Misplaced Pages

Talk:15.ai: Difference between revisions

Article snapshot taken from Wikipedia with creative commons attribution-sharealike license. Give it a read and then ask your questions in the chat. We can research this topic together.
Browse history interactively← Previous editContent deleted Content addedVisualWikitext
Revision as of 09:48, 20 December 2024 editAlalch E. (talk | contribs)Extended confirmed users, New page reviewers, Rollbackers29,920 edits OneClickArchived "Neutrality and COI cleanup" to Talk:15.ai/Archive 1← Previous edit Latest revision as of 14:01, 20 December 2024 edit undoSineBot (talk | contribs)Bots2,555,313 editsm Signing comment by GregariousMadness - "Reliable sources: " 
(19 intermediate revisions by 2 users not shown)
Line 1: Line 1:
{{Talk header | search= yes}} {{Talk header | search= yes}}
{{Old AfD multi |date=16 January 2023 |result='''keep''' |page=15.ai |date2=2 December 2024 |result2='''Delete''' |page2=15.ai (2nd nomination)}} {{Old AfD multi |date=16 January 2023 |result='''keep''' |page=15.ai |date2=2 December 2024 |result2='''Delete''' |page2=15.ai (2nd nomination)}}
{{Not a forum}}
{{DYK talk|9 July|2022|entry=... that the developer of ''']''' claims that as little as 15 seconds of a person's voice is sufficient to clone it up to human standards using ]?|nompage=Template:Did you know nominations/15.ai}} {{DYK talk|9 July|2022|entry=... that the developer of ''']''' claims that as little as 15 seconds of a person's voice is sufficient to clone it up to human standards using ]?|nompage=Template:Did you know nominations/15.ai}}


Line 96: Line 95:
::::Oh wait, there already is one. ] ] (]) 04:08, 15 November 2024 (UTC) ::::Oh wait, there already is one. ] ] (]) 04:08, 15 November 2024 (UTC)
{{Archive bottom}} {{Archive bottom}}
== Troy Baker / Voiceverse NFT plagiarism scandal section ==

This entire section appears to be about some twitter drama that is offtopic on this article. How is the fact some non-notable company plagiarized some voice lines from work that is probably already infringing on copyright worth mentioning on this article? ] (]) 04:16, 24 November 2024 (UTC)

:How do you mean? The event wasn’t mere twitter drama and has been documented extensively by independent outlets and primary sources. It clearly meets ]. ] (]) 20:22, 24 November 2024 (UTC)
::GNG is about the notability of topics for articles. That boring twitter drama is only tangentially related to 15.ai. It might be an important part of a potential article about Voiceverse NFT but that company is nonnotable. ] (]) 20:30, 24 November 2024 (UTC)
:::Strongly disagree. Not only are ] and ] for the event met, but 15.ai is the subject of plagiarism that is well documented. It isn’t mere tangential relation if the subject is the one getting plagiarized. ] (]) 20:54, 24 November 2024 (UTC)
::::Neither WP:GNG nor WP:EVENT is met for the event, and if that would be the case you could write an article about it. The story is about the fact that ''someone else'' potentially infringed on copyright. It is not about 15.ai, which ''probably also'' infringed on copyright. ] (]) 20:56, 24 November 2024 (UTC)
:::::Then by that logic, the section isn’t relevant on ]’s article either because of his mere relation to the scandal while ''someone else'' was doing the plagiarism, but I don’t think anyone would argue that. ] (]) 21:01, 24 November 2024 (UTC)
:::::There are many, many articles written about the event, so that’s just untrue. ] (]) 21:02, 24 November 2024 (UTC)
::::::Indeed, it should also be removed from the Troy Baker article (or at the very least drastically cut down). Troy, as a business partner, is of course more involved than 15.ai. But not much more. ] (]) 21:04, 24 November 2024 (UTC)
::::::Just because there are articles written about the Voiceverse event doesn't necessairly mean it belongs in the 15.ai Misplaced Pages article. See ], ] and ]. All of the many sources that detailed the Troy Baker/Voiceverse scandal occurred in a 3 day range in January. <b>]</b> 00:05, 25 November 2024 (UTC)
:::::::<s>The sources actually span a much longer period than just 3 days in January. Looking at the citations in the article, coverage continued through at least January 31st 2022 when Baker discontinued his partnership (covered by The Verge, Tweaktown, and others). See: (January 18), (January 31).</s>
:::::::<s>From what I can tell, this wasn't a sensationalized news cycle, as it was covered by a variety of legitimate sources. In addition, it's appropriate for inclusion in this article because it represented a significant moment in 15.ai's history that demonstrated real-world consequences of AI voice technology misuse. The incident contributed to ongoing discussions about voice actor rights, proper attribution of AI-generated content, and commercialization of voice synthesis technology. These are lasting impacts that help readers understand the broader context and implications of 15.ai's development.</s>
:::::::<s>That said, if there are concerns about the section's length or detail, consider condensing it while retaining the key points about the plagiarism incident and its significance instead of removing it outright. ] (]) 04:07, 25 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
::::::::Right, and the source from January 31st is used only to support the statement {{tq|Two weeks later, Baker discontinued his partnership with Voiceverse}} and doesn't mention 15.ai at all. {{tq|it was covered by a variety of legitimate}}. ] does quite literally state {{tq|Even in respected media, a 24-hour news cycle and other pressures inherent in the journalism industry can lead to infotainment and churnalism without proper fact checking, and they may engage in frivolous "silly season" reporting}}, being covered by seemingly reliable outlets doesn't make it inherently not sensationalist. Its relevance to 15.ai is tangential at best. See ] which reads {{tq|If reliable sources – other than the critics themselves – provide substantial coverage devoted to the controversies or criticisms, then sections and subarticles about them may be justified, but only within the limitations of WP:BLPGROUPS}} which indicates the controversy should be discussed at Troy Baker or Voicesense, not necessairly at considerable undue length at an article about 15.ai <b>]</b> 04:30, 25 November 2024 (UTC)
:::::::::<s>While you're right that the January 31st source focuses on Baker's withdrawal, the earlier coverage (like the NME article) shows that 15.ai wasn't tangential to this controversy - it was central to it as the victim of direct plagiarism. The initial reporting specifically focused on how 15.ai's technology was misappropriated, including technical details about voice line manipulation and verification through 15.ai's log files.</s>
:::::::::<s>Your citation of WP:CSECTION actually supports including this content in the 15.ai article. The guideline states that sections about controversies are justified when reliable sources provide substantial coverage devoted to the controversies. We have exactly that - reliable sources documenting how 15.ai's technology was specifically targeted and misused (NME is considered a reliable source as per ). This wasn't just about Baker or Voiceverse - it was about the unauthorized exploitation of 15.ai's platform.</s>
:::::::::<s>WP:CORG doesn't suggest we should only discuss this at Troy Baker or Voiceverse's articles - rather, it supports discussing it where relevant to the subjects involved. Since 15.ai was directly impacted as the plagiarized party, this controversy is relevant to its article. The fact that it's also relevant to Baker's article doesn't diminish its significance to 15.ai's history.</s>
:::::::::<s>I agree we could potentially condense the section, but suggesting it should only appear in other articles overlooks how central 15.ai was to this controversy. As a new editor I cannot directly edit the article, so I'd appreciate it if someone could make these suggestions on my behalf. Best regards. ] (]) 04:54, 25 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
::::::::::NME is considered reliable within its area of expertise, which is music. . My citation of CSECTION does not support its inclusion because the key is {{tq|substantial coverage}}. See ] which says news coverage of a {{tq|a '''prolonged''' controversy}} (emphasis mine) is considered substantial. Again, we have coverage over the course of 3 days and then articles 2 weeks later announcing that Troy Baker is dropping them. That is not evidence of a prolonged controversy by any stretch of the imagination. It was, as Polygnotus summarized, a controversy where someone who used plaigarized datasets that were furnished by 4chan had their work plaigarized by some NFT company. Were it not for the fact that Troy Baker was associated with the company, it likely wouldn't have even made it to the news that it did receive. <b>]</b> 06:29, 25 November 2024 (UTC)
:::::::::::<s>The noticeboard you posted states that NME is generally reliable as a source, not only in music. Besides, if we demanded absolute encyclopedia-like reliability and reputation for every single source and months of sustained coverage for every event and dismissed publications whenever they report outside a narrowly-defined expertise, we'd make Misplaced Pages nearly impossible to write and maintain. Many significant events have relatively brief news cycles but still merit inclusion for their broader impact and historical significance. It is clearly, at the very least, mentionable in the article, if not deserving of its own section in the article. ] (]) 15:58, 25 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
::::::::::::I have to agree with Brocade here. This was a flash-in-the-pan, and it only got any traction because Troy Baker was mentioned. It's an issue of ] here, adding this to the article is overstating its importance to 15.ai as a topic. — <b>]:<sup>]</sup></b> 17:37, 25 November 2024 (UTC)
:::::::::::::<s>Even so (I'm still of the opinion that the event warrants its own section, as an inclusionist Wikipedian myself), the existence of reliable sources connecting the event with 15.ai makes it ''at the very least'' mentionable in the 15.ai article. Maybe a fair compromise would be to have a redirect/main article template pointing to the ] article instead, and have much of the contents be in that article instead. ] (]) 17:53, 25 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
::::::::::::::Again, ] means no, it's really not mentionable. Not everything that happens should be included. — <b>]:<sup>]</sup></b> 18:15, 25 November 2024 (UTC)
:::::::::::::::<s>WP:DUE is actually in favor of including it as a mention, contrary to your claim. To quote, " fairly represent all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint in those sources.</s>
:::::::::::::::<s>All due respect, but your philosophy is one of many valid contrasting encyclopedia standards listed on Wikimedia ( and , for example). I see that you adopt an exclusionist stance on Misplaced Pages as is written on your user page, which is totally valid. I myself am more leaning towards the inclusionist side, but even so I've offered a potential compromise that would address both sides of the aisle. But what I don't agree with is asserting one side's opinion as the sole reigning truth - I'd wager there are many Wikipedians who would agree that the event deserves at least a mention on the article, especially with the existence of reliable sources headlining it. ] (]) 18:53, 25 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
::::::::::::::::{{ec}} At some point, when a bunch of people disagree with you, it is time to move on. {{tq|WP:DUE is actually in favor of including it as a mention}} That is incorrect. {{tq|I'd wager there are many Wikipedians who would agree that the event deserves at least a mention on the article}} I doubt it, so far it has been ] but if a ] forms that it should be included then it surely will. ] (]) 19:12, 25 November 2024 (UTC)
:::::::::::::::::Please don't close a discussion so soon. I have limited free time as a grad student, and I came back to this discussion just now. To explicitly state, I disagree that the section should be removed, as I thought the section was fine as it originally was before your edits. ] (]) 19:15, 25 November 2024 (UTC)
:::::::::::::::::<s>What "bunch of people"? Are you an admin? I see three people disagreeing with me, and three people (including myself) agreeing with me. What consensus is there? ] (]) 19:20, 25 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
::::::::::::::::You ignored {{tq|significant}}. There is nothing significant about an event that barely got reported, and ONLY got reported because of one actor's name being involved. — <b>]:<sup>]</sup></b> 19:49, 25 November 2024 (UTC)
:::::::::::::::::There's a number of articles that was reported on this very controversy and the fact that the voices were plagiarized from 15.ai. Plenty of other events on Misplaced Pages have sections on articles with a comparatively negligible number of sources, but I find it doubtful that an event like this that was on the front page of a number of tech outlets isn't as significant. To copy paste from what I posted before, see: . These are more than enough to establish notability. ] (]) 20:09, 25 November 2024 (UTC)
::::::::::::::::::Again, it was a flash-in-the-pan. Tons of news sites will pick up a minor thing to get clicks, that does not make the event significant as far as Misplaced Pages is concerned. — <b>]:<sup>]</sup></b> 21:05, 25 November 2024 (UTC)
:Ok, what even happened here while I was gone for the weekend? (o_O)
:I've read through it and think both sides make fair points - yes, it was a brief controversy that mainly blew up because of Baker's involvement, but it did directly involve 15.ai's tech being stolen and raise some important questions about AI voice synthesis ethics - plus, this controversy wasn't just a couple days long (it was at least several weeks long if I remember correctly), and there are notable sources that do specifically state that 15.ai's tech was stolen. Let's also remember ]: "Notability is not temporary; once a topic has been the subject of "significant coverage" in accordance with the general notability guideline, it does not need to have ongoing coverage."

