- nexa - server-nexa.polito.it

Fw: TR : INENT - LAW Research Seminar - Toshiko Takenaka, Atsushi Sato and Yoshiyuki Inaba
by Cristiana Sappa March 8, 2024

March 8, 2024

Buongiorno a tutti, Sperando di fare cosa gradita, trovate qui di seguito le informazioni per il seminario del 15 marzo alle ore 15h00 su: Patents in the generative AI era. E' possibile partecipare online o in presenza (nel qual caso Vi pregherei di farmi sapere entro giovedi' 14 marzo). Un saluto cordiale, Cristiana Sappa Oggetto: TR : INENT - LAW Research Seminar - Toshiko Takenaka, Atsushi Sato and Yoshiyuki Inaba De : research-seminars <research-seminars(a)ieseg.fr> Sujet : INENT - LAW Research Seminar - Toshiko Takenaka, Atsushi Sato and Yoshiyuki Inaba Quand : vendredi 15 mars 2024 15:00-16:30. Où : P400 + Zoom Zoom link : https://ieseg.zoom.us/j/84796394437

1 0

AI Act 1.01 ciclo di seminari online
by Antonio Vetro' March 8, 2024

March 8, 2024

Segnalo questa iniziativa all’interno del progetto FAIR: Ciclo di webinar "AI Act 1.01", organizzato dal PT1 di FAIR in previsione della prossima approvazione dell’AI Act. Gli incontri avranno luogo online i prossimi 22 marzo, 5 aprile, 11 aprile e 19 aprile (ore 17-18.30). Il fine dell'iniziativa è introdurre anche a non giuristi il regolamento e chiarirne i principi base e la struttura.  Gli incontri sono pensati secondo un approccio incrementale, avranno un taglio agile e ampio spazio sarà lasciato per la discussione. Programma completo: https://webmagazine.unitn.it/fileswebmagazine/download/119862/aiact101.pdf A.

1 0

"Time for a Pause: Without Effective Public Oversight, AI in Schools Will Do More Harm Than Good."
by J.C. DE MARTIN March 6, 2024

March 6, 2024

Time for a Pause: Without Effective Public Oversight, AI in Schools Will Do More Harm Than Good. Ben Williamson <https://nepc.colorado.edu/author/williamson-ben>, Alex Molnar <https://nepc.colorado.edu/author/molnar-alex>, and Faith Boninger <https://nepc.colorado.edu/author/boninger-faith> March 5, 2024 Publication Announcement <https://nepc.colorado.edu/node/13145> Press Citations <https://nepc.colorado.edu/publication/13144/citations> *Learn More:* NEPC Resources on Computing, Technology, and Information Systems <https://nepc.colorado.edu/topics/33> Ignoring their own well-publicized calls to regulate AI development and to pause implementation of its applications, major technology companies such as Google, Microsoft, and Meta are racing to fend off regulation and integrate artificial intelligence (AI) into their platforms. The weight of the available evidence suggests that the current wholesale adoption of unregulated AI applications in schools poses a grave danger to democratic civil society and to individual freedom and liberty. Years of warnings and precedents have highlighted the risks posed by the widespread use of pre-AI digital technologies in education, which have obscured decision-making and enabled student data exploitation. Without effective public oversight, the introduction of opaque and unproven AI systems and applications will likely exacerbate these problems. This policy brief explores the harms likely if lawmakers and others do not step in with carefully considered measures to prevent these extensive risks. The authors urge school leaders to pause the adoption of AI applications until policymakers have had sufficient time to thoroughly educate themselves and develop legislation and policies ensuring effective public oversight and control of school applications. *Suggested Citation*: Williamson, B., Molnar, A., & Boninger, F. (2024). /Time for a pause: Without effective public oversight, AI in schools will do more harm than good/. Boulder, CO: National Education Policy Center. Retrieved [date] from http://nepc.colorado.edu/publication/ai <https://nepc.colorado.edu/publication/ai> https://nepc.colorado.edu/publication/ai

1 0

Berlin blames Taurus call leak on officer logging in via insecure Singapore hotel line
by 380° March 5, 2024

