Chiming in as a journal editor here. Martin's last note resonated for me. I think there is something in this conversation as a provocation about where our limits of acceptability are in the absence of clear norms. I have the same reservations about disclosure statements. We have never (at least at my journal) asked authors to disclose things like statisticians, editors or other supporting characters. So why and what is so different about LLM use? I do think there is a fundamental difference, but until we can articulate an object (or objective ish) response that makes a material difference, I am wary of disclosure statements. I am also influenced by this paper about moving from disclosure to answerability: AI and the Future of Publishing <https://link.springer.com/article/10.1007/s13347-026-01148-8> While I am airing out my thoughts.. I also would not consent to be included in a research project, but I do like the suggestion from yesterday about pivoting to a more participatory conversation followed by a jointly produced paper. Have a wonderful weekend. Lauri _________________________________________ Lauri Goldkind, PhD ~Editor in Chief, Journal of Technology in Human Services <https://www.tandfonline.com/toc/wths20/current> *Founder ~ PAIRED Lab <https://www.non-profits.ai/>* <https://www.datajusticecollective.org> Professor, Graduate School of Social Service, Fordham Nonprofit AI Resources <https://www.notion.so/230ba982576680dbb568c1e2c977cd9d?v=230ba982576681649347000c86bafd31&source=copy_link> comPITence a project of the Public Interest Technology University Network <https://compitencies.org/nce.org> Read the latest: AI in the Nonprofit Human Services <https://www.tandfonline.com/doi/full/10.1080/23303131.2024.2427459#abstract> http://www.laurigoldkind.net/ *"Do what you can, with what you have, where you are." **-Theodore Roosevelt* *"You don't need what you don't need*." *Akira Kurosawa * *Illegitimi non carborundum* On Sat, Oct 17, 2026 at 9:29 AM Martin Berg via Air-L <air-l@lists.aoir.org> wrote:
Dear Jill and David (and everyone else),
Thanks to you both. I'm probably closer to David on this. I think there is something here worth exploring, and what interests me most is what actually happens between human and machine when we write, including what it does to genre and voice.
Of course we each have a voice of our own. I'm just less sure it was ever untouched by the technologies we write with or the people we write among. Scholarly texts go through co-authors, editors, copy-editors and Reviewer 2, and most of us learnt long ago to write towards conventions we never chose. So whose voice is on the page, even before a language model gets anywhere near it? And where would we stop declaring things? The spellchecker, Grammarly, Google Translate, the long departed and still slightly missed Clippy in Word?
I've been poking at these boundaries for a while. In the chapter From Mead to Machine: Conversational Choreographies with ChatGPT and the Emergence of the Socio-Technical Self in Generative AI <https://doi.org/10.1108/S0163-239620250000061003> I let ChatGPT write its own interjections into the text, marked as its own, and in the article Speculating with Machines: Generative AI in Futures Ethnography <https://doi.org/10.1177/10778004261434468> Sarah Pink and I followed what happened when ChatGPT took part in a futures workshop, including how it pulled people towards a more uniform register. Neither settled where the line between human and machine goes. I believe those boundaries are something to explore and reflect on, rather than something already settled and labelled an AI voice.
The About page <https://www.slightlyuncanny.com/about/> says how the blog is made, and also that the line between me and the machine is left blurry on purpose. I think of that as an artistic way of unsettling authorship, and I've found it useful for thinking about my own writing. Anyway, it's Saturday, so there's a new episode of the radio show as well, where my cloned voice (or I?) reads Saturday’s longer post over some cosmic country tunes. It's on Spotify <https://open.spotify.com/episode/31Qg0HMM4DocxPPSMujHW8?si=LfRL0wFIRF67vxUZi6P6jw&utm_source=copy-link> too, if you'd rather skip the blog.
Honestly, I see the whole thing more as a deliberate provocation than as research output. Whether it's art I don't know, but if it is, I'm not sure how much one ought to explain or apologise for it, or how seriously any of us should take ourselves.
Thanks for the reading suggestions, Jill. I may well give them to the machine to read and “reflect” on.
Best wishes,
Martin
-- Martin Berg, PhD Professor | Technology and Society, Malmö University Director, Digital Work Futures Research Lab Adjunct Professor (Research), Monash University (Emerging Technologies Research Lab) martin.berg@mau.se | +46 40 665 80 60 <+46%2040%20665%2080%2060> Profile: https://mau.se/en/persons/martin.berg DWF Lab: https://mau.se/dwf LinkedIn: https://www.linkedin.com/in/profberg/ _______________________________________________ The To unsubscribe send an email to air-l-leave@lists.aoir.org mailing list is provided by the Association of Internet Researchers http://aoir.org Subscribe, change options or unsubscribe at: Join the Association of Internet Researchers: http://www.aoir.org/