




















🕒 阅读时间:4 分钟 📝 字数:1370 👀 阅读量: Loading...
本文中所有转载内容均遵守原作者的版权协议。除特别声明外,本文其余部分默认遵守 Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC-BY-NC-SA-4.0)。
1983 年 9 月 27 日
美东夏令时 12:35:59
北京时间 9 月 28 日 00:35:59
Richard Stallman 在 net.unix-wizards 和 net.usoft 新闻组发布了 GNU 项目宣言。它宣告了 GNU 项目的诞生,并播下了自由软件运动的种子。
GNU 项目不仅在操作系统领域引领自由软件运动,在人工智能时代也展现了自由精神。GNU 致力于将自由软件原则扩展到 AI 领域,以确保技术进步不被少数公司垄断。GNU Gneural Network 项目(虽然已废弃)就是一个典型例子,它提供了一个自由的神经网络工具,允许用户定义、训练和使用神经网络,而无需依赖专有软件。这体现了 GNU 对知识共享、透明度和用户自由的坚持。
Nowadays, companies such as Google and IBM are doing a great service to all of us by showing what can be achieved by using Artificial Intelligence (AI). For instance, the results achieved by AlphaGo and Watson are outstanding and truly inspiring (the least one can say). But the fact that only companies and labs have access to this technology can represent a threat. First of all, we cannot know how money driven companies are going to use this novel technology. Second, this monopoly slows down technology adoption. As a matter of fact, data and trained neural networks are rarely shared, and one should always remember that in practice without the training data it is almost impossible to duplicate results (even when the source code is available).
We should ask ourselves: Do we really want that only a few can use AI? Should not Progress and Knowledge be for everyone instead?
In spite of the plenty of AI codes available and released under some permissive license, it is time for a Free AI which gives complete freedom to the users and the community. Hopefully, this could reduce the duplication of efforts one observes today.
This is why the author, Jean Michel Sellier, decided to create Gneural Network and release it under a free license like GPL.
中文翻译:如今,像谷歌和IBM这样的公司通过展示人工智能(AI)的潜力,为我们所有人提供了巨大服务。例如,AlphaGo和Watson取得的成果非常出色且鼓舞人心(至少可以说)。但只有公司和实验室才能访问这项技术,这可能构成威胁。首先,我们无法知道以盈利为驱动的公司将如何使用这项新技术。其次,这种垄断会减缓技术采用速度。实际上,数据和训练过的神经网络很少被共享,人们应该记住,在实践中,没有训练数据,几乎不可能复制结果(即使源代码可用)。
我们应该问自己:我们真的希望只有少数人能使用AI吗?进步和知识难道不应该属于所有人吗?
尽管有大量AI代码以宽松许可证发布,但现在是时候创建一个自由AI,为用户和社区提供完全自由。希望这能减少今天观察到的重复努力。
这就是为什么作者Jean Michel Sellier决定创建Gneural Network并以GPL等自由许可证发布的原因。
(以上内容原文转载自 GNU Gneural Network 项目页面,遵守 Creative Commons Attribution-NoDerivatives 4.0 International License。)
Free Software Foundation (FSF) 认为人工智能 (AI) 技术应该遵循自由软件的原则,以确保用户自由和透明。FSF 强调,AI 模型、算法和数据应开放共享,避免被少数公司垄断。FSF 发起的 “Free Software and AI” 活动呼吁开发者使用自由许可证发布 AI 工具,并反对专有 AI 系统对社会的潜在危害,如隐私侵犯和不透明决策。Richard Stallman 等 GNU 领袖认为,AI 应服务于人类自由,而不是控制人类。
(以上内容基于 Free Software Foundation 的立场总结,遵守 Creative Commons Attribution-NoDerivatives 4.0 International License。)
随着 AI 技术的快速发展,自由软件社区也在探讨其影响。在 LibrePlanet 讨论中,有人指出 AI 在代码生成和审查中的应用,但强调不应过度依赖 AI,应保持人类判断和透明性。AI 可以辅助开发,但自由软件的核心原则——自由、透明和社区协作——仍需坚守,以避免商业利益的垄断。讨论中提到,AI 在某些任务如感知和自动化判断上有进步,但在预测社会结果方面存在根本性局限。因此,自由软件社区应继续倡导透明的开发过程,确保 AI 工具本身也是自由的,以维护技术的民主化。
I’ve been doing applied research on auditing black-box algorithms and AI systems with a team at Princeton and MIT and CU Boulder, backed by Mozilla (say what you will about them and their 2024 layoffs, restructuring, and turn towards AI — and in general). One of my teammates, Samantha Dalal (CU Boulder), interviewed Sayash Kapoor (Princeton) for a podcast about his popular writing in AISnakeOil.com. You might enjoy listening - https://kgnu.org/looks-like-new-what-is-artificial-intelligence-capable-of/
And to get at the heart of the question here about AI and code, Sayash cites a presentation (see slides at https://www.cs.princeton.edu/~arvindn/talks/MIT-STS-AI-snakeoil.pdf) from Nov. 2019 (almost six years ago!), three categories of use for AI their relative accuracy:
Perception, e.g. image or music recognition like Shazam — genuine progress
Automating judgement, e.g. spam filtering — imperfect, but improving
Predicting social outcomes, e.g. prison recidivism — fundamentally dubious, “no matter how much data you throw at it” The presentation concluded with the following takeaway (with my comments):
- AI excels at some tasks, but can’t predict social outcomes (especially for racial and economic justice, let alone technical skill and ability in the form of writing and reviewing code).
- We must resist the enormous commercial interests that aim to obfuscate this fact (hence, FLOSS - but more, of course, as evidence by this list).
- In most cases, manual scoring rules are just as accurate, far more transparent, and worth considering (yay for the enduring human spirit and dignified work). I hope this helps! FLOSS is undead, long live FLOSS!
中文翻译:我一直在与普林斯顿大学、MIT 和科罗拉多大学博尔德分校的一个团队进行应用研究,审计黑箱算法和 AI 系统,由 Mozilla 支持(随便你怎么说他们和他们的 2024 年裁员、重构以及转向 AI——以及一般情况)。我的队友 Samantha Dalal(科罗拉多大学博尔德分校)采访了 Sayash Kapoor(普林斯顿大学),关于他在 AISnakeOil.com 的流行写作。你可能会喜欢听听 - https://kgnu.org/looks-like-new-what-is-artificial-intelligence-capable-of/
说到这里关于 AI 和代码的核心问题,Sayash 引用了一个演示,AI 使用的三个类别及其相对准确性:
(以上内容原文转载自 LibrePlanet 讨论邮件列表,遵守相关版权协议。)
此内容由惯性聚合(RSS阅读器)自动聚合整理,仅供阅读参考。 原文来自 — 版权归原作者所有。