Timeline for answer to Should we request a banner to communicate our policy regarding generative AI in posts? by Ewan
Current License: CC BY-SA 4.0
Post Revisions
30 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 29, 2024 at 9:02 | comment | added | Ewan | eg : softwareengineering.stackexchange.com/questions/453931/… | |
| Feb 6, 2024 at 19:43 | comment | added | Doc Brown | @ThomasOwens: to be fair, the policy for Software Engineering mentioned above is something you presented to us, nothing which was really created by the community. Not that I disagree to what you wrote there, I guess it is much better than the current Stack Overflow AI policy text. | |
| Feb 3, 2024 at 23:03 | comment | added | Thomas Owens Mod | What "machine"? The heuristics are applied by people. If enough of the heuristics match, the post gets deleted. If people claim otherwise, that will be the answer - it looks so much like generated content that it is likely to be generated content and doesn't belong here. | |
| Feb 3, 2024 at 21:54 | comment | added | Ewan | At some point you are going to get false flags and people will say "but my post isnt generated?!?" you cant just say the policy is anything caught by the machine is wrong. | |
| Feb 3, 2024 at 16:38 | comment | added | Thomas Owens Mod | There is no "primarily/substantively/mainly". There are heuristics for code and heuristics for English. If you make a post - a question or an answer - and sufficient heuristics for a strong enough claim that it is generated content, it gets removed. It doesn't matter if all those markers are in 1 paragraph out of 8 or spread across the whole post. If you keep posting generated content, you get suspended. The heuristics are designed to allow the use of spelling, grammar, translation tools to help people craft higher quality answers while prevent the use of tools that pose risks to users. | |
| Feb 3, 2024 at 15:23 | comment | added | Ewan | TBH I think in the near future all coding tools will have GenAI autocomplete built in. which means all the little example code snippets in answers will be generated using genAI, the policy should be written with that kind of thing in mind | |
| Feb 3, 2024 at 15:21 | comment | added | Ewan | thats why im suggesting its clean up to something a bit more concise "Answers must be primarily/substantively/mainly? written by the author" obvs needs a bit more with context about AI etc, but its clear and allows generated content to be used where reasonable. | |
| Feb 3, 2024 at 15:18 | comment | added | Ewan | Ok, this is why the clarity needs to be improved. I'm not talking about questions, im talking about answers with some content vs completely generated answers. The Software Eng linked policy talks about "Using AI-powered tools to generate multiple answers" and "posting machine-generated answers" its not the same as your statement in this question : "the use of generative AI to create content is prohibited" which would exclude all of my examples where the answer contains some AI content. The StackExchange linked policy talks about attribution of content "not created by you" different again. | |
| Feb 3, 2024 at 14:55 | comment | added | Thomas Owens Mod | On Software Engineering, answers are, by far, the biggest risk to consumers and the overall quality of the content here. I can't think of a case where posting a question generated by AI would not be handled under some other mechanism. The questions would need details, focus, clarity or be plagiarized, all of which would result in closure and deletion. That's not true for all sites in the network, though. The overall Network policy is that generative AI must be attributed. Our policy is that you don't post it - it will get deleted. That's pretty clear to me. | |
| Feb 3, 2024 at 14:50 | comment | added | Ewan | even you who wrote the policy are alternating between "no content" and "no answers" | |
| Feb 3, 2024 at 14:49 | comment | added | Ewan | I want more precision. and have given a suggestion. | |
| Feb 3, 2024 at 14:48 | comment | added | Thomas Owens Mod | We do have a clear policy. None of our policies can be algorithmically applied. Human judgement is always used. It's why we have human moderators looking at flags from human users and a Meta site to ask questions. If you are looking for legal precision in our policies, you won't find it and there's no need for that - it would hamstring the ability of moderators to make decisions that consider all factors, such as value and overall quality. | |
| Feb 3, 2024 at 14:44 | comment | added | Ewan | What i'm saying is we need to have a clear policy which allows those type of examples while banning the things you listed. I don't think the language in your linked meta question, or the language used in this question is good enough. Which I why I have made the suggested wording here | |
| Feb 3, 2024 at 14:42 | comment | added | Thomas Owens Mod | Only one of your examples is bad. The first is off-topic here, but would be allowed on the Generative AI Stack Exchange. Although the answer contains Generative AI, the answer was not generated by AI. The second is off-topic here, but would be removed on sites like Stack Overflow and probably Code Review if it met enough of the heuristics to detect generated code. The third would be permitted as translation tools, when used properly, are not generative AI. | |
