There are other tools available. Use them. Anything touched by Microsoft is poison. You have a choice. Choose “not evil”. https://t.co/FECYtqc4d8
— Beautyon (@Beautyon_) July 3, 2021
Copilot can’t run without the weights of the neural net and those weights cannot exist without the data. Is the model part of the “executable code” of Copilot? There are reasons why the answer could be yes, in which case Copilot would be GPL...
— Mark O. Riedl (@mark_riedl) June 30, 2021
Neural weights are not code themselves and do not resemble the surface form of any GPL code. GPL covers the possibility of "non-source conveyance". This would apply if neural weights are deemed a form of "object code". There are reasons to reject this line of argument, tho.
— Mark O. Riedl (@mark_riedl) June 30, 2021
IP lawyers, assemble! https://t.co/rsakdK28uo pic.twitter.com/SH9zv5JTrl
— Grady Booch (@Grady_Booch) June 30, 2021
What would be really funny
— Brian P. Hogan (@bphogan) July 2, 2021
Is if people who maintain popular repos
Started putting in wrong code on purpose.
copyright does not only cover copying and pasting; it covers derivative works. github copilot was trained on open source code and the sum total of everything it knows was drawn from that code. there is no possible interpretation of "derivative" that does not include this
— eevee (@eevee) June 30, 2021
Increasing number of developers are upset Github CoPilot is trained on Open Source code and then used to build a proprietary product.
— Dare Obasanjo (@Carnage4Life) July 4, 2021
Regardless of legality, this is similar to https://t.co/9OK968SH0v which was trained on public photos. Legal yet creepy.https://t.co/uqS7tYq0Pf pic.twitter.com/Mh6IkYctxG
I don't want to say anything but that's not the right license Mr Copilot. pic.twitter.com/hs8JRVQ7xJ
— Armin Ronacher (@mitsuhiko) July 2, 2021
Free Software advocates believed the GPL would end companies making money from proprietary software. Instead it simply enabled new types of proprietary software companies like Google & Amazon. Github Copilot is continuing that tradition.
— Dare Obasanjo (@Carnage4Life) July 2, 2021
Market incentives beat simple solutions. https://t.co/cCXYcKdVM6
Hi. I know you’re excited about copilot.
— Brian P. Hogan (@bphogan) July 2, 2021
GitHub scraped your code. And they plan to charge you for copilot after you help train it further.
It’s truly disappointing to watch people cheer at having their work and time exploited by a company worth billions.
github copilot has, by their own admission, been trained on mountains of gpl code, so i'm unclear on how it's not a form of laundering open source code into commercial works. the handwave of "it usually doesn't reproduce exact chunks" is not very satisfying pic.twitter.com/IzqtK2kGGo
— eevee (@eevee) June 30, 2021
Some thoughts about Github Copilot and copyright (I am not a lawyer).
— Mark O. Riedl (@mark_riedl) June 30, 2021
Copilot uses a version of GPT3 trained on GPL licensed code. GPL give everyone the right to copy and make derivatives. Derivatives inherit GPL.
Copilot can sometimes memorize and repeat snippets of code. pic.twitter.com/1JLwfQI65l
When it comes to digital policy, @Senficon is nearly always right.
— Max Andersson (@MaxAndersson) July 5, 2021
Placing AI-constructed works under copyright is a very bad idea. https://t.co/g7MWwtoNgv
i'm really tired of the tech industry treating neural networks like magic black boxes that spit out something completely novel, and taking free software for granted while paying out $150k salaries for writing ad delivery systems. the two have finally fused and it sucks
— eevee (@eevee) June 30, 2021
I really like this take on GitHub copilot: https://t.co/GtEzCzyZMP
— P. Oscar Boykin (@posco) July 5, 2021
GitHub Copilot has ruffled user feathers.https://t.co/7xA32nJsDu
— Mark Twomey (@Storagezilla) July 3, 2021
“I’m leaving GitHub because copilot uses my OpenSource code for training” is such an odd move. Anyone can fork it to there and GitHub can feed OpenSource code from anywhere to it and US copyright law permits this. I’m also pretty certain we should not strengthen copyright laws …
— Armin Ronacher (@mitsuhiko) July 3, 2021
More and more, well-established open-source developers are abandoning Github. This time it is about ML-powered GitHub Copilot product. https://t.co/rghTt0eUmK What do you think? pic.twitter.com/8BfgM0r7vh
— nixCraft (@nixcraft) July 3, 2021
Open Source means everyone can see, share, and use your code even if you disagree with them.
