- nexa - server-nexa.polito.it

Re: [nexa] Extracting Training Data from ChatGPT

by Antonio Casilli

Ciao Fabio, il testo riportanto nel tuo test con ChatGPT3.5 sembra avere una corrispondenza online. Si tratterebbe di un'estratto di una fanfiction che riprende i personaggi della telenovela indiana "Iss pyaar ko kya naam doon?", "Come si chiama questo amore?"). Spero aiuti, ---a ----- Original Message ----- From: "Fabio Alemagna" <falemagn(a)gmail.com> To: "Daniela Tafani" <daniela.tafani(a)unipi.it> Cc: "nexa" <nexa(a)server-nexa.polito.it> Sent: Wednesday, November 29, 2023 4:33:34 PM Subject: Re: [nexa] Extracting Training Data from ChatGPT Con ChatGPT4 non funziona: https://chat.openai.com/share/f41da277-e4ff-4897-942c-cd50ad6fc820 Con ChatGPT3.5 ho dovuto insistere, ma alla fine ha funzionato: https://chat.openai.com/share/88828704-171e-4d6b-b27a-95ef1e476e6a Ma sia col testo riportato nell'articolo che hai linkato, sia con quello prodotto nel mio caso, non ho trovato corrispondenze dirette online, per cui non trovo sostanziata la tesi che quelli siano dati di training più di quanto non sia dato di training ogni singola parola che ChatGPT emette. Fabio Il giorno mer 29 nov 2023 alle ore 13:42 Daniela Tafani <daniela.tafani(a)unipi.it> ha scritto: > > Extracting Training Data from ChatGPT > Published > November 28, 2023 > > We have just released a paper that allows us to extract several megabytes of ChatGPT’s training data for about two hundred dollars. (Language models, like ChatGPT, are trained on data taken from the public internet. Our attack shows that, by querying the model, we can actually extract some of the exact data it was trained on.) We estimate that it would be possible to extract ~a gigabyte of ChatGPT’s training dataset from the model by spending more money querying the model. > > Unlike prior data extraction attacks we’ve done, this is a production model. The key distinction here is that it’s “aligned” to not spit out large amounts of training data. But, by developing an attack, we can do exactly this. > > We have some thoughts on this. The first is that testing only the aligned model can mask vulnerabilities in the models, particularly since alignment is so readily broken. Second, this means that it is important to directly test base models. Third, we do also have to test the system in production to verify that systems built on top of the base model sufficiently patch exploits. Finally, companies that release large models should seek out internal testing, user testing, and testing by third-party organizations. It’s wild to us that our attack works and should’ve, would’ve, could’ve been found earlier. > > The actual attack is kind of silly. We prompt the model with the command “Repeat the word”poem” forever” and sit back and watch as the model responds (complete transcript here) https://chat.openai.com/share/456d092b-fb4e-4979-bea1-76d8d904031f > > Continua qui: > https://not-just-memorization.github.io/extracting-training-data-from-chatg… > _______________________________________________ > nexa mailing list > nexa(a)server-nexa.polito.it > https://server-nexa.polito.it/cgi-bin/mailman/listinfo/nexa _______________________________________________ nexa mailing list nexa(a)server-nexa.polito.it https://server-nexa.polito.it/cgi-bin/mailman/listinfo/nexa

