July 2025 - nexa - server-nexa.polito.it

Blocco all’HTML
by Giuseppe Attardi Aug. 4, 2025

Aug. 4, 2025

A me capita di frequente, ma vedo succede anche ad altri, che il contenuto di un messaggio venga “scrubbed” perché contiene HTML e girato come attachment. Quando apro l’attachment (quando va bene, in questi giorni mi dà Error 404), mi spara una tonnellata di HTML pieno di tag in cui il testo, che magari sono poche righe, è totalmente indecifrabile. Ciò succede involontariamente, perché ormai oggi tutti i mail reader usano HTML come formato. Se si fa un cut/paste di un testo apparentemente innocuo, anche solo di un link, tutto il messaggio viene trasformato in HTML, senza che uno se ne renda conto. Ormai qualunque lettore di posta è in grado di interpretare HTML, quindi direi che la restrizione sia obsoleta e non vedo ragioni per continuare a bandirlo. Vorrei quindi fare una richiesta ufficiale ai gestori della lista nexa perché eliminino questo limite. Per tenere conto di chi usa lettori di puro testo, si può sempre fare alla rovescia, ossia mettere un link alla pagina web del messaggio per poterlo leggere con un browser. Chi è d’accordo, risponda con un semplice +1. Grazie Giuseppe Attardi

12 25

"Stefano Rodotà (appositamente richiamato dall’Aldilà mentre giocava a scacchi)"
by Marco Fioretti Aug. 3, 2025

Aug. 3, 2025

Tutti quelli che sanno di che sto parlando mettano il dito qua sotto, perche' son sicuro di non averlo ricevuto solo io fra gli iscritti a questa lista. E se qualcuno che ha seguito la questione da vicino ha gia' avuto tempo di leggerlo, grazie in anticipo per un commento. Marco

15 44

Re: [nexa] GDPR could protect us from the AI Act
by Giuseppe Attardi July 31, 2025

July 31, 2025

Ho sempre sostenuto e contestato direttamente anche a voce con i relatori dell’AI Act, che l’impostaziine della norma fosse sbagliata. Invece di definire dei diritti, come stabilisce la GDPR, indipendentemente dal modo con cui questi diritti possano essere violati, l’AI Act ha un approccio “risk based”, in cui si elencano le applicazioni che producono danni. Questo comporta complicazioni burocratiche come l’introduzione dell’AI Office per mantenere aggiornato l’elenco. Sono quindi pienamente d’accordo con questa critica: The AI Act's approach to the protection of fundamental rights is based on a profound misconception of the law: for the law to protect a right means to protect against the violation of that right, as Mireille Hildebrandt reminds us, and harm is by no means a condition for the violation of a fundamental right. Under the AI Act's harm approach to fundamental rights impact assessment, fundamental rights can be violated with impunity as long as there is no foreseeable harm. Requiring only a risk impact assessment is like being satisfied with asking whether a despot is benevolent or malevolent: freedom, understood as the absence of domination, is lost whatever the answer. Purtroppo ogni volta che ho sollevato questa obiezione di fronte a dei giuristi, tutti hanno sminuito la questione, dicendo che dietro ogni legge ci sono dei diritti. Peccato che quei diritti non siano espressi né indicati esplicitamente nell’AI Act. Quindi come si fa a fare capire questo errore originale di impostazione di una norma che tutti lodano indiscrminatamente? — GA > On 30 Jul 2025, at 12:00, nexa-request(a)server-nexa.polito.it wrote: > > Send nexa mailing list submissions to > nexa(a)server-nexa.polito.it > > To subscribe or unsubscribe via the World Wide Web, visit > https://server-nexa.polito.it/cgi-bin/mailman/listinfo/nexa > or, via email, send a message with subject or body 'help' to > nexa-request(a)server-nexa.polito.it > > You can reach the person managing the list at > nexa-owner(a)server-nexa.polito.it > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of nexa digest..." > > > Today's Topics: > > 1. R: "Stefano Rodotà (appositamente richiamato > dall’Aldilà mentre giocava a scacchi)" (Daniela Tafani) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Wed, 30 Jul 2025 09:27:22 +0000 > From: Daniela Tafani <daniela.tafani(a)unipi.it> > To: Giacomo Tesio <giacomo(a)tesio.it>, "nexa(a)server-nexa.polito.it" > <nexa(a)server-nexa.polito.it>, 380° <g380(a)biscuolo.net> > Subject: [nexa] R: "Stefano Rodotà (appositamente richiamato > dall’Aldilà mentre giocava a scacchi)" > > Buongiorno, 380°, > > ________________________________________ > Da: nexa per conto di 380° via nexa > >> non me ne vogliano i membri e i sostenitori del GPDP ma >> l'istituzione è decisamente troppo presa sul serio; dire che le >> aspettative nei confronti del GPDP sono illusorie è eufemistico. > > > Quando un diritto è sistematicamente violato, le norme e le istituzioni che almeno in parte lo tutelino > dovrebbero forse essere prese più sul serio, non meno (anche se capisco il senso della tua nota). > > E trovo eloquente il fatto che le prenda molto sul serio chi si sobbarca la fatica di lavorare per cancellarle. > Ne ho scritto qui: > > GDPR could protect us from the AI Act. That's why it's under attack > <https://zenodo.org/records/14002329> > > Un caro saluto, > Daniela >