:Maybe we don't need a whole section on it, but removing it entirely seems like overkill. Could just mention it briefly in a broader section about impact/reception, possibly even within the "reaction from voice actors" section. That way we keep the historically relevant bits without giving Twitter drama too much weight. I like ]'s suggestion. I'll go ahead and edit the article to reflect that. ] (]) 19:09, 25 November 2024 (UTC)
::{{ping|GregariousMadness}} {{tq|it did directly involve 15.ai's tech being stolen}} That didn't happen. {{tq|raise some important questions about AI voice synthesis ethics}} That is incorrect, the important question raised is by 15.ai's use of copyrighted material by others, not by that NFT companies marketing person passing of a sample produced by 15.ai's work as someone elses. {{tq|we keep the historically relevant bits}} there are none. ] (]) 19:14, 25 November 2024 (UTC)
:::{{tq|That didn't happen.}} The NFT company used a non-commercial service for commercial purposes and passed it off as their own. If that isn't stealing, then I don't know what is.
:::{{tq|That is incorrect, the important questions raised is by 15.ai's use of copyrighted material by others, not by that NFT companies marketing person passing of a sample produced by 15.ai's work as someone elses.}} Not in this particular event.
:::{{tq|there are none.}} I disagree. ] (]) 19:18, 25 November 2024 (UTC)
::::Generating an audiofile or two via someone elses service and pretending your service generated them is not {{tq|stealing tech}} of course. {{tq|Not in this particular event.}} Yeah in this particular event no important questions were raised and nothing notable happened. {{tq|I disagree.}} that is allowed, others disagree with you. ] (]) 19:21, 25 November 2024 (UTC)
:::::That's fine! I think it's healthy for discussion that we have disagreements. But what isn't good is your trying to force consensus through repetitive disagreement and telling people to "move on". Let's remember to be civil. ] (]) 19:24, 25 November 2024 (UTC)
::::::I don't see anything Polygnotus has done that is incivil. — <b>]:<sup>]</sup></b> 19:50, 25 November 2024 (UTC)
:::::::I was mostly referring to {{tq|At some point, when a bunch of people disagree with you, it is time to move on.}} before anyone else had the chance to chime in. It feels like an attempt to prematurely close discussion when there's plenty of things to discuss, and I found that to be weird, especially when the discussion hadn't even been 24 hours old yet... ] (]) 20:06, 25 November 2024 (UTC)
::::::::That's still not ]. — <b>]:<sup>]</sup></b> 21:04, 25 November 2024 (UTC)
:@]: As per your latest edit: {{tq|also this is probably factually incorrect (the company didnt plagiarize work, someone in the marketing department used an audiofile generated by 15.ai and pretended it was generated via their software)}} Please note the definition of "plagiarism": ''the practice of taking someone else's work or ideas and passing them off as one's own'' (according to Oxford Dictionaries). This was by definition plagiarism, and I'm asking you to take a step back a little instead of editing so haphazardly when no consensus has been reached yet. ] (]) 19:35, 25 November 2024 (UTC)
::So according to your definition this isn't plagiarism. They didn't take work or ideas, they used an online service to generate an audiofile. ] (]) 19:39, 25 November 2024 (UTC)
:::The terms of service of the free service stipulated that any outputs ''must'' be used for non-commercial purposes, a violation of the terms, taking the work of 15.ai. See, for example, . ] (]) 19:42, 25 November 2024 (UTC)
::::Breaking the ToS != plagiarism. ] (]) 19:43, 25 November 2024 (UTC)
:::::I don't what to tell you. Every outlet that reported on this uses some sort of phrase meaning "stolen" or "taken without permission" or "plagiarized", and it's not up to you to decide whether a company plagiarized something or not when pretty much every news outlet universally agrees that some sort of unethical behavior took place.
::::: ] (]) 19:53, 25 November 2024 (UTC)
::::::Breaking the ToS ''could'' be {{tq|some sort of unethical behavior}} but that isn't the same thing as plagiarism and I searched for the string "plag" in those 4 links and Firefox found no results so I am unsure why you posted them. And of course "stolen" or "taken without permission" do not have the same meaning as "plagiarized". ] (]) 21:05, 25 November 2024 (UTC)
:::::::Plagiarism, specifically, involves presenting someone else's work as their own. Voiceverse did this when announcing their partnership with Troy Baker. If we're talking pure semantics, you could argue that instead of ''plagiarism'', the term ''misrepresentation'' could be a better fit, but then why remove that point altogether? You can easily change the term to "misrepresented" and it would still be appropriate for the article. Removing the incident entirely from the article would be throwing out important factual information. ] (]) 21:17, 25 November 2024 (UTC)
::::::::If I create an AI model from scratch, then the AI model is my work. But when I then train the AI model on data generated by others (sound files/artwork/whatever) it would be unreasonable for me to claim the copyright to what the AI model created. We ''are'' talking semantics, and words matter, specifically when we are accusing people or companies of stuff. We have to be exact. I don't think the information is important to this article because it has little to do with 15.ai. It would however be relevant if we would have an article about that NFT company, but we do not and it is not notable. ] (]) 21:25, 25 November 2024 (UTC)
:::::::::I've addressed this below, but this is still plagiarism by definition, as agreed upon by multiple academic sources. For example (from my own alma mater):
:::::::::{{tq|1. Plagiarism: Copying and pasting text, images, media, etc. generated by AI software into your document without attribution counts as plagiarism as defined by Duke. Repeating or slightly modifying phrases, sentences, or passages generated by AI tools without attribution is also plagiarism. Proper scholarly procedures require that all quoted material be identified by quotation marks or indentation on the page, and the source of information and ideas, if from another, must be identified and be attributed to that source. As described in the Duke Community Standard, plagiarism is not tolerated and may result in disciplinary action. Note that using AI as a grammar correction tool does not count as plagiarism. Duke library has a website with resources on proper AI citation.}} ] (]) 21:33, 25 November 2024 (UTC)
::::::::::Duke specifically says they are using their own definition of the word, which is fine of course, but that does not mean we have to. ] (]) 21:34, 25 November 2024 (UTC)
::::::::::Regardless of anyone's definition of plagiarism, Misplaced Pages cannot say that Voiceverse admitted to plagiarism when the sources being used to verify the claim do not at all contain Voiceverse admitting to that using any combination featuring that word. The sources all refer to a tweet where Voiceverse says {{tq|The voice was indeed taken from your platform, which our marketing team used without giving proper credit. Chubbiverse team has no knowledge of this. We will make sure this never happens again}} which reads more like they're tossing the blame on someone in the marketing team than anything. <b>]</b> 02:46, 27 November 2024 (UTC)