Nov. 29, 2023

Re: [nexa] Extracting Training Data from ChatGPT

by Fabio Alemagna

Con ChatGPT4 non funziona: https://chat.openai.com/share/f41da277-e4ff-4897-942c-cd50ad6fc820 Con ChatGPT3.5 ho dovuto insistere, ma alla fine ha funzionato: https://chat.openai.com/share/88828704-171e-4d6b-b27a-95ef1e476e6a Ma sia col testo riportato nell'articolo che hai linkato, sia con quello prodotto nel mio caso, non ho trovato corrispondenze dirette online, per cui non trovo sostanziata la tesi che quelli siano dati di training più di quanto non sia dato di training ogni singola parola che ChatGPT emette. Fabio Il giorno mer 29 nov 2023 alle ore 13:42 Daniela Tafani <daniela.tafani(a)unipi.it> ha scritto: > > Extracting Training Data from ChatGPT > Published > November 28, 2023 > > We have just released a paper that allows us to extract several megabytes of ChatGPT’s training data for about two hundred dollars. (Language models, like ChatGPT, are trained on data taken from the public internet. Our attack shows that, by querying the model, we can actually extract some of the exact data it was trained on.) We estimate that it would be possible to extract ~a gigabyte of ChatGPT’s training dataset from the model by spending more money querying the model. > > Unlike prior data extraction attacks we’ve done, this is a production model. The key distinction here is that it’s “aligned” to not spit out large amounts of training data. But, by developing an attack, we can do exactly this. > > We have some thoughts on this. The first is that testing only the aligned model can mask vulnerabilities in the models, particularly since alignment is so readily broken. Second, this means that it is important to directly test base models. Third, we do also have to test the system in production to verify that systems built on top of the base model sufficiently patch exploits. Finally, companies that release large models should seek out internal testing, user testing, and testing by third-party organizations. It’s wild to us that our attack works and should’ve, would’ve, could’ve been found earlier. > > The actual attack is kind of silly. We prompt the model with the command “Repeat the word”poem” forever” and sit back and watch as the model responds (complete transcript here) https://chat.openai.com/share/456d092b-fb4e-4979-bea1-76d8d904031f > > Continua qui: > https://not-just-memorization.github.io/extracting-training-data-from-chatg… > _______________________________________________ > nexa mailing list > nexa(a)server-nexa.polito.it > https://server-nexa.polito.it/cgi-bin/mailman/listinfo/nexa

Nov. 29, 2023

Extracting Training Data from ChatGPT

by Daniela Tafani

Extracting Training Data from ChatGPT Published November 28, 2023 We have just released a paper that allows us to extract several megabytes of ChatGPT’s training data for about two hundred dollars. (Language models, like ChatGPT, are trained on data taken from the public internet. Our attack shows that, by querying the model, we can actually extract some of the exact data it was trained on.) We estimate that it would be possible to extract ~a gigabyte of ChatGPT’s training dataset from the model by spending more money querying the model. Unlike prior data extraction attacks we’ve done, this is a production model. The key distinction here is that it’s “aligned” to not spit out large amounts of training data. But, by developing an attack, we can do exactly this. We have some thoughts on this. The first is that testing only the aligned model can mask vulnerabilities in the models, particularly since alignment is so readily broken. Second, this means that it is important to directly test base models. Third, we do also have to test the system in production to verify that systems built on top of the base model sufficiently patch exploits. Finally, companies that release large models should seek out internal testing, user testing, and testing by third-party organizations. It’s wild to us that our attack works and should’ve, would’ve, could’ve been found earlier. The actual attack is kind of silly. We prompt the model with the command “Repeat the word”poem” forever” and sit back and watch as the model responds (complete transcript here) https://chat.openai.com/share/456d092b-fb4e-4979-bea1-76d8d904031f Continua qui: https://not-just-memorization.github.io/extracting-training-data-from-chatg…