| Feb 3, 2024 at 14:41 | history | edited | Ewan | CC BY-SA 4.0 |
added 1038 characters in body
|
| Feb 3, 2024 at 14:39 | comment | added | Ewan | I have added three examples of what im talking about | |
| Feb 3, 2024 at 14:39 | comment | added | Ewan | Can you highlight where in the linked policy it excludes answers contain copilot content? I find that post unclear and confusing as a policy statement. | |
| Feb 3, 2024 at 14:36 | history | edited | Ewan | CC BY-SA 4.0 |
added 1038 characters in body
|
| Feb 3, 2024 at 14:34 | comment | added | Thomas Owens Mod | We have a set of heuristics, created by staff, moderators, and other select users used to detect Generative AI. Any post that meets sufficient heuristics can be deleted. Having multiple posts deleted for matching these heuristics warrants a suspension. | |
| Feb 3, 2024 at 14:33 | comment | added | Thomas Owens Mod | This is our policy for Software Engineering. It prohibits the use of generative AI. Single instances of using generative AI will result in deletion. Abuse of generative AI - such as continuing to post generative AI after being warned or flooding the site with generative AI content quickly - is ground for suspension.. (1/2) | |
| Feb 3, 2024 at 14:25 | comment | added | Ewan | and where do you say "no generative AI" as opposed to "no generative AI generated answers"? | |
| Feb 3, 2024 at 14:22 | comment | added | Ewan | sorry, which policy doesn't allow copilot? | |
| Feb 3, 2024 at 13:47 | comment | added | Thomas Owens Mod | But the policy doesn't allow for Copilot. And there are fundamental differences between spell check or translators and Generative AI. We aren't saying "no AI". We are saying "no Generative AI". And even if you attribute AI, that's not permitted here. | |
| Feb 3, 2024 at 13:22 | comment | added | Ewan | I don't think anyone disagrees with that, it's just wording the policy to allow (for example) use of copilot/autocomplete in human written example code, but not 100%/90%/80%? generated answers. If you just say "No AI content!" well there is spell check! if you say "NO AI answers" well I put 'The AI says this:' at the start of the answer! You want something short and sweet which allows some flexibilty | |
| Feb 3, 2024 at 13:18 | comment | added | Thomas Owens Mod | Moderators and users have consistently, across the Network, rejected GenAI because it floods the site with content that needs to be closely curated and can't easily be rejected. When content, on the surface, looks good and correct, but has subtle flaws, it's much easier to detect the class of content and reject it. That approach is less harmful. It reduces the risk of people being mislead by low quality or incorrect content and is more respectful of the time of the unpaid contributors and moderators since it's faster to detect generative AI than to validate the flood of generative AI content. | |
| Feb 3, 2024 at 13:16 | comment | added | Thomas Owens Mod | There's absolutely nothing wrong with using translation, grammar editing tools, and things that help a human write better content. No reputable tool that performs these functions should fall under any of the currently valid heuristics to detect generated content. Copilot - when used to generate large amounts of code - does fall into prohibited content, though. The biggest issue with generative AI is that it is incredibly easy to produce garbage. (1/2) | |
| Feb 3, 2024 at 13:11 | comment | added | Ewan | A key thing in future might be the use of copilot or similar tools to write example code which is then included in answers. I don't think anyone would want to exclude that kind of thing, it's just better autocomplete | |
| Feb 3, 2024 at 13:07 | comment | added | Ewan | I guess i was reading the summary in your question "the use of generative AI to create content is prohibited" rather than the linked "posting machine-generated answers is highly discouraged". I'm imagining an answer such as, "blah blah an example of machine generated content which proves this point might be : "machine generated content"", or the kind of things doc is talking about eg "....or as they say in french 'ce la vie!' (google translate used)" where the answer is human written, but contains machine generated parts | |
| Feb 3, 2024 at 12:48 | comment | added | Thomas Owens Mod | Our current policy is to delete all answers that match enough of the heuristics for generative AI. These policies and heuristics already exclude most, if not all, translations, since that's not generating new content. And nothing is said about examples of generative AI used to support existing answers, since that is not an answer itself. It should also be pointed out that SE has reviewed and all sites will be getting a banner. So it becomes a matter of which banner, and the only banner that matches our policy is the first one stating that generated answers are not allowed. | |
| Feb 3, 2024 at 12:23 | history | answered | Ewan | CC BY-SA 4.0 |