— Egee (@egee_irl) July 3, 2021
If you want to pick and choose who can use your code, don't fucking Open Source it. https://t.co/ITx80KeOZJ
Increasing number of developers are upset Github CoPilot is trained on Open Source code and then used to build a proprietary product.
— Dare Obasanjo (@Carnage4Life) July 4, 2021
Regardless of legality, this is similar to https://t.co/9OK968SH0v which was trained on public photos. Legal yet creepy.https://t.co/uqS7tYq0Pf pic.twitter.com/Mh6IkYctxG
「GitHubが著作権のあるコードを許可なく機械学習の訓練に使用し、ユーザの知らないうちに生成されたコードに勝手に利用することに反対します。これにともない、私は今後GitHubの利用を中止します」
— 新山祐介 (Yusuke Shinyama) (@mootastic) July 4, 2021
Copilotは昔からの「AIの生成物は著作物なのか」問題を再燃させてしまった。https://t.co/lLFNSjrzk4
まあ、こうなるわな。さてどうなるか。「Copilot の素材に使われるのが嫌なので Github の利用を止めます」https://t.co/xUqFMgDMzC
— Youhei SASAKI (@uwabami) July 5, 2021
Abandoning GitHubhttps://t.co/MkUt3ittT0
— fumi (@fumi_maker) July 5, 2021
激おこだ
When it comes to digital policy, @Senficon is nearly always right.
— Max Andersson (@MaxAndersson) July 5, 2021
Placing AI-constructed works under copyright is a very bad idea. https://t.co/g7MWwtoNgv
Important observations by @Senficon on the discussion about @github's #copilot feature and the intersection between #copyright, #AI and #open licenses. Expanding scope of © cannot be the solution if we want to protect the #publicdomainhttps://t.co/ohuUTgiT6I
— OpenFuture (@OpenFutureEU) July 5, 2021
Thanks for the great interest in my #GithubCopilot piece! Clearly, this is a complex issue with many facets. Sorry for not responding to a lot of comments yet, as I've got a lot going on at work. I missed this piece by @technollama, which is excellent: https://t.co/aoTlVjNYev
— Julia Reda (@Senficon) July 5, 2021
Is GitHub’s Copilot potentially infringing copyright? https://t.co/eSXHIx1OFR
— Andres Guadamuz (@technollama) June 30, 2021
This article https://t.co/nUXhDQa4zp is very well written and has strong arguments for why this is all technically legal. For me? This isn’t about being technically legal. I feel the same about Amazon ripping off open source projects. MS and GitHub are no better.
— Chris of Elixir, the 4th of his name (@ChrisKeathley) June 30, 2021
Here's a full response: https://t.co/eSXHIx1OFR
— Andres Guadamuz (@technollama) June 30, 2021
Abandoning GitHub https://t.co/JSQwc57WLX
— Rian Hunter (@cejetvole) July 3, 2021
EUの専門家による解釈: 「Copilotが生成するコードは派生物ではなく、GitHub (Microsoft) がGPLなコードを訓練データとして使用してもライセンス違反にはならない」
— 新山祐介 (Yusuke Shinyama) (@mootastic) July 6, 2021
もしCopilotの出力が派生物だとすると、AIが生成したデータはみんな著作物という扱いになってしまう。https://t.co/1Sqg7gkx6U
Fascinating read on the topic of GitHub Copilot and copyrights. This point seems crucial yet discutable in the current implementation: “The short code snippets that Copilot reproduces from training data are unlikely to reach the threshold of originality”. https://t.co/Uo4riNSkj5
— Arnaud Porterie (@arnaudporterie) July 5, 2021
This is the most cogent take I've seen thus far...https://t.co/tB2l2kQYs9 pic.twitter.com/b4ixf1TUF2
— Sam Kottler (@samkottler) July 5, 2021
Unfortunately, the other one doesn’t seem to behttps://t.co/aVTm5Ygghe
— payloadartist (@payloadartist) July 5, 2021