March 5, 2024

Buongiorno, spiegato l'arcano di una delle più importanti intercettazioni della storia recente. Sotto trovate il testo della notizia, ma prima metto un executive summary: il ministro Pistorious può così _frignare_ "Our communication systems were not compromised", ma poter instaurare a una call su Cisco WebEx [1] senza l'obbligo di usare la crittografia E2EE è una falla di sicurezza grande come una montagna: --8<---------------cut here---------------start------------->8--- Known limitations Transcoding, Automatic Closed Captioning, Transcription, PSTN, and other cloud-based services that require the cloud to access the media are not available at this time, as they are not supported by the Zero-Trust Security model for End-to-End Encryption v2 (E2EEv2). A participant joining from a Webex device must be one of the first 205 participants of any kind in the meeting, or their connection will require transcoding. As this is not supported, they will not be admitted to the meeting. The maximum number of participants in an E2EEv2 meeting is 1000. SIP video or telephone devices cannot join E2EEv2 meetings, as E2EEv2 is not available in the SIP protocol. --8<---------------cut here---------------end--------------->8--- (via [2]) Cioè in una conversazione E2EE *non* si può avvedere via PSTN (o SIP), come pare abbia fatto uno dei partecipanti. Non ho mai usaro quel servizio ma da quello che ho capito è "il moderatore" della call su WebEx che *deve* verificare lo stato della crittografia E2E con una semplice procedura [3]: --8<---------------cut here---------------start------------->8--- If you want to verify the security of the end-to-end encryption connection, communicate the security code to all participants, either verbally or in a message outside the Webex App. Ask all participants if they see the same code in the meeting. If any participants don’t see the same security code, ask them to leave the meeting and join again, and then verify that their security code matches the current code. The current code you see may be an updated one. If there are any issues, contact your administrator. --8<---------------cut here---------------end--------------->8--- Il traffico di quella conversazione era *in chiaro* su quella linea PSTN (SIP?) di Singapore, un gioco da bambini dell'asilo intercettarla. https://www.politico.eu/article/german-defense-minister-blames-taurus-call-… --8<---------------cut here---------------start------------->8--- Berlin blames Taurus call leak on officer logging in via insecure Singapore hotel line ═══════════════════════════════════════════════════════════════════════════════ Germany is scrambling to clean up the damage following an embarrassing leak of secret information to the Russians. The audio of German military officers discussing top secret information was intercepted by Russia only because one of them logged in through an insecure line from a hotel room in Singapore, Defense Minister Boris Pistorius said Tuesday. Addressing the leak in Berlin following an investigation by Germany's Military Counterintelligence Service, Pistorius insisted the incident was a one-off. "Our communication systems were not compromised," he said. On Friday, Russia's state-run media outlet Russia Today released details of a 38-minute call between senior German officers — including the chief of the air force — in which they discussed the hypothetical dispatch of Taurus cruise missiles to Ukraine as part of preparations for a meeting with Pistorius. Chancellor Olaf Scholz is opposed to sending the missiles, and the issue has splintered his governing coalition. One of the participants — [reported] to be Brigadier General Frank Gräfe — dialed into the WebEx call from a hotel room in Singapore where he was visiting an airshow. Over the weekend, the government [confirmed] the veracity of the recording. Pistorius on Sunday [called it] a "hybrid disinformation attack" by Russia, although the recording published by Moscow was not faked. [...] [Joshua Posaner] <https://www.politico.eu/author/joshua-posaner/> [reported] <https://www.politico.eu/article/the-russian-mfa-summoned-germanys-ambassado…> [confirmed] <https://www.politico.eu/article/germany-investigating-leak-in-russia-of-aud…> [called it] <https://www.politico.eu/article/german-defense-minister-accuses-moscow-of-a…> [Kremlin is celebrating] <https://www.politico.eu/article/the-russian-mfa-summoned-germanys-ambassado…> [according] <https://tass.com/politics/1755191> --8<---------------cut here---------------end--------------->8--- Saluti, 380° [1] https://en.wikipedia.org/wiki/Cisco_Webex [2] https://help.webex.com/en-us/article/5h5d8ab/End-to-end-encryption-with-ide… [3] https://help.webex.com/en-us/article/rzam8/Join-a-Webex-Meeting-with-End-to… -- 380° (Giovanni Biscuolo public alter ego) «Noi, incompetenti come siamo, non abbiamo alcun titolo per suggerire alcunché» Disinformation flourishes because many people care deeply about injustice but very few check the facts. Ask me about <https://stallmansupport.org>.