== Disproportionate scrutiny? ==

Having roamed around other Misplaced Pages articles to edit in my spare time, it feels like the level of scrutiny being applied to this article seems disproportionate compared to many other Misplaced Pages articles. While maintaining high standards is paramount, I notice that articles with far more questionable sources often go unchallenged. The discussion around the Troy Baker/Voiceverse incident involves multiple reliable sources (both primary and secondary) documenting a significant event involving 15.ai, yet faces intense push for removal, while similar incidents documented with comparable sources remain unchallenged in other articles. We should focus on improving content rather than removal, especially when reliable sources exist. Let me know what you think! ] (]) 19:59, 25 November 2024 (UTC)
:Its not an {{tq|intense push for removal}}, its just that if you say things that aren't true Wikipedians are possibly the group most likely to correct you. ] (]) 20:59, 25 November 2024 (UTC)
::@] I'm sorry, but this isn't a valid justification for undoing the latest revision. You insist that plagiarism never took place ({{tq|also this is probably factually incorrect (the company didnt plagiarize work, someone in the marketing department used an audiofile generated by 15.ai and pretended it was generated via their software}}) despite the reliable sources listed above that literally state that Voiceverse stole work from 15.ai ({{tq|Troy Baker's Partner NFT Company Voiceverse Reportedly '''Steals''' Voice Lines From 15.ai}}, {{tq|Voiceverse NFT admits to '''taking''' voice lines from non-commercial service}}, {{tq|Troy Baker-backed NFT company admits to '''using content without permission'''}}, {{tq|Voiceverse NFT Service Reportedly Uses '''Stolen''' Technology from 15ai}}), and you continue to dismiss the side disagreeing with you. The assertion that Voiceverse plagiarized (and to repeat, Oxford Dictionaries defines "plagiarism" as {{tq|the practice of taking someone else's work or ideas and passing them off as one's own}}) 15.ai is well-documented by multiple sources, and there is no consensus in this talk page whether to keep the section.
::I'm going to have to ask you ''why you believe that no plagiarism took place and why it isn't notable when the aforementioned plagiarism is the headline in multiple articles''. In the meantime, I'll revert your edits. As I don't want to start an edit war, I'm hoping that we will come to a consensus in a civil manner. Cheers! ] (]) 21:11, 25 November 2024 (UTC)
:::{{ping|GregariousMadness}} See ]. ] (]) 21:12, 25 November 2024 (UTC)
:::{{tq|this isn't a valid justification for undoing the latest revision}} if you are referring to that comment then I agree, but I explained my position elsewhere. {{tq|You insist that plagiarism never took place}} The quote you are using shows that I use the word probably, which people who insist on something never do. {{tq|reliable sources listed above that literally state that Voiceverse stole work}} if I create an AI model and you take some of its output while ignoring my ToS, that isn't plagiarism, right? Thats just breaking the ToS. If you steal my actual work, the AI model itself, things might be different. {{tq|there is no consensus in this talk page whether to keep the section}} see ]. {{tq|I'll revert your edits}} you shouldn't because editwarring is uncool and we haven't finished discussing things yet. ] (]) 21:20, 25 November 2024 (UTC)
::::Yes, you're right that you used the word "probably", which is why I stopped myself from reverting your edit. I'm hoping that we can come to a consensus about this topic. {{tq|if I create an AI model and you take some of its output while ignoring my ToS, that isn't plagiarism, right? Thats just breaking the ToS. If you steal my actual work, the AI model itself, things might be different.}} No, that's ''both'' stealing/plagiarism ''and'' breaking the TOS. This has happened many times in academia, where students have been expelled for plagiarism simply for misattributing sources, and recent orientation curriculums ''specifically include the misattribution/stealing of generative AI outputs in their research papers as an example of '''plagiarism''''' (they warn all students of this on day 1 of orientation). {{tq|you shouldn't because editwarring is uncool and we haven't finished discussing things yet.}} You're right, and I refrained from reverting your edit. ] (]) 21:27, 25 November 2024 (UTC)
::::See, for example:
::::{{tq|1. Plagiarism: Copying and pasting text, images, media, etc. generated by AI software into your document without attribution counts as plagiarism as defined by Duke. Repeating or slightly modifying phrases, sentences, or passages generated by AI tools without attribution is also plagiarism. Proper scholarly procedures require that all quoted material be identified by quotation marks or indentation on the page, and the source of information and ideas, if from another, must be identified and be attributed to that source. As described in the Duke Community Standard, plagiarism is not tolerated and may result in disciplinary action. Note that using AI as a grammar correction tool does not count as plagiarism. Duke library has a website with resources on proper AI citation.}} ] (]) 21:29, 25 November 2024 (UTC)
:::::I do not believe that the developer of an AI model owns the copyright to all output of that AI model, and certainly not if the AI model was trained on the copyrighted works of others. It would make more sense to give the owners of the copyrights of the works the model is based on the copyright. The law is slow to change. While human authors have copyright, AI models do not at this point. The example of students expelled for plagiarism is, in my view, not comparable because the university cares about whether they've done the work themselves, not if they are infringing on the rights of others. And the university is certainly not trying to protect the hypothetical rights of an AI model. ] (]) 21:31, 25 November 2024 (UTC)
::::::While you raise some interesting points about AI copyright law (and I tend to agree with some of the points you make), these are ultimately ''your personal interpretations''. {{tq|The example of students expelled for plagiarism is, in my view, not comparable because the university cares about whether they've done the work themselves, not if they are infringing on the rights of others. And the university is certainly not trying to protect the hypothetical rights of an AI model.}} The comparison to academic plagiarism remains relevant because plagiarism and copyright are distinct concepts - plagiarism is about misrepresenting the source or creator of work, regardless of copyright status.
::::::In the Voiceverse case, the key issue isn't about who owns copyright to AI outputs. It's that they explicitly claimed content generated by 15.ai was created by their own different system. That misattribution fits the basic definition of plagiarism as stated in the above PDF: presenting someone else's work as your own.
::::::{{tq|I do not believe that the developer of an AI model owns the copyright to all output of that AI model, and certainly not if the AI model was trained on the copyrighted works of others.}} Your view that AI model developers shouldn't own copyright to outputs is a legitimate position in an ongoing debate, ''but it doesn't change the fact that falsely claiming to be the source of content is plagiarism, regardless of copyright status.'' Even if content is in the public domain or has ambiguous copyright, presenting it as your own original work is still plagiarism. ] (]) 21:37, 25 November 2024 (UTC)
:::::::We have now reached what I believe to be the crux of the issue, in order to plagiarize the "victim" needs to be the copyright owner of the work. But since the "victim" is an AI model, a glorified Markov chain, it owns nothing and it cannot posses copyright. In the same way you can't steal physical possessions owned by an AI model you also can't steal intellectual possessions owned by an AI model. See ]. I can't plagiarize the sound of the waves, or the wind rustling some leaves. Mother Nature (although personified here) has no legal personhood. Claiming that your AI model generated somehing another AI model generated is possibly not nice, but I don't think it can qualify as plagiarism. As far as I know there is no copyright law against victimless plagiarism. Universities of course do have rules against victimless plagiarism (you are not allowed to pay someone else to do the work for you and then pretend you did it). ] (]) 21:42, 25 November 2024 (UTC)
::::::::{{tq|We have now reached what I believe to be the crux of the issue, in order to plagiarize the "victim" needs to be the copyright owner of the work.}} I disagree - the "victim" isn't just the AI model, but the combined set of the AI model and the developer. You wouldn't say you plagiarized Word or Microsoft when you copy someone's text, but you would say you plagiarized the author who wrote using Word. Similarly, 15.ai is both the tool AND the service created by its developer - when Voiceverse used 15.ai's output and claimed it as their own, they were plagiarizing both the system's output and misrepresenting the work of the developers who created that system.
::::::::{{tq|But since the "victim" is an AI model, a glorified Markov chain, it owns nothing and it cannot posses copyright. In the same way you can't steal physical possessions owned by an AI model you also can't steal intellectual possessions owned by an AI model. See Personhood.}} The AI model isn't just a "glorified Markov chain" operating in isolation - it's a complete service and system developed by people who specified terms for its use. When Voiceverse claimed output from this system was generated by their own technology, they were misrepresenting both the source of the content (15.ai) and implicitly the work of the developer who created that system.
::::::::{{tq|I can't plagiarize the sound of the waves, or the wind rustling some leaves. Claiming that your AI model generated somehing another AI model generated is possibly not nice, but I don't think it can qualify as plagiarism. }} Plagiarism, ultimately, is about honest attribution of sources and work. (And this is what they tell us at orientation, just to be clear.) Just as academic plagiarism can occur with public domain works that have no current copyright holder, plagiarism can occur when misrepresenting the source of AI-generated content, regardless of the complex questions around AI copyright law and who ''technically'' owns what.
::::::::The key issue is that Voiceverse claimed someone else's system's output was created by their own different system. That's plagiarism of both the output and misrepresentation of the developer's work, plain and simple. If I were a professor (I'm still a few years away from that, sadly), I would certainly categorize this as a cut-and-dry case of plagiarism - ''who'' was the victim of this plagiarism would be irrelevant. ] (]) 21:52, 25 November 2024 (UTC)
:::::::::The law currently does not give an AI model creator the copyright over the output of the AI model. {{tq|the "victim" isn't just the AI model, but the combined set of the AI model and the developer}} not according to the law as I understand it. {{tq|Plagiarism, ultimately, is about honest attribution of sources and work.}} that is not how the law works. For example, you can freely copy text from works in the public domain whose copyrights have expired without attribution. {{tq|If I were a professor}} thank god you are not studying law! ] (]) 21:57, 25 November 2024 (UTC)
::::::::::{{tq|The law currently does not give an AI model creator the copyright over the output of the AI model.}} Plagiarism and copyright infringement are two distinct concepts that you're conflating. You're correct about copyright law, but incorrect in using that to argue there was no plagiarism. You're right that current law doesn't give AI model creators copyright over AI output; this is a correct statement about copyright law. '''However, plagiarism is NOT defined by copyright law.''' Plagiarism is about presenting someone else's work as your own, ''regardless of copyright status.'' To reference your example, '''Students must cite public domain works even though they're not copyrighted.''' Academics must acknowledge their sources even when quoting copyright-expired materials. And finally, ''presenting someone else's public domain work as your own original creation is still plagiarism''.