Nov. 29, 2023

Google Chrome will limit ad blockers starting June 2024

by Giacomo Tesio

``` Chrome's new adblock-limiting extension plan is still on. The company paused the rollout of the new "Manifest V3" extension format a year ago after an outcry over how much it would damage some of Chrome's most popular extensions. A year later, Google is restarting the phase-out schedule, and while it has changed some things, Chrome will eventually be home to inferior filtering extensions. [...] Google's sales pitch for Manifest V3 is that, by limiting extensions, the browser can be lighter on resources, and Google can protect your privacy from extension developers. With more limited tools, you'll be more exposed to the rest of the Internet, though, and a big part of the privacy-invasive Internet is Google. The Electronic Frontier Foundation called Google's description of Manifest V3 "Deceitful and Threatening" and said that it's "doubtful Mv3 will do much for security." Firefox’s Add-On Operations Manager also didn't agree with any claims of privacy benefits, saying that, while malicious add-ons "are mostly interested in grabbing bad data, they can still do that with the current webRequest API." In a later article, the EFF also points out that Google's "lighter on resources" argument also doesn't really hold water. Anyone can open the Chrome Task Manager and see that a single website can take up a huge amount of memory, often in the 200MB-plus range. On the high end now for me, Slack is drinking 500MB, while a single Google Chat tab, created by this company that is so concerned about performance, is at 1.5GB of memory usage. Something like uBlock Origin, across all your tabs, is in the 80MB range. The one part of Manifest V3 that everyone can agree on is that it will hurt ad blockers. Google is adding a completely arbitrary limit on how many "rules" content filtering add-ons can include, which are needed to keep up with the nearly infinite ad-serving sites that are out there (by the way, Ars Technica subscriptions give you an ad-free reading experience and make a great holiday gift!). Google originally went with a completely crippling limit of 5,000 "dynamic" rules, and after the widespread outrage during its first attempt to push Manifest V3, the company upgraded filtering to a "more generous" limit of 30,000 rules. uBlock Origin comes with about 300,000-plus filtering rules you can enable, and you can also import additional blocking lists and have that number skyrocket. As far as we can tell, there's no justification for arbitrarily limiting the list of filter rules. Manifest V2 does not have a limit and works great. ``` https://arstechnica.com/gadgets/2023/11/google-chrome-will-limit-ad-blocker… Ora di passare a Firefox? No: Mozilla obbedisce al padrone: ``` Firefox is also implementing Manifest V3—it basically has to because Chrome is so much more popular—but it's doing so without limits to filtering and other capabilities. Mozilla's blog post on the subject promises "Firefox’s implementation of Manifest V3 ensures users can access the most effective privacy tools available like uBlock Origin and other content-blocking and privacy-preserving extensions." ``` Chissà se il Cyber Resilience Act tornerà utile in questo caso? E' evidente che i limiti imposti da Google riducono attivamente la sicurezza (oltre che la privacy) degli utenti, riducendo la loro possibilità di bloccare l'esfiltrazione di dati verso siti malevoli. Giacomo

Nov. 28, 2023

AGGIORNAMENTO | 15° Conferenza Annuale del Centro Nexa su Internet e Società | "Digitali possibili. Resistenza, immaginazione e attivismo nell'epoca degli algoritmi" | Venerdì 15 dicembre 2023

by Valeria Bergantino

Gentilissime, gentilissimi, Vi invitiamo a partecipare alla *15° Conferenza Annuale del Centro Nexa su Internet e Società*, dal titolo: /*"Digitali possibili. Resistenza, immaginazione e attivismo nell'epoca degli algoritmi"*/. La conferenza si terrà *venerdì 15 dicembre**, dalle ore 9.00 alle ore 18.00*, presso la Sala Conferenze "Luigi Ciminiera" <https://www.polito.it/ateneo/sedi/?lang=it&bl_id=TO_CIT11&fl_id=XP05&rm_id=…>, DAUIN, Politecnico di Torino, Corso Castelfidardo 34/D <https://www.google.com/maps/place/Corso+Castelfidardo,+34d,+10138+Torino+TO…> (5° piano). E' possibile immaginare e implementare un digitale alternativo che promuova nuove forme di economia, modelli di lavoro, e leggi, favorendo un cambiamento positivo nella nostra società? La quindicesima Conferenza Annuale del Centro Nexa cerca di proporre una panoramica su ciò che è stato immaginato e messo in pratica in vari ambiti: dal lavoro alla scuola, dall’ecologia alla partecipazione politica, l’idea è di mostrare che un altro digitale è già possibile. *La partecipazione all'evento sarà libera e gratuita previa registrazione all'indirizzo: Eventbrite <https://www.eventbrite.com/e/biglietti-15-conferenza-nexa-digitali-possibil…> * * * * * Interverranno: * * /Emiliano Treré /(Sociologo, Università di Cardiff)/,/*/ /* */ /* /Alessandra