2 1

‘Musk needs to be adored … Zuckerberg is out of his depth’: Kara Swisher on the toxic giants of Big Tech | Technology | The Guardian
by Alberto Cammozzo March 4, 2024

March 4, 2024

‘Musk needs to be adored … Zuckerberg is out of his depth’: Kara Swisher on the toxic giants of Big Tech <https://www.theguardian.com/technology/2024/mar/04/musk-needs-to-be-adored-…> John Harris “This is about love gone wrong,” says Kara Swisher, looking back on a life spent studying the giants of Big Tech. “I saw the possibilities of tech being the saviour of humanity – or at the very least, really helping people, in terms of community and knowledge and education. And instead, you know …” She pauses, and wearily exhales. “It’s like that old expression: ‘They promised us jetpacks, and this is what we got?’ Like, are you kidding me?” But, she adds: “The problem isn’t tech. It’s people.” Swisher is essentially a business journalist, but her speciality is human beings and what they do with wealth and power. She has been scrutinising Silicon Valley for around three decades – writing ferociously and insightfully for the Washington Post, the Wall Street Journal and the New York Times, founding the tech news website Recode, and blazing a trail into podcasting. Her writing has always struck a delicate balance between insider knowledge and biting irreverence. Now, though, her iconoclastic side has won out, given free rein in an extremely readable memoir, Burn Book. The title, she tells me, comes from the 2004 movie Mean Girls, in which the leading characters keep a shared diary full of slights and gossip about their classmates. As well as telling her own story, the book centres on pen-portraits of people – men, mostly – she has closely observed as their wealth and influence has ballooned: Bill Gates, Steve Jobs, Mark Zuckerberg, Jeff Bezos, Elon Musk, and Sergey Brin and Larry Page, the founders of Google. A few of them emerge as rounded, deep-thinking people who at least understand the huge questions of power that swirl around them. Others, by contrast, are more cuttingly portrayed – as “fresh-faced wunderkinds I had mostly rooted for” who eventually made the author “feel like a parent whose progeny had turned into, well, assholes”. Swisher talks to me on video call from her home in Washington DC, but she still has a house in San Francisco, and says she spends as much time as she can there. And she insists, again and again, that she still has faith in tech’s liberating, knowledge-enhancing potential: “I still believe in it. Stuff around generative AI, for example: I’ve talked to so many doctors and they’re like: ‘This could change the whole game on cancer entirely.’ Or climate change technologists who say: ‘We can now really begin to understand solutions to this.’” She also says she remains a believer in what she calls “the most vaunted parts of the American experience”: the US system of democracy, and the idea that anyone can become a success (“It’s full of hypocrisy, and at the same time it’s very true”). But in Burn Book, these two articles of faith are sorely tested by the moment she uses to symbolise tech’s spectacular fall from grace: 14 December 2016, when many of the executives and company founders she had on speed-dial met the newly elected president on the 25th floor of Trump Tower, in New York. Even now, she talks about this in the pained manner of someone recalling eating rotten food: it was a hugely significant moment, in the worst possible way. “My issue, if you can believe it, wasn’t really with Trump,” she says. “I get him. He’s very easy to read. He’s a racist, he’s a homophobe – he’s everything terrible about the United States. My issue was with the tech people who knew better, didn’t like him, but wanted money. More money. They wanted less regulation and the ability to grow, unrestricted. They knew he was bad. He was very anti-immigration, and this was an industry built on immigration. All I wanted them to do was say something publicly, like: ‘We’re going, but let me tell you, Mr president-elect, we are going to do everything to fight you on your immigration statements.’ They could have done anything. But they snuck in. They really did.” She then mentions the pithy statement of supposed corporate ethics still enshrined in Google’s code of conduct. “‘Don’t be evil.’ Well, evil’s sitting in front of you. You might want to have a word.” Swisher is 61, a bit older than most of the tech bros she has spent so much time chronicling. The sense of an insider-outsider is compounded by her upbringing in Long Island, which seems to have leavened her embrace of California optimism with a very north-eastern kind of sarcasm and scepticism. In a world still dominated by straight white men, moreover, her gender and sexuality also set her apart, having played a big role in her pre-tech backstory. Her initial ambition, she says, was to follow the example set by her father, who had served in the US Navy, and work as a strategic analyst for either the military or the CIA. But as she writes, “pushing against the anti-gay tide was nearly impossible at the time, and the ferreting out of gays in the military continued for over a decade”. She began making her way in journalism. By 1996, she was dedicatedly reporting about the new world taking shape in northern California, and what she calls “supremely odd but compelling people”. “Like the Google guys – they always had strange clothes and they said odd things to you and … wandered away,” she says. “They would have kind of weird and wacky headquarters. A lot of toys, which actually was a signal to me, and not a good one. I was like: ‘What are they doing?’ The idea of being childlike – they loved that. A lot of their clothes were very juvenile. So was their food. There’d be, like, pogo sticks. A lot of graffiti on the walls, but paid-for graffiti. And ping-pong tables.” In the book, these regressive tendencies reach their jaw-dropping nadir at the baby shower held in 2008 by Google co-founder Brin and his wife. Guests, Swisher recounts, could wear “a diaper with an oversized comical pin”, “a ruffled baby hat that came with a rattle” or “adult-sized footy pyjamas accessorised with a teddy bear and a sucker”. She refused all the sartorial options, but soon encountered Wendi Deng – then the wife of Rupert Murdoch, who Swisher calls “Uncle Satan” – wearing “leather pants and stiletto boots under the giant Pampers”. All this weirdness, Swisher says, was there, whether consciously or not, to smooth over the fact that the new tech industry was not the big-hearted humanitarian project its founders often talked about, but something much more straightforward: the latest iteration of rapacious capitalism. “I thought it was all performative. It was like: ‘Aren’t we different?’; and I was like: ‘You’re not really that different.’ I was irritated by the performative nature of it all, you know: all soft and squishy, but hard as nails on the inside. And that’s what these people were, right? They were always killers. Every one of them.” This leads on to another of her book’s big themes. If people like this acquired any degree of power, they were probably always going to be out of their depth – a point that seems to apply particularly vividly to Zuckerberg. When she first met the-then boss of Facebook, she was struck by the fact that he looked “like a newborn something, all fawn-like eyes and wide forehead”, and was painfully socially awkward. But he also “craved power and historical significance from the get-go”. Does she think he ever feels any fear about the huge responsibilities that ought to come with what he has built? She answers emphatically. “Yes. I think every now and then you see that he knows he’s in over his head.” She mentions an interview she did with Zuckerberg in 2018, for her Recode Decode Podcast, in which he expressed the somewhat startling opinion not only that Facebook (since folded into the giant company that Zuckerberg named Meta) should host content put up by Holocaust deniers, but that such people were “not intentionally getting it wrong”. “He was so out of his depth, and you could see that he kind of knew it,” she says. “But he kind of walked into it: like: ‘I can handle this.’ I’m like: ‘You cannot handle hundreds of years of antisemitism – I’m telling you that you can’t. You need some real experience.’ And that was my issue: someone who was so ill-prepared was making decisions that affected all kinds of people and unleashed an enormous amount of toxic waste.” She describes his recent appearance in front of the judiciary committee of the Senate, also attended by parents who had lost their children to suicide after horrific experiences online. “They were there,” Swisher tells me, “and they had pictures of their kids. They held them up, these photographs. I saw him look at the entire group. And you could see – like: ‘Oh my God.’ But he couldn’t bring himself to apologise directly. He had to say: ‘I’m sorry for what was done to you.’ And that’s not an apology. I don’t know what that is.” She pauses. “It’s an acknowledgment of pain.” What does that say about him? “It says he still can’t take responsibility. You could see in his eyes that he was just like: ‘Whoa, whoa, whoa.’” He could see the impact. But he couldn’t say: ‘I’m sorry.’ They were saying: ‘Things you did directly, decisions you made, helped kill my child.’ And he couldn’t address that. It’s so passive.” Meta’s president of global affairs, let us not forget, is the former Lib Dem leader Nick Clegg. Has she had any dealings with him? “Yes. Of course. Nick. Smoothie Nick. I’ve never interviewed him: he won’t give me an interview because he’s smart. I just feel like: ‘Do you believe anything you’re saying?’ I don’t know. He’s such a pretty sayer of things. I’m sort of like: ‘What do you actually believe in here? What is your goal?’ That’s what I often think of with him.” In the book, Swisher says Zuckerberg is “the most damaging man in tech”. Elon Musk, by contrast, is maligned as the “most disappointing”, which reflects Swisher’s long period of thinking of the founder of Tesla and SpaceX as one of the tech industry’s most promising sons. In 2016, she contacted him ahead of the big meeting with Trump, warning that the president-elect would “screw” him; two years later, Musk told her she had been right. All told, she seemed to believe that he operated on a higher level than most of his peers. “Here’s someone who actually was doing serious things,” she says. “There’s a lot of people in Silicon Valley who are always doing a dry cleaning app. He was thinking of everything from cars to space to solar. Even the silly stuff like [his imagined high-speed transport system] Hyperloop: what a great idea. What an interesting idea.” She also mentions Neuralink, the venture working on computer interfaces that can be implanted in people’s brains. “How could we upgrade our intelligence? That’s a big, fascinating problem.” Initially, she thought his ownership of X would be a good thing: “He used the product and he understood it. And he’s a guy who got things done.” But then came a rupture: in October 2022, she tweeted about a Washington Post article that questioned his work on the satellite-based internet system Starlink, and he emailed her the eloquent words: “You’re an asshole.” Soon after, he began opening X up to the kind of far-right voices that had been banned, and plunging into conspiracy theories – as she puts it, “going off the rails every day of the week and twice on Sunday”. She says she is saddened by all this. “The drug that’s hurting Elon Musk is all his enablers, who suck up to him. And needing to be adored. That’ll kill you.” And she watched it happen, tweet by tweet: “You could see him getting more and more radicalised. He suddenly got obsessed with the woke mind virus, whatever the hell that is, and angry all the time.” Given its huge financial problems, does she think X will eventually fold? “Why would it? He’s the world’s richest man. It’s like having a mega-yacht: it’s expensive, but he’s got the money to keep it up. And they’ll give him loans … so it will only go down when he decides to stop paying for it.” Does he think he will? “No! I think he loves it. He desperately needs attention. Trump is running for president because he desperately needs attention, among other things. And Musk really needs attention. If he owns X, he becomes quantumly more interesting to people across the globe.” Which brings us to one of the biggest questions of all. As that summit with the titans of Big Tech proved, Trump is arguably the quintessential politician of the internet age. Whether Musk explicitly supports him or not, X’s reshaping as something of a right-wing hellscape will be a sizable help to his re-election campaign. Come November, does Swisher think he’ll win? “No. No. This is where I believe in the American people. There’s an angry strain through the American experience that’s never left us. We’re an angry people, in many ways. But we’re also hopeful people. And people have had enough of him, especially women. “They’re tired of being groped. They’re tired of being cheated. They’re tired of loudmouths. Biden is not the ideal candidate, but he’s a decent man and he’s done a lot of stuff, and it will sink in enough to get rid of Trump.” Swisher has not spent the last 90 minutes mincing her words, and so it proves again. “He might be in jail, and I hope he is,” she says. “I hope they put him away and throw away the key.” Burn Book by Kara Swisher (Piatkus, £25) is out now. To support the Guardian and the Observer, buy a copy at guardianbookshop.com. Delivery charges may apply.

