other platform that satisfies the specs needed to run these models.
>

Platform portability does not solve for how to improve libre
neural networks in an open, community guided way.

To collaboratively develop better open models we'd need the generated
usage data to be publically shareable. Attempts like open-assistant
(https://open-assistant.io) that share usage data under cc-by-sa maybe
a good enough solution for this. But it'll fall on the server owners
to get explicit user consent and clean sensitive usage data to share
this data publically without liability.

--
Debanjum Singh Solanky
Founder, Khoj (https://khoj.dev/)

[-- Attachment #2: Type: text/html, Size: 3980 bytes --]

  reply	other threads:[~2023-09-10  2:18 UTC|newest]

Thread overview: 68+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-07 23:54 [NonGNU ELPA] New package: llm Andrew Hyatt
2023-08-08  5:42 ` Philip Kaludercic
2023-08-08 15:08   ` Spencer Baugh
2023-08-08 15:09   ` Andrew Hyatt
2023-08-09  3:47 ` Richard Stallman
2023-08-09  4:37   ` Andrew Hyatt
2023-08-13  1:43     ` Richard Stallman
2023-08-13  1:43     ` Richard Stallman
2023-08-13  2:11       ` Emanuel Berg
2023-08-15  5:14       ` Andrew Hyatt
2023-08-15 17:12         ` Jim Porter
2023-08-17  2:02           ` Richard Stallman
2023-08-17  2:48             ` Andrew Hyatt
2023-08-19  1:51               ` Richard Stallman
2023-08-19  9:08                 ` Ihor Radchenko
2023-08-21  1:12                   ` Richard Stallman
2023-08-21  8:26                     ` Ihor Radchenko
2023-08-17 17:08             ` Daniel Fleischer
2023-08-19  1:49               ` Richard Stallman
2023-08-19  8:15                 ` Daniel Fleischer
2023-08-21  1:12                   ` Richard Stallman
2023-08-21  4:48               ` Jim Porter
2023-08-21  5:12                 ` Andrew Hyatt
2023-08-21  6:03                   ` Jim Porter
2023-08-21  6:36                 ` Daniel Fleischer
2023-08-22  1:06                 ` Richard Stallman
2023-08-16  2:30         ` Richard Stallman
2023-08-16  5:11           ` Tomas Hlavaty
2023-08-18  2:10             ` Richard Stallman
2023-08-27  1:07       ` Andrew Hyatt
2023-08-27 13:11         ` Philip Kaludercic
2023-08-28  1:31           ` Richard Stallman
2023-08-28  2:32             ` Andrew Hyatt
2023-08-28  2:59               ` Jim Porter
2023-08-28  4:54                 ` Andrew Hyatt
2023-08-31  2:10                 ` Richard Stallman
2023-08-31  9:06                   ` Ihor Radchenko
2023-08-31 16:29                     ` chad
2023-09-01  9:53                       ` Ihor Radchenko
2023-09-04  1:27                     ` Richard Stallman
2023-09-04  1:27                     ` Richard Stallman
2023-09-06 12:25                       ` Ihor Radchenko
2023-09-06 12:51                       ` Is ChatGTP SaaSS? (was: [NonGNU ELPA] New package: llm) Ihor Radchenko
2023-09-06 16:59                         ` Andrew Hyatt
2023-09-09  0:37                           ` Richard Stallman
2023-09-06 22:52                         ` Emanuel Berg
2023-09-07  7:28                           ` Lucien Cartier-Tilet
2023-09-07  7:57                             ` Emanuel Berg
2023-09-09  0:38                         ` Richard Stallman
2023-09-09 10:28                           ` Collaborative training of Libre LLMs (was: Is ChatGTP SaaSS? (was: [NonGNU ELPA] New package: llm)) Ihor Radchenko
2023-09-09 11:19                             ` Jean Louis
2023-09-10  0:22                             ` Richard Stallman
2023-09-10  2:18                               ` Debanjum Singh Solanky [this message]
2023-08-27 18:36         ` [NonGNU ELPA] New package: llm Jim Porter
2023-08-28  0:19           ` Andrew Hyatt
2023-09-04  1:27           ` Richard Stallman
2023-09-04  5:18             ` Andrew Hyatt
2023-09-07  1:21               ` Richard Stallman
2023-09-12  4:54                 ` Andrew Hyatt
2023-09-12  9:57                   ` Philip Kaludercic
2023-09-12 15:05                   ` Stefan Kangas
2023-09-19 16:26                     ` Andrew Hyatt
2023-09-19 16:34                       ` Philip Kaludercic
2023-09-19 18:19                         ` Andrew Hyatt
2023-09-04  1:27         ` Richard Stallman
2023-08-09  3:47 ` Richard Stallman
2023-08-09  4:06   ` Andrew Hyatt
2023-08-12  2:44     ` Richard Stallman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CAM_WPJStUsCM0jSyPEBjnjFwg-aV=fL7JHkSEGFT=_Xir_qRGA@mail.gmail.com' \
    --to=debanjum@khoj.dev \
    --cc=ahyatt@gmail.com \
    --cc=emacs-tangents@gnu.org \
    --cc=jporterbugs@gmail.com \
    --cc=rms@gnu.org \
    --cc=team@khoj.dev \
    --cc=yantar92@posteo.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this external index

	https://git.savannah.gnu.org/cgit/emacs.git
	https://git.savannah.gnu.org/cgit/emacs/org-mode.git

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.