domingo, 26 de abril de 2026

FSF Condemns Responsible AI Licences as Unethical

The Free Software Foundation has taken a firm stance against so-called 'Responsible AI' Licences, declaring them both unethical and incompatible with software freedom. In a blog post published this week, the FSF's Licensing and Compliance Manager explicitly criticised RAIL licences, which attempt to restrict artificial intelligence and machine learning software from being used in potentially harmful applications such as surveillance and crime prediction.

FSF Condemns Responsible AI Licences as Unethical

Whilst these licences are marketed as addressing ethical concerns in AI development, the FSF argues they fundamentally miss the mark. The organisation points out that RAIL licences fail to require the essential components necessary for users to truly control their computing with machine learning tools. These missing elements include complete training inputs, training configuration settings, trained models, and crucially, the source code of software used for training, testing, and running machine learning-based tools.

According to the FSF, this means that machine learning systems under RAIL can still be unethical, as use restrictions alone do not prevent these licences from being wielded as a means of exercising power over users. The Foundation further contends that RAIL licences contribute to unethical marketing of machine learning, hiding behind morally-loaded restrictions that they claim to enforce but cannot truly guarantee.

Instead of restricting how software can be used, the FSF advocates for more effective approaches to addressing social injustice through technology. These include government and community support for freedom-respecting tools and services, releasing programmes under strong copyleft licences, and entrusting copyrights to organisations with the resources to enforce copyleft provisions. The Foundation maintains that software freedom must be defended rather than denied, arguing that greater availability of free software encourages collaboration on tools that avoid moral dangers whilst solving existing problems. By denying people the freedom to use particular programmes, RAIL licences ultimately prevent software from being used for the common good.

Fuente Original: https://news.slashdot.org/story/26/04/25/079207/free-software-foundation-says-responsible-ai-licenses-which-restrict-harmful-uses-are-unethical-and-nonfree?utm_source=rss1.0mainlinkanon&utm_medium=feed

Artículos relacionados de LaRebelión:

Artículo generado mediante LaRebelionBOT

No hay comentarios:

Publicar un comentario