1 0

170° Mercoledì di Nexa | 13 marzo 2024, ore 17.00
by Nexa - Media March 4, 2024

March 4, 2024

Gentilissime, gentilissimi, Vi invitiamo a partecipare al 170° Mercoledì di Nexa, che si terrà mercoledì 13 marzo, alle ore 17.00, con un incontro dal titolo "A fin di bene: il nuovo potere della ragione artificiale". Ospite dell'incontro: Stefano Isola (Università di Camerino). L'incontro si terrà IN PRESENZA e ONLINE. SEDE FISICA dell'incontro: Centro Nexa su Internet e Società, Politecnico di Torino, Via Boggio 65/a, Torino (1° piano). Per accedere alla sala si raccomanda di suonare al citofono Portineria e di seguire le indicazioni segnalate lungo il percorso. QUI<https://nexa.polito.it/contatti> maggiori informazioni su come raggiungerci. STANZA VIRTUALE dell'incontro: https://didattica.polito.it/VClass/NexaEvent Di seguito maggiori dettagli: https://nexa.polito.it/mercoledi-170 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/ mail: valeria.bergantino(a)polito.it<mailto:valeria.bergantino@polito.it> tel: 3473443585

1 0

"Posta Elettronica" e Atenei: la situazione è peggiore di quanto sembrasse...
by Damiano Verzulli March 4, 2024

March 4, 2024

Lo scorso dicembre segnalavo in lista [1] un aggiornamento di un mio vecchio censimento dal quale emergeva che negli ultimi tre anni 6 atenei avessero abbandonato il servizio di posta on-prem, favorendo la soluzione cloud. Dovendo discutere dell'argomento in un contesto per me importante, ho voluto verificare meglio lo stato di quegli Atenei che, ai miei occhi, risultavano ancora "on-prem". *IL RISULTATO È STATO AGGHIACCIANTE* Oltre ad UniMilano (della quale gia' sapevamo che la migrazione era in corso) ed ad un mio errore/refuso sulla Scuola Universitaria Superiore Sant'Anna, di Pisa (per la quale avevo erroneamente verificato il dominio "sssup.it" anziché l'ufficiale "santannapisa.it", chiaramente gestito da Microsoft), ho riscontrato che altri 7 Atenei, inclusi POLITO e UNIPI, sono passati a Microsoft. Di fatto, dalla mia analisi, attualmente emerge che soltanto 4 Atenei gestiscono la posta internamente: * *uniba.it* - Università degli Studi di Bari Aldo Moro * *unich.it* - Università degli Studi Gabriele d’Annunzio Chieti Pescara * *unina.it* - Università di Napoli Federico II * *uniroma2.it* - Università di Roma Tor Vergata La situazione dettagliata è sempre disponibile al vecchio indirizzo. Mi chiedo (...e vi chiedo) se questo scenario è ormai da considerare irrimediabilmente compromesso (e quindi è il caso di supportare lo spegnimento di queste ultime 4 infrastrutture), oppure se ci possano essere le condizioni per considerare questi ultimi 4 atenei una sorta di "avanguardia" da cui ripartire.... La domanda nasce dal fatto che... potrei avere un qualche ruolo nel determinare il futuro di quella relativa ad "unich.it"... Un saluto, DV [1] https://server-nexa.polito.it/pipermail/nexa/2023-December/052038.html [2] https://dvblog.soabit.com/la-posta-elettronica-negli-atenei-italiani/ -- Damiano Verzulli e-mail:damiano@verzulli.it --- possible?ok:while(!possible){open_mindedness++} --- "...I realized that free software would not generate the kind of income that was needed. Maybe in USA or Europe, you may be able to get a well paying job as a free software developer, but not here [in Africa]..." -- Guido Sohne - 1973-2008 http://ole.kenic.or.ke/pipermail/skunkworks/2008-April/005989.html

7 8

Re: [nexa] LMs and AI make software development harder
by Giuseppe Attardi March 4, 2024