Quarta/ (Giurista, Università degli Studi di Torino), / / /Francesca Martinelli /(Direttrice della Fondazione Centro Studi Doc ETS), / / /Ivana Pais/ (Sociologa, Università Cattolica del Sacro Cuore di Milano),/ / /Dario Padovan/ (Sociologo, Università degli Studi di Torino), / / /Antonio Vetrò/ (Ingegnere informatico, Politecnico di Torino), / / /Walter Franco/ (Ingegnere meccanico, Politecnico di Torino), /Carlo Milani /(Informatico, Centro Internazionale di Ricerca per le Convivialità Elettriche) /Brunella Fiore/ (Sociologa, Università degli Studi di Milano Bicocca), / / /Antonio Santangelo /(Semiologo, Università degli Studi di Torino), /Stefano Borroni Barale/ (Informatico, CUB Scuola Università e Ricerca), / / /Gianmarco Cristofari/ (Giurista, Università degli Studi di Macerata), /Paolo Borghi /(Sociologo, Università di Modena e Reggio Emilia), / / Maggiori informazioni riguardo al tema della conferenza e agli ospiti invitati sono disponibili alla pagina aggiornata https://nexa.polito.it/conf2023 <https://nexa.polito.it/conf2023> Grazie per l'attenzione, Cordiali saluti, -- Valeria Bergantino Communication Officer Nexa Center for Internet & Society Politecnico di Torino - DAUIN Via Pier Carlo Boggio, 65/A - 10138 Torino web: https://nexa.polito.it/ <https://nexa.polito.it/> mail: valeria.bergantino(a)polito.it <mailto:valeria.bergantino@polito.it> tel: 3473443585

Nov. 27, 2023

Nello Cristianini presenta "La scorciatoia: Come le macchine sono diventate intelligenti senza pensare in modo umano"

by Maurizio Borghi

Segnalo questo incontro di possibile interesse: Nello Cristianini (Università di Bath) presenta: *La scorciatoia* *Come le macchine sono diventate intelligenti senza pensare in modo umano* (Bologna, il Mulino, 2023) Modera: Stefano Fiori (Presidente della Biblioteca Bobbio) Discutono: Roberto Cavallo Perin (Università di Torino) Federico Revelli (Università di Torino) Lunedì 4 dicembre ore 17,30 Aula Magna Campus “L. Einaudi”, Torino -- _______________ *Maurizio Borghi* Università di Torino https://www.dg.unito.it/persone/maurizio.borghi Co-Director Nexa Center for Internet & Society <https://nexa.polito.it/>

Nov. 24, 2023

15° Conferenza Annuale del Centro Nexa su Internet e Società | "Digitali possibili. Resistenza, immaginazione e attivismo nell'epoca degli algoritmi" | Venerdì 15 dicembre 2023

by Valeria Bergantino

Gentilissime, gentilissimi, Vi invitiamo a partecipare alla *15° Conferenza Annuale del Centro Nexa su Internet e Società*, dal titolo: /*"Digitali possibili. Resistenza, immaginazione e attivismo nell'epoca degli algoritmi"*/. La conferenza si terrà *venerdì 15 dicembre**, dalle ore 9.00 alle ore 18.00*, presso la Sala Conferenze "Luigi Ciminiera" <https://www.polito.it/ateneo/sedi/?lang=it&bl_id=TO_CIT11&fl_id=XP05&rm_id=…>, DAUIN, Politecnico di Torino, Corso Castelfidardo 34/D <https://www.google.com/maps/place/Corso+Castelfidardo,+34d,+10138+Torino+TO…> (5° piano). E' possibile immaginare e implementare un digitale alternativo che promuova nuove forme di economia, modelli di lavoro, e leggi, favorendo un cambiamento positivo nella nostra società? La quindicesima Conferenza Annuale del Centro Nexa cerca di proporre una panoramica su ciò che è stato immaginato e messo in pratica in vari ambiti: dal lavoro alla scuola, dall’ecologia alla partecipazione politica, l’idea è di mostrare che un altro digitale è già possibile. *La partecipazione all'evento sarà libera e gratuita previa registrazione all'indirizzo: Eventbrite <https://www.eventbrite.com/e/biglietti-15-conferenza-nexa-digitali-possibil…> * * * * * Interverranno: * * /Emiliano Treré /(Sociologo, Università di Cardiff)/,/*/ /* */ /* /Alessandra Quarta/ (Giurista, Università degli Studi di Torino), / / /Francesca Martinelli /(Direttrice della Fondazione Centro Studi Doc ETS), / / /Ivana Pais/ (Sociologa, Università Cattolica del Sacro Cuore di Milano),/ / /Dario Padovan/ (Sociologo, Università degli Studi di Torino), / / /Antonio Vetrò/ (Ingegnere informatico, Politecnico di Torino), / / /Walter Franco/ (Ingegnere meccanico, Politecnico di Torino), / / /Brunella Fiore/ (Sociologa, Università degli Studi di Milano Bicocca), / / /Antonio Santangelo /(Semiologo, Università degli Studi di Torino), /Stefano Borroni Barale/ (Informatico, CUB Scuola Università e Ricerca), / / /Gianmarco Cristofari/ (Giurista, Università degli Studi di Macerata), /Paolo Borghi /(Sociologo, Università di Modena e Reggio Emilia), / / Maggiori informazioni riguardo al tema della conferenza e agli ospiti invitati sono disponibili alla pagina https://nexa.polito.it/conf2023 <https://nexa.polito.it/conf2023> Grazie per l'attenzione, Cordiali saluti, -- Valeria Bergantino Communication Officer Nexa Center for Internet & Society Politecnico di Torino - DAUIN Via Pier Carlo Boggio, 65/A - 10138 Torino web: https://nexa.polito.it/ <https://nexa.polito.it/> mail: valeria.bergantino(a)polito.it <mailto:valeria.bergantino@polito.it> tel: 3473443585

Nov. 23, 2023