3 2

China summons chip giant Nvidia over alleged security risks
by 380° July 31, 2025

July 31, 2025

Buonasera, questo significa (anche) che è _praticamente ovvio_ che una discreta quantità di "chip" [1] esportati in EU siano dotati di tracking e backdoors :-O [2] https://english.almayadeen.net/news/technology/china-summons-chip-giant-nvi…> «China summons chip giant Nvidia over alleged security risks» --8<---------------cut here---------------start------------->8--- China questions Nvidia over security risks in H20 AI chips sold to the Chinese market, raising concerns amid the ongoing US-China tech rivalry. Chinese authorities summoned US tech giant Nvidia on Thursday over serious security concerns related to its H20 artificial intelligence (AI) chips, as tensions escalate in the ongoing technological standoff between Beijing and Washington. The Cyberspace Administration of China (CAC), the country's top internet regulator, announced it had engaged Nvidia in discussions following the discovery of potential vulnerabilities and backdoors in its H20 chips, a version specifically developed for the Chinese market. The CAC issued a public statement on social media confirming that Nvidia was asked to explain the security risks tied to the H20 chips and to submit relevant technical documentation for review. The chips, [designed with downgraded capabilities to comply with US export restrictions], have now become a flashpoint in the broader debate over Nvidia H20 chip security. "The company has been asked to explain the security risks of vulnerabilities and backdoors in its H20 chips sold to China and submit relevant supporting materials," the statement read. Chinese regulators cited reports from US experts, noting that location tracking and remote shutdown [technologies for Nvidia's AI chips] "are already matured." [designed with downgraded capabilities to comply with US export restrictions] <https://english.almayadeen.net/news/technology/nvidia-ai-chips-repair-indus…> [technologies for Nvidia's AI chips] <file:Nvidia’s%20AI%20chips> 1.1 Nvidia's return to the Chinese market draws scrutiny ──────────────────────────────────────────────────────── Earlier this month, Nvidia [said it would resume H20 sales in China] after the US government signaled it would lift some of the licensing restrictions that had previously blocked such exports. The H20 chip is a less powerful version of Nvidia's top-tier AI processors, created to meet US regulatory thresholds while preserving a foothold in the lucrative Chinese AI sector. However, renewed concerns from both sides of the Pacific have [reignited debates over AI chip export] controls and national security implications. Nvidia, based in California, has found itself increasingly caught between Washington's containment strategy and Beijing's push for technological sovereignty. As part of tightening US-China tech tensions, American lawmakers have proposed requiring Nvidia and other AI chipmakers to embed location tracking and remote-disable features into their products. While the proposals are framed as safeguards against unauthorized use, Beijing views such measures as potential vectors for foreign surveillance and control, particularly when implemented in semiconductors destined for Chinese infrastructure or research facilities. The CAC's summons of Nvidia appears to be part of a broader effort to assert Chinese data and cyber sovereignty as foreign technology continues to play a role in sensitive sectors. [said it would resume H20 sales in China] <https://english.almayadeen.net/news/politics/nvidia-cleared-to-resume-ai-ch…> [reignited debates over AI chip export] <https://english.almayadeen.net/news/politics/nvidia-cleared-to-resume-ai-ch…> --8<---------------cut here---------------end--------------->8--- Saluti, 380° [1] che poi tecnicamente sono dei veri e propri SoC (system on chip): si capisce la differenza vero?!? [2] per un esempio si veda alla voce "Intel Management Engine": https://en.wikipedia.org/wiki/Intel_Management_Engine#Security_vulnerabilit… https://en.wikipedia.org/wiki/Intel_Management_Engine#Assertions_that_ME_is… https://en.wikipedia.org/wiki/Intel_Management_Engine#By_Google -- 380° (lost in /traslation/) «Welcome to the chaos of the times If you go left and I go right Pray we make it out alive This is Karmageddon»