March 4, 2024

Questo signore si dichiara Agilist, una delle metodologie moderne di sviluppo software, ma sembra non aderire ad alcuni suoi principi fondamentali, quali: 1. Pairwise programming: qui un copilot funge benissimo da secondo nella coppia 2. Deliver working software frequently: il che significa dedicare gran parte del tempo a scrivere codice — Beppe > On 29 Feb 2024, at 12:32, nexa-request(a)server-nexa.polito.it wrote: > > From: Daniela Tafani <daniela.tafani(a)unipi.it <mailto:daniela.tafani@unipi.it>> > To: "nexa(a)server-nexa.polito.it <mailto:nexa@server-nexa.polito.it>" <nexa(a)server-nexa.polito.it <mailto:nexa@server-nexa.polito.it>> > Subject: [nexa] LMs and AI make software development harder > Message-ID: <1ff9c3d20cd643eabb1bbe07a054d2ba(a)unipi.it <mailto:1ff9c3d20cd643eabb1bbe07a054d2ba@unipi.it>> > Content-Type: text/plain; charset="Windows-1252" > > LLMs and AI make software development harder > > LLMs and AI make software development harder. Wait, what? Isn’t the whole point of AI to make writing code easier? Well, yes. But writing code is the easy part of software development. The hard part is understanding the problem, designing business logic and debugging tough bugs. And that’s where AI code assistants like copilot or chatgpt make our job harder, as they strip a way the easy parts of our job and only leave us with the hard parts and make it harder for new developers to master the craft of software development. > Coding is the easy part? > > Is coding really that easy? No, not exactly easy - mastering a programming language still takes years of practice. But when looking at software development as a whole, writing code is one of the easier part and it is no wonder that chatgpt and copilot can write decent code. First, they have trained on millions of lines of code and second, code is by its nature very easy to understand for a machine as programming languages are very structured languages with limited vocabularies. For a LLM it is probably much easier to learn than natural language. > > Programming languages are just very powerful tools that we use to solve problems > > In the end, programming languages are just very powerful tools that we use to solve problems. And the hard part is not the learning tool, but understanding the problem and designing a solution for it. This is instantly obvious as most software engineering problems could be solved by a lot of different programming languages, which one to pick is a matter of context or even personal preference. > > Another indicator that programming is that easy part is, that the more senior a software developer gets, the less time they usually spend writing code. Instead seniors spend more time understanding the problem, designing the solution, jumping in to debug tough bugs or doing design decisions and of course mentoring junior team members. While this might not be true for every senior developer, when looking at my software development bubble this is a clear trend. > The hard parts of software development > > Copilot and other AI assistants are a great help for developers, but they are not flawless. A part of it is natural, as they are trained on existing code without any context and there are also some bad habits from the training data that code assistants might have picked up. And while this might get optimized over time, at the moment it means that developers still have to review the code that is generated by the AI code assistants. And reviews are hard - especially if one cannot query the author of the code for their intent. > > And even if the code is good enough, it might still introduce flaws into the control logic of a program, might be missing edge cases or introduce a regression bug when integrated into an existing code base. This means that developers have to debug the code that is generated by the AI code assistants in case of an error. And debugging is hard - especially for these kind of problems where it might be hard to recreate the circumstances that cause the bug in the first place. > > As the generated code heavily depends on the context we give the AI code assistants, this means that we have to be very precise in our descriptions which means that we have to understand the problem very well, which requires domain knowledge and context awareness on the side of the developer. Even if we just focus on the technical part, being aware of the surrounding architecture and the existing code is crucial to get good results. > > Granted we could ask LLMs like chatgpt for help with integration into the codebase or we could just pass it the whole codebase and let it redesign everything. But apart from requiring lot of input to give enough context debugging in an unfamiliar codebase is even tougher than debugging stuff that you wrote yourself. > > And then there is the whole thing about figuring out what exactly our product should do, how it should behave and how it should look like. At the moment this still requires a lot of human smarts and while AI tools might allow us to iterate faster on figuring out what we want to build in the end it is still a human that has to make the decision. > AI generated software development is exhausting > > It seems a given that AI assistants will change our job by automating away writing code and even helping us with some design decisions. It is very convenient that we can ask chatgpt questions regarding system design and get reasonable answers. What is still left to us is making the decision on which answer to pick and which prompt to give to the LLM to get the results that we need. And this is very exhausting - decision fatigue is a thing and it is very real. Already before AI code assistants the limiting factor in the speed of delivering software was not the often the decision making process of an organization or a team - not writing the code. > > The limiting factor in delivery speed is decision making, not writing code > > On top of that is that current company structures will most likely still hold software developers accountable for the code that is running in a product, not the AI code assistants that wrote them in the first place. This will add another layer of stress on it, not just do we need to make more decisions faster, we are also to blame if the AI code assistants make a mistake. > > And if there is a mistake then the debugging needs to be done, which often needs a lot of context and background knowledge to be efficient. AI tools are of less help there, because they cannot figure out context changes by themselves. They might help us with the easy part of debugging like running tests with different variations, to narrow down the cause but finding the prompt for an LLM to generate the fix will still be on us. > Are AI tools replacing developers? > > AI assistants might lower the initial hurdle to get into software development, but they will not make it easier to become a good, experienced software developer. Most of the senior developers I know gained the background knowledge and context needed to formulate complex solutions from years of slogging through (bad) code and learning from their mistakes. This might be an inefficient way of learning but it is very effective in building up the domain knowledge that is needed for software development. This knowledge is also something that is very hard to teach in a formal way, as books or online tutorials by nature are somewhat generic and and adaption to real life situations still needs hands-on experience. > > As I see it, broad usage of AI tools will change the the skill distribution of software developers. We might end up with a lot more junior developers that are able to write code - or at least prompt the LLMs write the code - but lack the deep understanding of software development to be efficient in decision making. On the other hands senior developers that have acquired the context and domain knowledge will be fewer and fewer as the effort to acquire this knowledge will be higher as AI tools will hide away the parts that would enable us to learn unless the generated code is reviewed in-depth, which then raises the question if we gain that much efficiency through the tools at all. > > So are AI tools replacing developers? Currently no, they will transform the job of a developer but they will not replace them. The question is how we as an industry will make sure that we retain the knowledge and experience that we have gained over the years. It will also raise the question how we handle the human side of software development, as the job will either become more boring because we just feed machines with prompts, yet more stressful because we have to make more hard decisions faster. Or maybe AI tools are really just a hype and a fad and nothing will change at all. > Written on November 3, 2023 > > > https://dominikberner.ch/ai-tools-make-our-job-harder/