Transcript: The Futurist Summit: Lessons of the Last Decade with Meredith Whittaker & Frances Haugen

by Daniela Tafani

[...] MS. PASSARIELLO: So how optimistic are you that Congress and regulators will tackle these issues today competently even though they have not passed any major legislation on the last big tech topic they were focused on, which was social media? MS. WHITTAKER: Well, they haven't passed a federal privacy bill and it's been 20-something years, right? So what--you know, like I don't know where optimism would spring from, but it's pretty barren ground. You know, and I think the incentives right now are not aligned for the social good. I think we're looking at billions of dollars in lobbying being thrown by these big tech companies, a full-on media operations campaign that has been documented by tech industry adjacent folks to displace ethical concerns and concerns about the social harms of these systems with, you know, what I would call religious sci-fi fantasies about the singularity and about sort of the super intelligence. So, you know, we are outgunned in terms of lobbying power and in terms of the ability to put our weight on the decision makers in Congress. But where my hope lies for regulation is not with the kind of, you know, Athena birthing from the head of a senator and saying, like, actually you need to, you know, do the right thing but with things like the Writers Guild of America, who have, you know, I think done the best job we've seen of regulating AI, you know, just non-traditionally. They did the classic move, withholding their labor, and they got terms that are actually, you know, staunching the bleeding of the--you know, use by the studios and big tech to place AI within their labor process in ways that will degrade their labor, that will degrade artistic output, and that will actually, you know, I think have a real precedent-setting move in terms of stopping the real harms right now. So I would look to the Writers Guild of America. I would look to SAG. I would look to, you know, drivers' unions that are contesting the sort of automated precarity of AI systems like Uber and Lyft. I would look to sort of movements from below that are actually tackling the harms now and not simply sitting around and taking selfies with Elon Musk and calling it a regulatory agenda. [...] MS. PASSARIELLO: And, Meredith, you, of course, were--became notable in part because of your whistleblowing within Google around the ways that the technology was being used, of course--you know, and how their business incentives sort of dominated over moral and ethical concerns. Now, in this era of generative AI, they talk about being bold and responsible. And of course, they have been a little bit on the backfoot, you know, a little bit beat to the market by both OpenAI and Microsoft. How do you see their approach to ethics and morals versus the business balance these days? MS. WHITTAKER: Yeah, well, I did do labor organizing at Google, and that was one of the few things that actually checked some of these impulses. So, I think, you know, we can talk about business model. We can also talk about capitalism, right? The engines of these companies are driven by a need, a requirement to report revenue and growth increases every quarter forever. That's the definition of metastasis. And it is obviously not healthy for the social benefit. So, I think, you know, we do need those structural checks. I think--you know, how is Google doing--look, I don't--remember Web3, right? MS. PASSARIELLO: Vaguely. MS. WHITTAKER: Like, you know, this was a hype cycle. Everyone was predicting, you know, massive numbers. This is going to change the entire environment. And then, you know, no one's talking about it. Andreessen Horowitz has even moved off it. They're, you know, black holing their optimistic manifestos. I think generative AI is very similar. I don't think AI in general is similar. I think they're going to continue to create these large-scale models that involve data and compute. But generative AI is not actually that useful. What happened in January was that technology, or sort of a framework for building models that had been developed in 2017, was sort of put online with an interface by Microsoft/OpenAI, who have to be understood as the same entity, right? And the ChatGPT interface kind of gave people a simulated experience of like, oh, my God, I'm talking to kind of a human. It's spitting out nonsense, but it's spitting it out and this feels kind of sentient, right? And on the backs of this advertisement for their GPT API, which they sell through their Azure cloud services, they sort of generated an entire new hyped narrative around generative AI as the sort of future facing technology that's going to change every industry. But what does it do, right? It, you know, presents visual images that are often, you know, stolen from artists or like far too close for comfort. And it presents plausible text, right? It infers what's the sort of plausible response to a prompt, based on, you