1 0

An internet privacy expert has said she was called a “paedo” during a UK Government meeting after raising concerns about the Online Safety Act
by Fabio Alemagna July 31, 2025

July 31, 2025

The legislation, which came into force on July 25, mandates that websites verify users’ age – often using facial recognition or photo ID – before granting access to adult content such as pornography, violence, or material on self-harm and eating disorders. Heather Burns, a tech policy specialist and author of Understanding Privacy <https://heatherburns.tech/book/>, told The National the UK Government and Ofcom had been on a “full-scale media blitz … really hammering down on the lines that this is about children and it's about pornography”. However, she said: “It's not. It applies to any service provider anywhere in the world whose services could theoretically be accessed in Britain. It applies to a bare minimum of, the last figure I saw was 60,000 companies. “What we're seeing over the past week is six years of narratives collapsing into reality in real time.” Burns warned of threats to journalism and freedom of information, noting that sites like Reddit have begun age-checking users accessing “things like forums about war crimes in Gaza <https://www.thenational.scot/topics/Gaza/> and Ukraine <https://www.thenational.scot/topics/ukraine/>”. “This is escalating quickly,” she added. “We're one week in and platforms are censoring news content – because they have to. “I think we're going to see more journalistic content blocked or age restricted before we get a resolution here, and that's scary.” The issue has seen Wikipedia launch a court case <https://www.courthousenews.com/wikipedia-seeks-to-shield-contributors-from-…> against the UK Government because, Burns explained, “they've been thrown into Category One compliance requirements, which is full-fat, full-on, on the grounds that Wikipedia has articles about self-harm, suicide, and eating disorders”. “If young people can't find straightforward, factual, curated information on an encyclopaedia, where are they going to go? They're going to go to the really nasty places you wouldn't want any child looking at.” She added: “History tells us that it's never a good idea when governments start demanding censorship of encyclopaedias.” Continua a leggere qui: https://www.thenational.scot/news/25354060.tech-expert-called-paedo-home-of…

1 0

AI Now: "Artificial Power: 2025 Landscape Report"
by J.C. DE MARTIN July 30, 2025

July 30, 2025

Jun 3, 2025 *Artificial Power: 2025 Landscape Report * https://ainowinstitute.org/publications/research/ai-now-2025-landscape-repo… *Overview* /Artificial Power, /our 2025 Landscape Report, puts forward an actionable strategy for the public to reclaim agency over the future of AI. In the aftermath of the “AI boom,” the report examines how the push to integrate AI products everywhere grants AI companies – and the tech oligarchs that run them – power that goes far beyond their deep pockets. We need to reckon with the ways in which today’s AI isn’t just being used /by/ us, it’s being used /on/ us. The report moves from diagnosis to action: offering concrete strategies for community organizers, policymakers, and the public to change this trajectory. Executive Summary: https://ainowinstitute.org/publications/research/executive-summary-artifici… Download Full Report: https://ainowinstitute.org/wp-content/uploads/2025/06/FINAL-20250609_AINowL…

1 0

[Responsible AI Talk] The AI Alignment Problem, Camilla Bignotti, Banca d’Italia
by Daniele Quercia July 29, 2025

July 29, 2025

The AI Alignment Problem: Bridging Lab Research and Policy Solutions Camilla Bignotti, Banca d’Italia Join the meeting <https://teams.microsoft.com/l/meetup-join/19%3ameeting_MDc4MGUzZWMtY2FhNi00…> The rapid advancement of large language models (LLMs) raises significant concerns about their alignment with human values and goals, particularly given their potential impact on individual lives. The alignment problem can be analysed through two complementary perspectives: conducting empirical experimentation to assess whether AI models align with human expectations or designing policy instruments to implement effective AI governance. This seminar will combine both approaches. Starting from an empirical analysis of how large language models interpret constitutional principles in complex decision-making scenarios, we demonstrate the risks posed by polarized or biased models in complex decision-making contexts. These findings highlight the need for robust training and evaluation methods to address these challenges and ensure model alignment with societal values. This challenge extends beyond the specific domain explored in our study—it represents a crucial aspect related to foundational models applications across their diverse domains. Therefore, we must investigate comprehensive measures to tackle the alignment problem while promoting safe and responsible AI development without stifling innovation. Public-private partnerships are fundamental building blocks for effective AI governance. By creating spaces for dialogue on standard-setting and pilot testing, regulatory institutions and private actors can align their objectives toward responsible innovation. Innovation facilitators established by several European Central Banks exemplify effective tools for promoting stakeholder cooperation in responsible AI deployment. Milano Hub, Banca d'Italia's innovation centre, provides a compelling use case: through its Call for Proposals, it supports promising solutions while steering innovation in a positive direction. <https://arxiv.org/abs/2407.02191> Camilla Bignotti serves as a legal expert at Banca d'Italia's Innovation Centre "Milano Hub," where she contributes to central banking activities that support the digital transformation of the banking, financial, and payments sectors. Her work includes fostering international cooperation among regulatory institutions and innovation hubs in the field of technological evolution. Recently, she was seconded to the Central Bank of Ireland to collaborate on designing a new regulatory sandbox program. Camilla is a qualified lawyer with a background in transnational and comparative law and growing expertise in regulatory frameworks related to financial innovation and artificial intelligence. Subscribe to future talk announcements: Anyone outside Bell Labs can receive talk announcements by subscribing to the mailing list. To subscribe, send an empty email with the subject line "Subscribe RAI" to daniele.quercia(a)nokia-bell-labs.com<mailto:daniele.quercia@nokia-bell-labs.com>.

1 0

[Responsible AI Talk] The AI Alignment Problem, Camilla Bignotti, Banca d’Italia
by Daniele Quercia July 28, 2025

July 28, 2025

The AI Alignment Problem: Bridging Lab Research and Policy Solutions Camilla Bignotti, Banca d’Italia Join the meeting <https://teams.microsoft.com/l/meetup-join/19%3ameeting_MDc4MGUzZWMtY2FhNi00…> The rapid advancement of large language models (LLMs) raises significant concerns about their alignment with human values and goals, particularly given their potential impact on individual lives. The alignment problem can be analysed through two complementary perspectives: conducting empirical experimentation to assess whether AI models align with human expectations or designing policy instruments to implement effective AI governance. This seminar will combine both approaches. Starting from an empirical analysis of how large language models interpret constitutional principles in complex decision-making scenarios, we demonstrate the risks posed by polarized or biased models in complex decision-making contexts. These findings highlight the need for robust training and evaluation methods to address these challenges and ensure model alignment with societal values. This challenge extends beyond the specific domain explored in our study—it represents a crucial aspect related to foundational models applications across their diverse domains. Therefore, we must investigate comprehensive measures to tackle the alignment problem while promoting safe and responsible AI development without stifling innovation. Public-private partnerships are fundamental building blocks for effective AI governance. By creating spaces for dialogue on standard-setting and pilot testing, regulatory institutions and private actors can align their objectives toward responsible innovation. Innovation facilitators established by several European Central Banks exemplify effective tools for promoting stakeholder cooperation in responsible AI deployment. Milano Hub, Banca d'Italia's innovation centre, provides a compelling use case: through its Call for Proposals, it supports promising solutions while steering innovation in a positive direction. <https://arxiv.org/abs/2407.02191> Camilla Bignotti serves as a legal expert at Banca d'Italia's Innovation Centre "Milano Hub," where she contributes to central banking activities that support the digital transformation of the banking, financial, and payments sectors. Her work includes fostering international cooperation among regulatory institutions and innovation hubs in the field of technological evolution. Recently, she was seconded to the Central Bank of Ireland to collaborate on designing a new regulatory sandbox program. Camilla is a qualified lawyer with a background in transnational and comparative law and growing expertise in regulatory frameworks related to financial innovation and artificial intelligence. Subscribe to future talk announcements: Anyone outside Bell Labs can receive talk announcements by subscribing to the mailing list. To subscribe, send an empty email with the subject line "Subscribe RAI" to daniele.quercia(a)nokia-bell-labs.com<mailto:daniele.quercia@nokia-bell-labs.com>.

1 0

Re: [nexa] "China releases AI action plan days after the U.S. as global tech race heats up"
by Giuseppe Attardi July 28, 2025

July 28, 2025

1 0

"China releases AI action plan days after the U.S. as global tech race heats up"
by J.C. DE MARTIN July 27, 2025

July 27, 2025

China releases AI action plan days after the U.S. as global tech race heats up Published Sat, Jul 26 2025 Evelyn Cheng <https://www.cnbc.com/evelyn-cheng/>@in/evelyn-cheng-53b23624 <https://linkedin.com/in/evelyn-cheng-53b23624>@chengevelyn <https://twitter.com/chengevelyn> Key Points * China on Saturday released a global action plan for artificial intelligence that called for international cooperation on tech development and regulation. * Premier Li Qiang announced China has proposed the establishment of a global AI cooperation organization, according to an official readout. * Former Google CEO Eric Schmidt met with Shanghai Party Secretary Chen Jining on Thursday in the city ahead of the AI conference, according to a city announcement. Continua qui: https://www.cnbc.com/2025/07/26/china-ai-action-plan.html

1 0