::::::::::{{tq|thank god you are not studying law!}} Seriously? That was uncalled for. ] (]) 22:03, 25 November 2024 (UTC)
:::::::::::{{tq|Plagiarism is not illegal in the United States in most situations. Instead it is considered a violation of honor or ethics codes and can result in disciplinary action from a person’s school or workplace. However, plagiarism can warrant legal action if it infringes upon the original author’s copyright...}} If there is no "original author", because the AI model has whitewashed copyrighted stuff from other sources and can't itself be an author, how can it be plagiarism? Not sure how you interpreted that, I was just joking about the fact that the law is confusing and weird and less fun than AI. ] (]) 22:06, 25 November 2024 (UTC)
::::::::::::{{tq|Plagiarism is not illegal in the United States in most situations. Instead it is considered a violation of honor or ethics codes and can result in disciplinary action from a person’s school or workplace. However, plagiarism can warrant legal action if it infringes upon the original author’s copyright... }} This only strengthens the argument that Voiceverse committed plagiarism, even while we debate the copyright implications. I'm glad we agree that plagiarism is an ''ethical violation'', not a legal one. Copyright infringement is a separate legal issue that may or may not accompany plagiarism; therefore, ''whether 15.ai has legal copyright over its outputs is irrelevant to whether Voiceverse committed plagiarism''. They could still have committed plagiarism (an ethical violation) even if no copyright was infringed.
::::::::::::{{tq|Not sure how you interpreted that, I was just joking about the fact that the law is confusing and weird and less fun than AI.}} Not a very funny joke. Students like me spend over a decade working to become a professor in their area of expertise, and I don't like it when someone I don't know on the Internet jokes about my credentials. ] (]) 22:13, 25 November 2024 (UTC)
:::::::::::::But I didn't joke about your or anyone else's credentials? ] (]) 22:14, 25 November 2024 (UTC)
::::::::::::::You wrote {{tq|thank god you are not studying law!}} on my mention that I am working to become a professor. You don't see how that sounds insulting? ] (]) 22:22, 25 November 2024 (UTC)
:::::::::::::::No, that does not sound insulting at all to me. Can you explain why you react this way? ] (]) 22:32, 25 November 2024 (UTC)
::::::::::::::::If you truly didn't mean any offense, then I'll consider it water under the bridge. For most of this talk page discussion, it felt like people weren't being friendly towards me (made worse because of a prior interaction on this article), so I took that statement to imply that I was so stupid in my arguments that studying the law would have been a disservice to the law. ] (]) 22:37, 25 November 2024 (UTC)
:::::::::::::::::Oh man I apologize if I come across ''that'' dickish, but that was not what I meant at all! I just think studying law is like bashing your head against a wall. If people are unfriendly what usually helps is ] and ]. And sometimes an admin dropping by to remind everyone how to behave helps. ] (]) 22:39, 25 November 2024 (UTC)
:::::::::::::::Those do not correlate. There are professors for everything under the sun, no one here had any way to know you were studying to be a law professor until you just now brought it up. — <b>]:<sup>]</sup></b> 22:24, 25 November 2024 (UTC)
::::::::::::::::I don't know what you're talking about. I'm not studying to be a law professor... I'm a machine learning grad student working to be a professor, and professors and colleges are the ones who decide whether a student has been plagiarizing or not, not the law. May I ask, why are you confusing the conversation all of a sudden? ] (]) 22:32, 25 November 2024 (UTC)
:::::::::::::::::I'm confused because you're not being clear. You're really pissed off over a law joke, when it's got absolutely nothing to do with your studies?
:::::::::::::::::Also, I'm ''really'' getting sick of your failure to ] and making accusatory statements towards others. — <b>]:<sup>]</sup></b> 23:13, 25 November 2024 (UTC)
::::::::::::::::::<s>Both parties were clear enough, @] and @] had already cleared up the confusion and made up before your unnecessarily confusing and aggressive statements. "You're really pissed off over a law joke", and "Also, I'm really getting sick of your failure to WP:AGF and making accusatory statements towards others." is not civil behavior. ] (]) 01:03, 26 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
:::::::::::::::::::Says the ]. Regardless, I'm done with this discussion. There's no consensus for adding this to the article, and you can attempt an ] if you really want to add it. — <b>]:<sup>]</sup></b> 17:19, 26 November 2024 (UTC)
::::::::::::::::::Please calm down and remember to be civil. We (Polygnotus and I) already cleared up the confusion and are now attempting to reach a compromise before you commented, so this abrasiveness is unnecessary. ] (]) 19:15, 26 November 2024 (UTC)
::::::::::::::::I had read their userpage so I knew they were ''not'' studying to become a law professor. ] (]) 22:32, 25 November 2024 (UTC)
::::::::::::::::So if we agree that there is legally nothing wrong with what they did, and it was only an "ethical violation" by someone working in marketing, then its a non-story right? Storm in a teacup. Ethical violations and marketing go hand in hand, right? ] (]) 22:36, 25 November 2024 (UTC)
:::::::::::::::::{{tq|So if we agree that there is legally nothing wrong with what they did, and it was only an "ethical violation" by someone working in marketing, then its a non-story right?}} No, absolutely not. As a personal aside, ethical violations in marketing shouldn't just be dismissed as normal or acceptable just because they're common. But on top of that, this wasn't just a minor marketing mishap/strategy. Voiceverse misrepresented their technological capabilities to potential investors/customers, used another company's technology while claiming it was their own, did this during a major partnership announcement with Troy Baker, and violated terms of service in the process. The fact that something might not be illegal doesn't make it a "non-story." Again, the notability of the story is supported by multiple sources directly implicating Voiceverse in the plagiarism of 15.ai's work, which should be the only thing that matters. ] (]) 22:42, 25 November 2024 (UTC)
::::::::::::::::::In my view those are a lot of reasons why this is important to mention on an article about Voiceverse, but not 15.ai. What impact did it have on 15.ai? Perhaps we can compromise on something much shorter and less prominent that mentions the impact of the plagiarism? It may be possible to get consensus to include something like that. ] (]) 22:46, 25 November 2024 (UTC)
:::::::::::::::::::I don't mean to be dismissive, but I have to go soon. But being the victim of plagiarism is absolutely notable for 15.ai, as the incident gained significant media attention specifically because it involved 15.ai being exploited by a larger commercial entity, making it a defining moment in its history.
:::::::::::::::::::I'll give a personal analogy: if one of my academic articles was plagiarized by a more important researcher or publisher, it would have a ''significant'' impact on myself and my "history". It would definitely be something that, if I had my own Misplaced Pages article in the future, would be mentioned as an aside. ] (]) 22:50, 25 November 2024 (UTC)
::::::::::::::::::::It is always a bit sad when David's Misplaced Pages article is defined by Goliath's actions imo. I think there was previously an undue amount of focus on something that, in my view, was not very important, but if you disagree you may have some suggestions how to incorporate it into the article in a less prominent way. ] (]) 22:53, 25 November 2024 (UTC)
:::::::::::::::::::::Thank you for the engaging conversation yesterday. Apologies for jumping the gun on your joke - I was feeling antsy because this is my first time engaging in a serious debate like this on Misplaced Pages, and I'm not very good at serious debates. I took some time to think about it myself, and as you suggested, I'll be working on an edit that addresses your concerns while not making the event a centerpiece of the article. I'll let you know once I finish my edits! ] (]) 19:14, 26 November 2024 (UTC)
:::::::::::::::::::::I've significantly trimmed down the section to only include a paragraph and a half of text. I'm thinking of trimming it down even further, but I believe it no longer puts an undue amount of focus on the event. Let me know if I should change anything further! Cheers. ] (]) 19:30, 26 November 2024 (UTC)
:::::::::::::::::::<s>I agree with a compromise similar to this - it was what I suggested earlier, so I'm happy we're coming to a consensus! We can still include the citations but condense the section a little bit. I'm sure we can find a good middle ground. ] (]) 01:14, 26 November 2024 (UTC)</s> {{sockvote|HackerKnownAs}}
:There's nothing unusual here, especially for an article that recently had multiple ] issues. As to {{tq| I notice that articles with far more questionable sources often go unchallenged}}, that's an ] argument. Feel free to bring up the problems those articles have so they get some attention. It's a volunteer project, so people pick and choose what they get involved with. Until they know there's a problem, they can't to do anything about it. — <b>]:<sup>]</sup></b> 21:13, 25 November 2024 (UTC)
::I see. I was beginning to think there was a coordinated effort to remove content from an article that I drafted up as my first Wikipedian project, but that makes more sense. I still think that the contentious section has more than enough references and relevant to keep, but I understand the scrutiny now. ] (]) 21:19, 25 November 2024 (UTC)
:::Please list or tag problems in other articles, and if you ping me I'd be happy to take a look. ] (]) 21:21, 25 November 2024 (UTC)
::::Sure. One article that I recently looked at that has a ''lot'' of unsourced text is ], but having been a participant in Olympiad mathematics, I can safely say that it is notable and much of the information there is correct. If we followed the standards exhibited in this talk page, however, I would feel like over 70% of the article would have to be removed. The 15.ai article is extremely well-sourced in comparison. ] (]) 21:54, 25 November 2024 (UTC)
:::::Thank you. I have added the {{tl|Sources}} template which adds it to ]. ] (]) 22:01, 25 November 2024 (UTC)
:{{tq|We should focus on improving content rather than removal}}
:Removing content '''''is''''' sometimes the same as improving the article. Mentions of a scandal, complete with tweets blasted across the article, does not improve the encyclopedic content of the article. It isn't a major event for 15.ai, the topic of the article, and it wasn't a prolonged controversy with sustained converage that warrants anything more than a brief mention. An entire section of the article about it is ]. Moreover, ] states {{tq|Information should not be included solely because it is true or useful. An article should not be a complete presentation of all possible details, but a summary of accepted knowledge regarding its subject}}. A brief scandal of another company using a voiceline that was generated by 15.ai has functionally very little to do with 15.ai itself. It is, essentially, a conflict between 15(the creator of 15.ai), who for the purposes of Misplaced Pages is considered unnotable, , and Voiceverse, who for the purposes of Misplaced Pages is also considered unnotable. <b>]</b> 08:25, 26 November 2024 (UTC)