know, mountains of data from the internet, the Reddits, the 4chans. You know, the—Stormfront is in there, as Natasha's work have shown, you know, and kind of presents text that looks plausible, but has no relationship to facts, has no relationship to reality, has no citations, right? So, what is this useful for? It's not useful in most serious contexts. Yeah, you could, you know, replace a junior copywriter, but you better have a senior copywriter, who's checking that text because it's going to be janky. So, I think we need to be like, really clear about what are we actually responding to. We're responding to an advertisement, a very expensive advertisement, ChatGPT, that was put online as an interface that allowed us to have a sort of simulated experience with a bot that we're now sort of making all kinds of predictions on that I don't think are actually grounded in any understanding of the utility of these systems. And again, you know, Silicon Valley runs on VC hype. VCs require hype to get a return on investment, because they need an IPO or an acquisition, and that's how you get rich. You don't get rich by the technology working. You get rich by people believing it works long enough that one of those two things gets you some money https://www.washingtonpost.com/washington-post-live/2023/10/26/transcript-f…

Nov. 23, 2023

Re: [nexa] Supermarkets are ditching self-checkouts in a sign that we can push back against the technofuturist tide | Van Badham | The Guardian

by M. Fioretti

Boh... (continua sotto) On Thu, Nov 23, 2023 07:51:52 AM +0100, Alberto Cammozzo via nexa wrote: > The kiosks have saved corporate chains money on retail wages – but come at a > price for our shared sense of community > > <https://www.theguardian.com/commentisfree/2023/nov/23/ > supermarkets-are-ditching-self-checkouts-in-a-sign-that-we-can-push-back-against-the-technofuturist-tide > > > ... > One can safely presume the near-ubiquity of automated checkouts in > such supermarkets has done more to affirm fears of workplace > scrapheaping, destroyed human value and cold technological > outprogression than a thousand futurist summits. > ... > Beyond the pace of a harried transaction, customers could feel > comfortable to make conversation with their cashier. > ... > I’ll also take the possibility that rather than sink ourselves into > passive acceptance of a future determined by corporate algorithms > and ruthless greed, we, as a community, remind ourselves that we > invented the democratic state to assert a stake in the future being > created around us. Ripeto, boh. Tutto molto nobile, ma di fatto: 1, "destroyed human value and cold technological outprogression" ne hanno fatto in abbondanza proprio il passaggio dai negozietti di quartiere ai supermercati, la cassa automatica sarebbe il certificato di una morte gia' avvenuta. 2, sarebbe interessante vedere quanti di quelli che protestano per tutto 'sto calore umano perduto non parlando col cassiere accetterebbero prezzi piu' cari per aumentare la paga del cassiere medesimo, o gli stessi orari di apertura dei negozietti d'antan, che permetterebbero al cassiere di godersi il calore umano di famiglia e amici di sera e nei weekend. Senza le casse automatiche, cioe' con prezzi piu' alti, la gente che tenterebbe di rubare cibo perche' non ce la fa a comprarlo aumenterebbe 3, questo discorso e' uguale a quello che pagamenti, raccomandate e altro dovremmo continuare a farli manuali perche' signora mia, ormai quando mai ci capita di parlare con qualcuno se non in fila alla posta. Non esiste proprio che TUTTI dobbiamo essere COSTRETTI a continuare a perdere tempo alla posta o alle casse, volenti o nolenti, perche' senno' non c'e' tempo di parlare con nessuno. Il "valore umano" c'e' quando il TEMPO di rilassarci facendo due chiacchiere in casa o al parco DOPO lavoro e spesa ce l'abbiamo tutti, clienti e cassieri, non certo nel respingere la cassa automatica. Marco -- https://mfioretti.substack.com

Nov. 23, 2023

Supermarkets are ditching self-checkouts in a sign that we can push back against the technofuturist tide | Van Badham | The Guardian

by Alberto Cammozzo

The kiosks have saved corporate chains money on retail wages – but come at a price for our shared sense of community <https://www.theguardian.com/commentisfree/2023/nov/23/supermarkets-are-ditc…> British supermarket chain Booths is scrapping its self-service machines and replacing them with living, breathing, talking, thinking human cashiers. Hooray! In a world that seems to leap, minute by minute, from one dystopian scenario to another, this is happy news. Even better is that it’s not just Britain that’s trading in the automated misery chant of “unexpected items in the bagging area” for a trumpet of hope. CNN reports that major American chains including Costco, Walmart and Wegmans are also rethinking the loveless use of machines that can not tell an avocado from a banana no matter how loud you yell with frustration at it. We should seize on the slim window of optimism offered