2 1

White Paper - Rethinking Privacy in the AI Era: Policy Provocations for a Data-Centric World
by Daniela Tafani March 3, 2024

March 3, 2024

White Paper February 22, 2024 Rethinking Privacy in the AI Era: Policy Provocations for a Data-Centric World Jennifer King, Caroline Meinhardt Executive Summary ➜ In this paper, we present a series of arguments and predictions about how existing and future privacy and data protection regulation will impact the development and deployment of AI systems. ➜ Data is the foundation of all AI systems. Going forward, AI development will continue to increase developers’ hunger for training data, fueling an even greater race for data acquisition than we have already seen in past decades. ➜ Largely unrestrained data collection poses unique risks to privacy that extend beyond the individual level—they aggregate to pose societal-level harms that cannot be addressed through the exercise of individual data rights alone. ➜ While existing and proposed privacy legislation, grounded in the globally accepted Fair Information Practices (FIPs), implicitly regulate AI development, they are not sufficient to address the data acquisition race as well as the resulting individual and systemic privacy harms. ➜ Even legislation that contains explicit provisions on algorithmic decision-making and other forms of AI does not provide the data governance measures needed to meaningfully regulate the data used in AI systems. ➜ We present three suggestions for how to mitigate the risks to data privacy posed by the development and adoption of AI: 1. Denormalize data collection by default by shifting away from opt-out to opt-in data collection. Data collectors must facilitate true data minimization through “privacy by default” strategies and adopt technical standards and infrastructure for meaningful consent mechanisms. 2. Focus on the AI data supply chain to improve privacy and data protection. Ensuring dataset transparency and accountability across the entire life cycle must be a focus of any regulatory system that addresses data privacy. 3. Flip the script on the creation and management of personal data. Policymakers should support the development of new governance mechanisms and technical infrastructure (e.g., data intermediaries and data permissioning infrastructure) to support and automate the exercise of individual data rights and preferences. https://hai.stanford.edu/white-paper-rethinking-privacy-ai-era-policy-provo…

1 0

Vi racconto com'è “vendere” la propria voce per addestrare un'intelligenza artificiale
by don Luca Peyron March 2, 2024

March 2, 2024

Tra costume e futuro https://www.wired.it/article/intelligenza-artificiale-addestramento-voce-fr… Buona lettura dl _________________________ don Luca Peyron Pastorale Universitaria - Apostolato Digitale Arcidiocesi di Torino www.universitari.to.it via XX settembre 83, Torino tel. 011 5156239

2 1