== Regarding The Batch ==

I have been removing all instances of this source from the article. It was deemed unreliable when the article was a draft. They were re-added after the article was released from draftspace by a now vanished user who flagged them as a minor edit. <b>]</b> 00:02, 28 November 2024 (UTC)

:That's a great point! I do see that the source was deemed unreliable when the article was a draft, but I also see that that was back in 2021, only two years after the newsletter was founded. Nevertheless, I think that in 2024, the newsletter has a much better reputation and is now considered rather reliable among the AI/ML community, as I see prominent leaders in the AI space referring to deeplearning.ai and ''The Batch'' from time to time. I've made a post in the Noticeboard to gather others' opinions here: ] ] (]) 18:41, 28 November 2024 (UTC)
::As per the discussion in the above link, I've re-added Andrew Ng's quote to the section. Cheers! ] (]) 22:12, 28 November 2024 (UTC)
:::Cheers! <b>]</b> 18:47, 2 December 2024 (UTC)

== 15.ai finally responds == == 15.ai finally responds ==


Line 235: Line 110:
I'm posting this as a reference. According to archived discussions on the Teahouse, a good rule of thumb is three independent and reliable sources that demonstrate significant coverage to establish ]. Here are the major reliable sources that provide significant coverage for the subject: I'm posting this as a reference. According to archived discussions on the Teahouse, a good rule of thumb is three independent and reliable sources that demonstrate significant coverage to establish ]. Here are the major reliable sources that provide significant coverage for the subject:


*'']'' . Reliable as one of the largest and oldest-running newspapers in Taiwan. Listed as one of the three major Chinese-language newspapers in ]. Significant coverage includes an overview of the technology behind 15.ai, particularly noting its ease of use and limited data, and also discusses how 15.ai works, its features, and the viral videos that have spawned using 15.ai. Over 400 (approximate since the article is written in Chinese) words of coverage. *'''''' '']'' . Reliable as one of the largest and oldest-running newspapers in Taiwan. Listed as one of the three major Chinese-language newspapers in ]. Significant coverage includes an overview of the technology behind 15.ai, particularly noting its ease of use and limited data, and also discusses how 15.ai works, its features, and the viral videos that have spawned using 15.ai. Over 400 (approximate since the article is written in Chinese) words of coverage.
*'']'' . Reliable as listed in ]. Significant coverage includes an overview of the DeepMoji technology used for emotiveness, applications of the voices not restricted to viral videos, and how to use it. Over 400 (approximate since the article is written in Japanese) words of coverage. *'']'' . Reliable as listed in ]. Significant coverage includes an overview of the DeepMoji technology used for emotiveness, applications of the voices not restricted to viral videos, and how to use it. Over 400 (approximate since the article is written in Japanese) words of coverage.
*'']'' . While not listed in ], ''AUTOMATON'' is one of the largest and reputable gaming news outlets in Japan, and has been used in multiple GA's like '']'', '']'', and ]. Significant coverage includes DeepMoji, a list of characters available on the application, examples of video content users have created with the platform, an overview of the pronunciation capabilities of the model, as well as a mention of how to use ARPAbet strings. Almost 800 (approximate since the article is written in Japanese) words of coverage. *'']'' . While not listed in ], ''AUTOMATON'' is one of the largest and reputable gaming news outlets in Japan, and has been used in multiple GA's like '']'', '']'', and ]. Significant coverage includes DeepMoji, a list of characters available on the application, examples of video content users have created with the platform, an overview of the pronunciation capabilities of the model, as well as a mention of how to use ARPAbet strings. Almost 800 (approximate since the article is written in Japanese) words of coverage.
*''Towards Data Science'' . While the website itself is written on Medium (which is not considered reliable), Medium is only being used as a vessel to host the article itself (similar to how Google Docs can be used to host an article), which is part of a very popular newsletter called ''Towards Data Science'', which has almost 800K followers on social media. Following alone means nothing in determining the reliability of a source, but Rionaldi Chandraseta, the author of the article, is an ]-published machine learning specialist who has published papers that are listed on Google Scholar . The newsletter has a solid editorial board that consists of multiple masters and PhD's in machine learning and computer science. '''Over 1,000''' words of English-language coverage detailing every facet of 15.ai, from its capabilities to its underlying research. *'''''' Rionaldi Chandraseta . While the article itself is written on Medium (which is not considered reliable), Medium is only being used as a vessel to host the article itself (similar to how Google Docs can be used to host an article), which is part of a very popular newsletter called ''Towards Data Science'', which has almost 800K followers on social media. Following alone means nothing in determining the reliability of a source, but Rionaldi Chandraseta, the author of the article, is an ]-published machine learning specialist who has published papers that are listed on Google Scholar . The newsletter has a solid editorial board that consists of multiple masters and PhD's in machine learning and computer science. '''Over 1,000''' words of English-language coverage detailing every facet of 15.ai, from its capabilities to its underlying research.
*Yongqiang Li . Since the article is locked to foreigners without an account, I asked a friend to translate this for me. The article goes into great detail about the technology behind 15.ai and talks about its features, its future, and potential problems. The author is a professor at the Harbin Institute of Technology and has multiple publications listed on Google Scholar . *'''''' Yongqiang Li . Since the article is locked to foreigners without an account, I asked a friend to translate this for me. The article goes into great detail about the technology behind 15.ai and talks about its features, its future, and potential problems. The author is a professor at the ] and has multiple publications listed on Google Scholar and ResearchGate .
*'']'' . Reliable as listed in ]. While the main focus of the article isn't 15.ai, it goes into detail the controversy and Twitter exchange that happened when Voiceverse NFT misappropriated 15.ai's work. From {{tq|However, in now-deleted tweets, Voiceverse was found to have boasted about using its tech for the voice of a cartoon character - which was in fact created using 15.ai, a popular non-commercial text-to-speech service.}} to {{tq|"Hey @fifteenai we are extremely sorry about this," Voiceverse NFT wrote. "The voice was indeed taken from your platform, which our marketing team used without giving proper credit. Chubbiverse team has no knowledge of this. We will make sure this never happens again."}}, this is about 300 words of coverage. *'']'' . Reliable as listed in ]. While the main focus of the article isn't 15.ai, it goes into detail the controversy and Twitter exchange that happened when Voiceverse NFT misappropriated 15.ai's work. From {{tq|However, in now-deleted tweets, Voiceverse was found to have boasted about using its tech for the voice of a cartoon character - which was in fact created using 15.ai, a popular non-commercial text-to-speech service.}} to {{tq|"Hey @fifteenai we are extremely sorry about this," Voiceverse NFT wrote. "The voice was indeed taken from your platform, which our marketing team used without giving proper credit. Chubbiverse team has no knowledge of this. We will make sure this never happens again."}}, this is about 300 words of coverage.
*''Stevivor'' . After doing more research, I found that Steven Wright, the author of this article, ''also'' writes for '']'', a solid and well-known technology and gaming publication. In addition, ''Stevivor'' is reliable and independent, and it is the most-read independent gaming news network in the Oceanic region. *''Stevivor'' . After doing more research, I found that Steven Wright, the author of this article, ''also'' writes for '']'', a solid and well-known technology and gaming publication. In addition, ''Stevivor'' is reliable and independent, and it is the most-read independent gaming news network in the Oceanic region.
*'']'' . While ''Kotaku'' is in ], it's still debated whether an article from the "Odds and Ends" category is considered "News". There is no clear consensus to this, but the description for the "Odds and Ends" category is "Odds and Ends - Gaming Reviews, News, Tips and More.". The argument in the previous AfD was that this article did not meet reliability, although it met independence and significant coverage. *'']'' . While ''Kotaku'' is in ], it also states {{tq|News posts from Kotaku between 2010 and 2022 are considered reliable,}} (the article is from 2021, so it meets this criteria) but also states {{tq|although editors are cautioned of blog/geeky posts that have little news or reporting significance}. It's still debated whether an article from the "Odds and Ends" category is considered "News", and the entry in ] says {{tq|articles should be evaluated on a case-by-case basis.}} There is no clear consensus to this, but the description for the "Odds and Ends" category is "Odds and Ends - Gaming Reviews, News, Tips and More.". The argument in the previous AfD was that this article did not meet reliability, although it met independence and significant coverage. I personally believe that this article is reliable.
*'']'', '']'', and '']'' All three of these sources are found under ], but there has been a debate whether these three met significant coverage. While they all pass ], it is not a Misplaced Pages policy and their significant coverage can be debated. *'']'', '']'', and '']'' All three of these sources are found under ], but there has been a debate whether these three met significant coverage. While they all pass ], it is not a Misplaced Pages policy and their significant coverage can be debated.
*'']'' . ] notes ''NME'' is reliable in its expertise, and it has been debated whether gaming is one of their areas of expertise. The Misplaced Pages article for NME states that this is so, and gaming is listed as one of NME's header sections, but there has been debate whether NME's expertise extends outside of music. Similar coverage to ''Eurogamer'', but with fewer words, but still above the threshhold for ] (which, again, is not Misplaced Pages policy). *'']'' . ] notes ''NME'' is reliable in its expertise, and it has been debated whether gaming is one of their areas of expertise. The Misplaced Pages article for NME states that this is so, and gaming is listed as one of NME's header sections, but there has been debate whether NME's expertise extends outside of music. Similar coverage to ''Eurogamer'', but with fewer words, but still above the threshhold for ] (which, again, is not Misplaced Pages policy).
*] . The author, Andrew Ng, is one of the most famous and influential artificial intelligence researchers in the world, with a healthy Google Scholar profile and was included in the Time 100 Most Influential People in AI list in 2023. While 15.ai is mentioned as a blurb and likely does not meet significant coverage, it shows that the subject wasn't a mere curiosity and was under the radar for a large number of prominent figures in AI while the service was active. <!-- Template:Unsigned --><small class="autosigned">—&nbsp;Preceding ] comment added by ] (] • ]) 14:00, 20 December 2024 (UTC)</small> <!--Autosigned by SineBot-->