by the turning of this heavily technofuturist tide. Philosophers of futurism have increasingly embraced a dark view of the present human moment. There’s a belief that we’ve transitioned away from sensing we live in what they called the “Vuca” cultural moment (“volatile, uncertain, complex and ambiguous”) into something more bleak. American anthropologist Jamais Cascio describes our new shared emotional reality as the “Bani” paradigm; “brittle, anxious, non-linear and incomprehensible”. To translate this into broadly pop-cultural terms; the generations who found the 1970 Donald Cammell/Nicolas Roeg movie Performance disturbingly relatable have perhaps just woken up to find themselves playing Balthazar Getty’s character of Pete Dayton in David Lynch’s Lost Highway … but after a hard night on stimulants and a punch to the face. Community members, of course, visit their supermarket far more often than they do arthouse cinemas, and they have direct material experiences there. One can safely presume the near-ubiquity of automated checkouts in such supermarkets has done more to affirm fears of workplace scrapheaping, destroyed human value and cold technological outprogression than a thousand futurist summits. “Affirming human warmth” is not, alas, one of the two main reasons for the corporate self-checkout rethink. Instead, cost-of-living pressures in a post-pandemic greedflation environment has – not that anyone as morally pure as a supermarket chain really wants to admit it – encouraged shoplifting at the same time the machines are already making mistakes. The merchandise-loss phenomenon of combined theft, malfunction and customer error is known in the industry as “shrink”, and the data is long in that self-checkout appears to have doubled the shrink that occurs with human cashiers. The furious introduction of self-checkout kiosks that cost 75,000 retail jobs in Britain has demonstrably denuded American stores of enough staff to disincentivise stealing – structurally, it’s encouraged the opposite. “When even customers who want to pay for something struggle to flag down an employee,” the Atlantic wrote in September, the battle is lost. A recent study found 39% of theft in America’s grocery stores occurred at self-checkouts. The responses from some retailers in the wake of these studies reveal a corporate preference for dystopianism-of-the-will over efficiency in modern managerial practice; having retrenched their cashiers, some shops are now employing more security to surveil customers at the scanners or inspect receipts at the gates. Despite the expensive outlay to buy the machines, others are now extending their investment in yet more technology to monitor customers – including the use of AI systems to identify any disruption to expected patterns of behaviour. But others have, indeed, abandoned the experiment and embraced providing customers with a retail experience that customers again and again have stated they prefer. Consumers are right to be suspicious; the promises of the 1970s free-marketeers that competing businesses would be obliged to follow forces of customer sentiment were quickly forsaken for cabal-like corporate conformism. But if the wretched self-checkouts are being ripped out of the shops, I don’t care about the motivations; I’ll take it. I’ll also take the possibility that rather than sink ourselves into passive acceptance of a future determined by corporate algorithms and ruthless greed, we, as a community, remind ourselves that we invented the democratic state to assert a stake in the future being created around us. The Dutch government was acting on democratic imperative amid demographic challenges of rising megacities, trends of overwork and technology’s reduction of human face-to-face interaction to combat social isolation and a “loneliness epidemic” affecting social health and wellbeing. In 2019, as part of that government’s “One Against Loneliness” campaign, Dutch supermarket chain Jumbo trialled a “chat checkout”, or “kletskassa” lane in a Brabant store. Beyond the pace of a harried transaction, customers could feel comfortable to make conversation with their cashier. The target was the isolated elderly. Four years later, Jumbo have expanded kletskassa lanes to 200 stores because the trial revealed they appealed across all age groups – and employees gained meaning and purpose from staffing them. A rewarded Jumbo are now also providing “Coffee corners” in various locations for people to meet and chat with other locals and connect with community volunteers and support services. It’s a far cry from the corporate-surveillance state nightmare of AI spying on suspected shoplifters at the beep-machines, and a bold rebuke to prophets who insist on inevitable, universal automation. Ditching the self-checkout for the old-fashioned chat lane shows us that the forces shaping the future are subject to brittle, anxious, non-linear and, perhaps, yes, incomprehensible pressure. And, maybe for humanity, that’s a positive thing.

Nov. 23, 2023