Open Source is an idea more than it is a set of specs, and that idea is now incompatible with AI, according to the moderators of the r\/opensource subreddit.<\/p>\n
\u201cThe long story short, if your project uses AI, is made by AI, or is intended for AI, it\u2019s not welcome on r\/opensource<\/a>. AI, as it exists today, is incongruent to this Open Source community,\u201d wrote<\/a> moderator Wolvereness on Saturday, in a post spotted by the Verge<\/a>.<\/p>\n The new batch of subreddit rules Wolvereness announced says projects posted to r\/opensource must not make obvious use of AI. \u201cTokens are cheap, and we don\u2019t care how many you spent to make your amazing project,\u201d the post explains. Another rule says a project must not be \u201cintended for AI,\u201d meaning no human-written AI model context protocols<\/a> (MCPs) or other AI-adjacent projects.<\/p>\n Wolvereness\u2019 post also includes rules unrelated to AI, such as bans on posts from users who don\u2019t speak English, projects less than a year old, and posts from Reddit accounts less than a year old.<\/p>\n As many on both the pro-<\/a> and anti-AI<\/a> sides have already pointed out, projects thought of as core to the open source movement now make use of AI coding, which arguably makes this subreddit an outlier. For instance, open source pioneer Linus Torvalds announced<\/a> that he used Claude Code to write for Linux kernel earlier this month.<\/p>\n With Reddit banning open-source projects that have ANY AI contributions, here’s a list of projects you can no longer discuss there:<\/p>\n Linux, Chromium, Firefox, WordPress, VS Code, Python, VLC, Blender, Git, Docker, LibreOffice, React, Node.js, PHP, TypeScript, Rust, Home\u2026 https:\/\/t.co\/96gsFVAXAb<\/a><\/p>\n \u2014 chasm (@chasmmmmmmmmmmm) October 25, 2026<\/a><\/p>\n The developer world is in an ecstatic \u201copen source\u201d frenzy at the moment as consumer-grade AI coding has reached levels of sophistication that allow hobbyists to decompile programs at their leisure<\/a>. Users willing and able to expend mass quantities of tokens can write prompts that ask models to work backwards, starting with published code that can only be read by computers and writing human-readable\u2014and therefore editable\u2014versions.<\/p>\n In some cases this results in clean-room\u2014meaning theoretically legal<\/a>\u2014projects like ArtCraft<\/a>, a free, vibe coded recreation of Adobe Creative Suite, published with a new name. ArtCraft is labeled as open source, and open to contributions on GitHub, but could never be allowed on r\/opensource according to the new rules.<\/p>\n all games are effectively open source atp, opus can reverse engineer any game effectively by faking controller\/keyboard inputs and observing state! Do this with enough instances\/compute you can reverse engineer any game! https:\/\/t.co\/a1g67EJj4f<\/a><\/p>\n \u2014 Rehan Sheikh (@rehan_shei) September 28, 2026<\/a><\/p>\n \u00a0<\/p>\n And a common refrain online lately is \u201call games are now open source\u201d\u2014but that\u2019s entirely because of AI. Decompiling\u00a0 with AI has resulted in a flood of more legally questionable projects in which video games are decompiled, turning them into de facto free software\u2014which is often posted on the subreddit r\/decomps<\/a>. That subreddit has also become inundated with AI lately, but in recent weeks it has developed its own AI-inclusive open source ethos<\/a>.<\/p>\n Broadly speaking, the community on r\/opensource appears receptive to the new rule<\/a>, but with many exceptions<\/a>. Those who feel alienated by the change have already jumped ship to r\/TrueOpenSource<\/a>, an AI-tolerant open source subreddit created yesterday.<\/p>\n An edit from Wolvereness embraces the schism, offering a \u201cshout out\u201d to the creator of r\/TrueOpenSource, and saying they hope it will be \u201ca great refuge for those affected by this policy.\u201d<\/p>\n","protected":false},"excerpt":{"rendered":"Open Source is an idea more than it is a set of specs, and that idea is now…\n","protected":false},"author":2,"featured_media":813832,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[554,733,4308,29721,621,86,56,54,55],"class_list":["post-813831","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-open-source-software","tag-reddit","tag-technology","tag-uk","tag-united-kingdom","tag-unitedkingdom"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/813831","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/comments?post=813831"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/813831\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media\/813832"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media?parent=813831"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/categories?post=813831"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/tags?post=813831"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}