However, some of these have been contested in ''some form'', the arguments for which I personally disagreed with. Think of these sources as you will. I still haven't gone over many of the the sources that are used in the newly written page, but I will continue to do my research and update this. However, some of these have been contested in ''some form'', the arguments for which I personally disagreed with. Think of these sources as you will. I still haven't gone over many of the the sources that are used in the newly written page, but I will continue to do my research and update this.
Line 252: Line 128:
:These sources look largely reliable to me. It was a large part of why I accepted the original draft in the first place, since it's rare to see a draft with such good sourcing. I personally believe this meets notability, especially since it's been clarified outright this is much expanded from the deleted article. ''] Considerer:'' ''']''' (]) (]) 18:45, 19 December 2024 (UTC) :These sources look largely reliable to me. It was a large part of why I accepted the original draft in the first place, since it's rare to see a draft with such good sourcing. I personally believe this meets notability, especially since it's been clarified outright this is much expanded from the deleted article. ''] Considerer:'' ''']''' (]) (]) 18:45, 19 December 2024 (UTC)
::Thank you. Sorry for all the trouble. I never meant for all this to happen. I can get carried away at times and I felt a fire light up inside me when I was writing the new article. I really hope this version gets to stay. ] (]) 21:42, 19 December 2024 (UTC) ::Thank you. Sorry for all the trouble. I never meant for all this to happen. I can get carried away at times and I felt a fire light up inside me when I was writing the new article. I really hope this version gets to stay. ] (]) 21:42, 19 December 2024 (UTC)

== Please fill out the missing stuff in the Japanese source citations ==

@]: Please see the example for a foreign-language reference below. What's needed are the trans-title, language, quote, and trans-quote params
<pre>
{{cite news
|last=Vučković
|first=Branko
|date=28 September 2013
|title=Železnice Srbije u sve lošijem stanju
|trans-title=Serbian Rail in Increasingly Worse Condition
|url=https://www.slobodnaevropa.org/a/zeleznice-srbije-u-sve-losijem-stanju/25120244.html
|access-date=5 November 2024
|website=Radio Slobodna Evropa |language=sh
|quote=Koliko je železnica u Srbiji zaostala, najbolje se vidi iz poređenja sa Evropskom unijom, gde se vozovi kreću prosečnom brzinom između 200 i 300 kilometara na sat, dok je prosečna brzina na domaćim prugama 44 kilometra, samo nekoliko kilometara brže od prvog voza koji je saobraćao na tek sagrađenoj pruzi Beograd - Niš u septembru 1884. godine.
|trans-quote=The extent to which the railway system in Serbia has fallen behind is best illustrated by comparing it with the European Union, where trains travel at an average speed of between 200 and 300 kilometers per hour, while the average speed on domestic tracks is just 44 kilometers, only a few kilometers faster than the first train that operated on the newly constructed Belgrade–Niš line in September 1884.}}
</pre> —] 12:41, 20 December 2024 (UTC)

:Done, thank you! ] (]) 13:05, 20 December 2024 (UTC)
::Thanks a lot. Having translated titles is really important, but it would also be very nice and ''genuinely helps verifiability'' to have translated relevant quotes. If it needs to be multiple quotes from the same source, cite the same source multiple times each time with a different suitable quote and I might convert the citation method to ]. If you determine that any of the Asian sources aren't essential, and the statements are already made verifiable using English-language sources, please remove them. —] 13:10, 20 December 2024 (UTC)
:::Oh... I just saw this message after I submitted an edit translating all of the other sources. What should I do? Should I revert my edit? ] (]) 13:18, 20 December 2024 (UTC)
::::No worries, please see ]. Basically, there's no need to use {{tl|efn}} for translations of quotes. —] 13:39, 20 December 2024 (UTC)
::::In other words, what Japanese text in the reference no. 2 (as of ]) supports the statement "15.ai was conceived as a research project by a developer known as "15" during their undergraduate studies at the Massachusetts Institute of Technology, and was later implemented following the developer's successful exit from a startup venture." and what is the English translation of that Japanese text. —] 13:42, 20 December 2024 (UTC)
:::::I see, I'll get to that in a bit. I'm expanding some other sections but I'll definitely make those edits soon. ] (]) 13:49, 20 December 2024 (UTC)

Latest revision as of 14:01, 20 December 2024

This is the talk page for discussing improvements to the 15.ai article.
This is not a forum for general discussion of the article's subject.
Article policies
Find sources: Google (books · news · scholar · free images · WP refs· FENS · JSTOR · TWL
Archives: 1Auto-archiving period: 30 days 
Articles for deletionThis article was nominated for deletion. Please review the prior discussions if you are considering re-nomination:
A fact from 15.ai appeared on Misplaced Pages's Main Page in the Did you know column on 9 July 2022 (check views). The text of the entry was as follows:
  • Did you know... that the developer of 15.ai claims that as little as 15 seconds of a person's voice is sufficient to clone it up to human standards using artificial intelligence?
A record of the entry may be seen at Misplaced Pages:Recent additions/2022/July. The nomination discussion and review may be seen at Template:Did you know nominations/15.ai.
Misplaced Pages
Former good article15.ai was one of the Engineering and technology good articles, but it has been removed from the list. There are suggestions below for improving the article to meet the good article criteria. Once these issues have been addressed, the article can be renominated. Editors may also seek a reassessment of the decision if they believe there was a mistake.
Article milestones
DateProcessResult
June 11, 2022Good article nomineeListed
November 18, 2024Good article reassessmentDelisted
Current status: Delisted good article
This article has not yet been rated on Misplaced Pages's content assessment scale.
It is of interest to the following WikiProjects:
WikiProject iconArticles for creation
WikiProject iconThis article was reviewed by member(s) of WikiProject Articles for creation. The project works to allow users to contribute quality articles and media files to the encyclopedia and track their progress as they are developed. To participate, please visit the project page for more information.Articles for creationWikipedia:WikiProject Articles for creationTemplate:WikiProject Articles for creationAfC
Note icon
This article was accepted from this draft on 19 December 2024 by reviewer Pokelego999 (talk · contribs).
WikiProject iconInternet
WikiProject iconThis article is within the scope of WikiProject Internet, a collaborative effort to improve the coverage of the Internet on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.InternetWikipedia:WikiProject InternetTemplate:WikiProject InternetInternet
???This article has not yet received a rating on the project's importance scale.
WikiProject iconInternet culture
WikiProject iconThis article is within the scope of WikiProject Internet culture, a collaborative effort to improve the coverage of internet culture on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Internet cultureWikipedia:WikiProject Internet cultureTemplate:WikiProject Internet cultureInternet culture
???This article has not yet received a rating on the project's importance scale.
WikiProject Internet culture To-do:

Here are some tasks awaiting attention:
WikiProject iconTechnology
WikiProject iconThis article is within the scope of WikiProject Technology, a collaborative effort to improve the coverage of technology on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.TechnologyWikipedia:WikiProject TechnologyTemplate:WikiProject TechnologyTechnology
WikiProject iconMy Little Pony
WikiProject iconThis article is within the scope of WikiProject My Little Pony, a collaborative effort to improve the coverage of My Little Pony related articles on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.My Little PonyWikipedia:WikiProject My Little PonyTemplate:WikiProject My Little PonyMy Little Pony
???This article has not yet received a rating on the importance scale.
WikiProject iconArtificial Intelligence
WikiProject iconThis article is within the scope of WikiProject Artificial Intelligence, a collaborative effort to improve the coverage of Artificial intelligence on Misplaced Pages. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Artificial IntelligenceWikipedia:WikiProject Artificial IntelligenceTemplate:WikiProject Artificial IntelligenceArtificial Intelligence

RFC on Status of Web Site

WP:SNOW CLOSE After a month, an overwhelming majority voted for Option C to omit the information from the infobox. While I am involved, there is a strong, clear consensus to omit from the infobox. Brocade River Poems (She/They) 17:59, 4 December 2024 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.


How should the current status of the 15.ai web site be listed in the infobox?

  • A. Under maintenance.
  • B. Abandoned.
  • C. Omit the Current Status field from the infobox.

Robert McClenon (talk) 01:46, 4 November 2024 (UTC)


Please enter A, B, or C, with a brief statement in the Survey. Please not reply to other editors in the Survey. That is what the Discussion section is for.

Survey

I'd say something like "under maintenance since <date>" to avoid original research. Aaron Liu (talk) 02:45, 4 November 2024 (UTC)
(invited by the bot) Leave it out. WP:Ver requires sourcability for whatever is put in there and there is no source in the article for any such characterization. Sincerely, North8000 (talk) 20:35, 4 November 2024 (UTC)
However, there is a source for the website being down for maintenance since 2022.
Since this was clearly from DRN, I wonder what its participants have to say and am surprised they have not commented yet. Aaron Liu (talk) 21:36, 4 November 2024 (UTC)
There is a source? Where? 2400:79E0:8041:4880:1804:EAEA:346E:9670 (talk) 00:59, 5 November 2024 (UTC)
This throws a wrench in things... the inline citation confused me into thinking it cited the entire sentence. I'm now not sure what we should do. Aaron Liu (talk) 12:17, 7 November 2024 (UTC)
I think we can use https://archive.ph/sk2VL as a source. Aaron Liu (talk) 12:20, 7 November 2024 (UTC)
Regarding DRN. Two of the participants at the DRN have said the conversation went on too long and noted they did not wish to participate in the RfC or continued dispute, one of the editors was indef blocked for a different issue, one of the editors didn't participate in the DRN at all basically, and I have been busy (as my userpage indicates) with school. That should explain why they have not commented. Brocade River Poems (She/They) 02:15, 9 November 2024 (UTC)
Abandoned., by a creator who has disappeared entirely from the internet – SJ + 22:21, 8 November 2024 (UTC)
We'd need a source to say that the creator has disappeared entirely (which isn't true either; their 𝕏 audience has found https://pony.best/ with their byline) Aaron Liu (talk) 22:51, 8 November 2024 (UTC)
I stand corrected! Abandoned, nonetheless. That's not a permanent state, it can be revised if that ever changes. – SJ + 03:33, 11 November 2024 (UTC)

Discussion

I wonder if this RfC is still needed. I haven't followed the dispute but it looks to me like it might only exist because a sock-farm was trying to keep option A and with this sock farm now hopefully gone it might be unnecessary. I appreciate one editor is supporting B above while the rest are supporting C and maybe the RfC having started it's too late but it just looks to me like the sort of thing which could have been resolved via normal discussion were it not for the sock farm. Nil Einne (talk) 17:40, 17 November 2024 (UTC)

It seems the original cause of this entire thing was caused just because of the current status section- It is very disputed between multiple people and many accounts can be created for spamming/reverting the final decision- This dispute has been active for about a month, and it is taking way too long in my opinion Cooldudeseven7 join in on the tea talk 17:48, 17 November 2024 (UTC)
Even if there is a claim, it is hard to find one that follows reference guidelines- and it also probably isnt WP:NPOV. Cooldudeseven7 join in on the tea talk 17:49, 17 November 2024 (UTC)
it just looks to me like the sort of thing which could have been resolved via normal discussion were it not for the sock farm. Incidentally, it had actually been resolved normally. The entire reason it became an RfC was because the sockfarm returned and reverted the edits that were decided upon at DRN. Everyone who participated in the DRN case had no problems with the proposed solution that it should be removed from the infobox. Brocade River Poems (She/They) 05:25, 19 November 2024 (UTC)

Past tense

Should the opening sentence of the article refer to the subject in the past tense? —⁠ ⁠BarrelProof (talk) 11:23, 7 November 2024 (UTC)
Yes, updated. – SJ + 22:21, 8 November 2024 (UTC)
Yes. Brocade River Poems (She/They) 02:01, 9 November 2024 (UTC)
I should have warned you sooner, but there will likely be one person who will resist and attempt that you make to do such edit. I personally am fine with this decision. Thought 1915 (talk) 18:43, 9 November 2024 (UTC)
It does seem that user has tied up the talk page for over a month. – SJ + 03:33, 11 November 2024 (UTC)
I agree that it should be rephrased to past tense -- that user does seem to be edit warring against the general interest of most people here? Any thoughts? DrawWikiped 05:06, 11 November 2024 (UTC)
I would like to mention that this is the fifth (5th) talk page topic regarding this very question, and that each time, a majority of editors seemed to agree with the suggested change. Thought 1915 (talk) 16:17, 11 November 2024 (UTC)
They are engaged in a very slow, protracted edit-war outside of the one time in October where 3RR was flagrantly violated by both sides. Brocade River Poems (She/They) 18:53, 13 November 2024 (UTC)
I don't see why not, so, sure. Aaron Liu (talk) 16:34, 11 November 2024 (UTC)
Software can exist in the present tense long past its last update, as people continue to use it. Web apps can not. I see no meaningful sense in which this still exists today. I converted the rest of the article to past tense. @RocketKnightX: don't get into a revert war; if you want the project to still be active, convince its authors to bring it back to life. – SJ + 17:50, 11 November 2024 (UTC)
Yes, and Im very close to bringing the people who constantly revert against consensus to ANI. 180.129.92.142 (talk) 03:19, 15 November 2024 (UTC)
180., that seems appropriate at this point. – SJ + 03:22, 15 November 2024 (UTC)
Oh wait, there already is one. Misplaced Pages:Administrators'_noticeboard#15.ai_behavioral_issues. 180.129.92.142 (talk) 04:08, 15 November 2024 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

15.ai finally responds

https://x.com/fifteenai/status/1865439846744871044 RocketKnightX (talk) 10:27, 15 December 2024 (UTC)

Contested deletion

User:Sirfurboy and User:Pokelego999, I rewrote this article from scratch last night. You can see the edit history to prove it. I even asked the closing admin to not give me the old version of the article because I wanted to do the subject justice. I even asked User:Liz what the proper steps I should take to make sure that I was doing everything correctly. I spent hours writing up a neutral, original version of the article as a draft, submitted it to AfC, and continued to make edits throughout the day. Please reconsider the speedy deletion, because it isn't an unaltered and unimproved version of the article. GregariousMadness (talk to me!) 15:40, 19 December 2024 (UTC)

I'm admittedly not too involved in this discussion, so I'll refrain from saying much, but I have no issues with keeping this article around since it seems to meet bare notability and can be verified as being a new version of the article via the creator's edit history. I'd appreciate an admin verifying the edit histories are different so as to dispel all reasonable doubt. However, I do have concerns about the fact there's an ongoing deletion review for the subject. I'm admittedly unsure as to how it's handled when the article is recreated while the deletion review is ongoing, but I do have concerns it may interfere with the discussion. It's still not grounds for a speedy to my knowledge, but there may be problems there. Magneton Considerer: Pokelego999 (Talk) (Contribs) 15:50, 19 December 2024 (UTC)
That's my mistake. When I was creating the draft, it said that approvals could take up to 8 weeks, so I decided to submit it and continue editing it. I wasn't expecting it to be approved so quickly. But it's definitely not true that what I wrote is an unchanged draft, which was the rationale for this speedy delete. I spent so much time researching and writing this up, and this version of the article has a lot of sources that weren't present in the now-deleted one. The second AfD for the original article was closed as a "delete" due to the sockpuppetry and canvassing, so my new research that was posted after the re-listing was never taken into consideration in the AfD. This version of the article includes all of the new sources that meet reliability and significant coverage. GregariousMadness (talk to me!) 16:13, 19 December 2024 (UTC)

Reliable sources

I'm posting this as a reference. According to archived discussions on the Teahouse, a good rule of thumb is three independent and reliable sources that demonstrate significant coverage to establish WP:GNG. Here are the major reliable sources that provide significant coverage for the subject:

  • United Daily News . Reliable as one of the largest and oldest-running newspapers in Taiwan. Listed as one of the three major Chinese-language newspapers in List of newspapers in Taiwan. Significant coverage includes an overview of the technology behind 15.ai, particularly noting its ease of use and limited data, and also discusses how 15.ai works, its features, and the viral videos that have spawned using 15.ai. Over 400 (approximate since the article is written in Chinese) words of coverage.
  • Den Fami Nico Gamer . Reliable as listed in WP:VG/RS. Significant coverage includes an overview of the DeepMoji technology used for emotiveness, applications of the voices not restricted to viral videos, and how to use it. Over 400 (approximate since the article is written in Japanese) words of coverage.
  • AUTOMATON . While not listed in WP:VG/RS, AUTOMATON is one of the largest and reputable gaming news outlets in Japan, and has been used in multiple GA's like Only Up!, Visions of Mana, and Sprigatito, Floragato, and Meowscarada. Significant coverage includes DeepMoji, a list of characters available on the application, examples of video content users have created with the platform, an overview of the pronunciation capabilities of the model, as well as a mention of how to use ARPAbet strings. Almost 800 (approximate since the article is written in Japanese) words of coverage.
  • Rionaldi Chandraseta . While the article itself is written on Medium (which is not considered reliable), Medium is only being used as a vessel to host the article itself (similar to how Google Docs can be used to host an article), which is part of a very popular newsletter called Towards Data Science, which has almost 800K followers on social media. Following alone means nothing in determining the reliability of a source, but Rionaldi Chandraseta, the author of the article, is an IEEE-published machine learning specialist who has published papers that are listed on Google Scholar . The newsletter has a solid editorial board that consists of multiple masters and PhD's in machine learning and computer science. Over 1,000 words of English-language coverage detailing every facet of 15.ai, from its capabilities to its underlying research.
  • Yongqiang Li . Since the article is locked to foreigners without an account, I asked a friend to translate this for me. The article goes into great detail about the technology behind 15.ai and talks about its features, its future, and potential problems. The author is a professor at the Harbin Institute of Technology and has multiple publications listed on Google Scholar and ResearchGate .
  • Eurogamer . Reliable as listed in WP:VG/RS. While the main focus of the article isn't 15.ai, it goes into detail the controversy and Twitter exchange that happened when Voiceverse NFT misappropriated 15.ai's work. From However, in now-deleted tweets, Voiceverse was found to have boasted about using its tech for the voice of a cartoon character - which was in fact created using 15.ai, a popular non-commercial text-to-speech service. to "Hey @fifteenai we are extremely sorry about this," Voiceverse NFT wrote. "The voice was indeed taken from your platform, which our marketing team used without giving proper credit. Chubbiverse team has no knowledge of this. We will make sure this never happens again.", this is about 300 words of coverage.
  • Stevivor . After doing more research, I found that Steven Wright, the author of this article, also writes for Inverse, a solid and well-known technology and gaming publication. In addition, Stevivor is reliable and independent, and it is the most-read independent gaming news network in the Oceanic region.
  • Kotaku . While Kotaku is in WP:VG/RS, it also states News posts from Kotaku between 2010 and 2022 are considered reliable, (the article is from 2021, so it meets this criteria) but also states {{tq|although editors are cautioned of blog/geeky posts that have little news or reporting significance}. It's still debated whether an article from the "Odds and Ends" category is considered "News", and the entry in WP:VG/RS says articles should be evaluated on a case-by-case basis. There is no clear consensus to this, but the description for the "Odds and Ends" category is "Odds and Ends - Gaming Reviews, News, Tips and More.". The argument in the previous AfD was that this article did not meet reliability, although it met independence and significant coverage. I personally believe that this article is reliable.
  • Game Informer, PC Gamer, and Rock, Paper, Shotgun All three of these sources are found under WP:VG/RS, but there has been a debate whether these three met significant coverage. While they all pass WP:100WORDS, it is not a Misplaced Pages policy and their significant coverage can be debated.
  • NME . WP:RS notes NME is reliable in its expertise, and it has been debated whether gaming is one of their areas of expertise. The Misplaced Pages article for NME states that this is so, and gaming is listed as one of NME's header sections, but there has been debate whether NME's expertise extends outside of music. Similar coverage to Eurogamer, but with fewer words, but still above the threshhold for WP:100WORDS (which, again, is not Misplaced Pages policy).
  • Andrew Ng . The author, Andrew Ng, is one of the most famous and influential artificial intelligence researchers in the world, with a healthy Google Scholar profile and was included in the Time 100 Most Influential People in AI list in 2023. While 15.ai is mentioned as a blurb and likely does not meet significant coverage, it shows that the subject wasn't a mere curiosity and was under the radar for a large number of prominent figures in AI while the service was active. — Preceding unsigned comment added by GregariousMadness (talkcontribs) 14:00, 20 December 2024 (UTC)

However, some of these have been contested in some form, the arguments for which I personally disagreed with. Think of these sources as you will. I still haven't gone over many of the the sources that are used in the newly written page, but I will continue to do my research and update this.

Yes, I'm aware that this is a contentious article that was submitted to AfD. But the AfD was closed largely due to the misbehaving of new Misplaced Pages editors, who are likely to be children, which is not surprising given the popularity of the application among younger people. I'm committed to doing this subject justice, and I argue that this subject not only meets the bare minimum of notability, but meets it well-within question. GregariousMadness (talk to me!) 17:11, 19 December 2024 (UTC)

These sources look largely reliable to me. It was a large part of why I accepted the original draft in the first place, since it's rare to see a draft with such good sourcing. I personally believe this meets notability, especially since it's been clarified outright this is much expanded from the deleted article. Magneton Considerer: Pokelego999 (Talk) (Contribs) 18:45, 19 December 2024 (UTC)
Thank you. Sorry for all the trouble. I never meant for all this to happen. I can get carried away at times and I felt a fire light up inside me when I was writing the new article. I really hope this version gets to stay. GregariousMadness (talk to me!) 21:42, 19 December 2024 (UTC)

Please fill out the missing stuff in the Japanese source citations

@GregariousMadness: Please see the example for a foreign-language reference below. What's needed are the trans-title, language, quote, and trans-quote params

{{cite news 
|last=Vučković 
|first=Branko 
|date=28 September 2013 
|title=Železnice Srbije u sve lošijem stanju
|trans-title=Serbian Rail in Increasingly Worse Condition 
|url=https://www.slobodnaevropa.org/a/zeleznice-srbije-u-sve-losijem-stanju/25120244.html 
|access-date=5 November 2024
|website=Radio Slobodna Evropa |language=sh 
|quote=Koliko je železnica u Srbiji zaostala, najbolje se vidi iz poređenja sa Evropskom unijom, gde se vozovi kreću prosečnom brzinom između 200 i 300 kilometara na sat, dok je prosečna brzina na domaćim prugama 44 kilometra, samo nekoliko kilometara brže od prvog voza koji je saobraćao na tek sagrađenoj pruzi Beograd - Niš u septembru 1884. godine. 
|trans-quote=The extent to which the railway system in Serbia has fallen behind is best illustrated by comparing it with the European Union, where trains travel at an average speed of between 200 and 300 kilometers per hour, while the average speed on domestic tracks is just 44 kilometers, only a few kilometers faster than the first train that operated on the newly constructed Belgrade–Niš line in September 1884.}}

Alalch E. 12:41, 20 December 2024 (UTC)

Done, thank you! GregariousMadness (talk to me!) 13:05, 20 December 2024 (UTC)
Thanks a lot. Having translated titles is really important, but it would also be very nice and genuinely helps verifiability to have translated relevant quotes. If it needs to be multiple quotes from the same source, cite the same source multiple times each time with a different suitable quote and I might convert the citation method to shortened footnotes. If you determine that any of the Asian sources aren't essential, and the statements are already made verifiable using English-language sources, please remove them. —Alalch E. 13:10, 20 December 2024 (UTC)
Oh... I just saw this message after I submitted an edit translating all of the other sources. What should I do? Should I revert my edit? GregariousMadness (talk to me!) 13:18, 20 December 2024 (UTC)
No worries, please see Special:Diff/1264103607. Basically, there's no need to use {{efn}} for translations of quotes. —Alalch E. 13:39, 20 December 2024 (UTC)
In other words, what Japanese text in the reference no. 2 (as of special:permalink/1264103607) supports the statement "15.ai was conceived as a research project by a developer known as "15" during their undergraduate studies at the Massachusetts Institute of Technology, and was later implemented following the developer's successful exit from a startup venture." and what is the English translation of that Japanese text. —Alalch E. 13:42, 20 December 2024 (UTC)
I see, I'll get to that in a bit. I'm expanding some other sections but I'll definitely make those edits soon. GregariousMadness (talk to me!) 13:49, 20 December 2024 (UTC)
Categories: