July 2024 - nexa - server-nexa.polito.it

Jacobin Italia, "L'intelligenza artificiale e la nostra"
by J.C. DE MARTIN July 28, 2024

July 28, 2024

Mi sembra che non sia ancora stato segnalato in lista il fatto che l'ultimo numero della rivista "Jacobin Italia" è dedicato all'"intelligenza artificiale": https://jacobinitalia.it/rivista/lintelligenza-artificiale-e-la-nostra/ Qui sotto l'indice. Per chi ama la carta, il numero costa 13 euro. juan carlos ----- *L’intelligenza artificiale e la nostra * Il capitale nel cervello Redazione Jacobin Italia Operai e algoritmi Matteo Pasquinelli La profezia di Ned Ludd Alessandro Nuvolari L’IA non risolve i problemi essenziali Alexander Brentler Si scrive IA si legge capitale e lavoro Armanda Cetrulo Dario Guarascio Jelena Reljic La lotta per liberare il tempo Giorgio Maran Il lavoro non riconosciuto Marco Bertorello Christian Marazzi Il limite ecologico Lorenzo Tecleme Il complesso militare-digitale Andrea Coveri Claudio Cozza Dario Guarascio Intimità automatizzata Anton Jäger L’allarme dell’Ue servirà a poco Lizzie O’Shea I dati che discriminano Teresa Numerico Un nuovo campo di disuguaglianze Carlotta Caciagli Per un modernismo senza mercato Evgenij Morozov Simón Vázquez La politica e il deepfake Andrea Natella Governare il digitale dal basso Giorgio Pirina Laura Eccher Nella trama dell’algoritmo Gianmarco Peterlongo La techno e il suono della fine di Detroit Gavin Mueller L’oracolo digitale ha le allucinazioni Gino Roncaglia Antonio Montefusco Tiziana Mancinelli Le psicopatologie tecnologiche di J. G. Ballard Owen Hatherley L’umanità sopravviverà all’IA? Garrison Lovely

1 0

Why Zuckerberg’s multibillion-dollar gamble doesn’t just matter to Meta | Artificial intelligence (AI) | The Guardian
by Alberto Cammozzo July 28, 2024

July 28, 2024

Tempo di cominciare a valutare il ritorno degli investimenti fatti. Ma se la competizione si sposta sui prezzi, è forse segno che l'innovazione radicale ha esaurito il suo slancio? E le innovazioni incrementali renderanno davvero redditivi i prodotti che finora non sembrano maturi? <https://www.theguardian.com/technology/article/2024/jul/26/why-zuckerbergs-…> Why Zuckerberg’s multibillion-dollar gamble doesn’t just matter to Meta Dan Milmo Spending on artificial intelligence could hit a staggering $1tn, according to analysts concerned about whether there will be a return on such a spree. Mark Zuckerberg’s answer this week to such jitters was to release his latest AI system for free. Meta’s Llama 3.1 405B is its most powerful yet, it says, and one of the most capable in the world. While the tech company didn’t disclose how much it cost to train, Zuckerberg, its co-founder and chief executive, has previously disclosed a $10.5bn (£8.9bn) investment in just the chips required to power its AI data centres – with the rest of the electronics, the electricity itself, and the physical building an additional cost on top of that. Yet despite the exorbitant outlay, the parent of Facebook and Instagram will charge you nothing for it. If you can get hold of a computer powerful enough to run it, you don’t have to pay Zuckerberg a cent. Whether that gamble will pay off matters to more than just Meta, though: a big bet by investors and Meta’s tech peers is hingeing on the same question. In June, analysts at Goldman Sachs published a note with the sceptical title: “Gen AI: too much spend, too little benefit?” It pointed to a $1tn investment in AI over the next few years by the tech industries, other companies and utilities on infrastructure, including chips and power grids. That prompted the question: “Will this large spend ever pay off?” The note covers a range of views on whether the spree will bring acceptable returns, including an economics professor at Massachusetts Institute of Technology, Daron Acemoglu, who argues that “truly transformative changes” brought about by AI “won’t happen quickly”. In other words, the beneficial economic return from this boom might take longer than investors expect. The research also questions whether power supply can keep up with the demand related to training and operating AI systems – and asks the same about the chips needed to power those models. There are also more optimistic takes from Goldman analysts, who argue that AI will ultimately automate 25% of all work tasks in the US (thus making the economy more productive but also creating new tasks and products) and that the spending boom is not out of whack with previous tech investment sprees. However, Sequoia Capital, an early investor in ChatGPT developer OpenAI, has made it clear that AI companies need to work hard to pay back all that investment in infrastructure such as chips and data centres. A prior estimate that AI companies will need to earn $200bn to pay back their investment has risen to $600bn, wrote Sequoia partner David Cahn. Cahn stressed that backing AI so strongly would “almost certainly” be worthwhile but “the road ahead is going to be a long one”. Benedict Evans, a tech analyst, asked in a note this month whether large language models – the technology underpinning tools such as ChatGPT – “might also be a trap”, because while ChatGPT might look like a full-fledged product, it isn’t. Evans draws an analogy between LLMs and the first iPhones and the appearance of the internet – the potential is there, but other things need to happen as well. “When most technologies first appear they aren’t ready yet and it’s not clear why they’re useful, and they need a bunch more work,” he says. “The iPhone didn’t have 3G and it didn’t have apps. The web arrived in the early 90s, but no one had a modem, let alone broadband. A whole bunch of stuff had to happen before the web could take off.” Evans adds that chatbots have worked well for people who need time-saving fixes for coding and marketing, but it has yet to bring fixes in other areas. “There are very specific cases where this already saves a huge amount of time, but it hasn’t generalised to everybody yet.” He adds: “The feeling that I think most people have looking at ChatGPT is, OK, this is amazingly cool … but what am I supposed to do with this?” This week, OpenAI announced it is testing a search engine in the US, in what could become a direct challenge to search behemoth Google, which has launched AI-generated search answers, too. OpenAI has also hired a former Meta and Twitter executive, Kevin Weil, as its chief product officer to help answer the question: “What am I supposed to do with this?” This week, tech news site the Information reported that OpenAI could lose as much as $5bn this year, based on analysis of internal financial figures and interviews with sources, as it amasses $8.5bn-worth of operating costs including $4bn a year on renting servers from Microsoft and $3bn a year on training models. Full-year revenue could be anywhere between $3.5bn and $4.5bn per year the Information reported, implying a loss of up to $5bn. San Francisco-based OpenAI makes money from charging companies to build consumer products using its models, or letting companies build in-house chatbots with its technology, as well as generating revenue from charging users $20 a month to access a more powerful version of ChatGPT. Google and Anthropic also sell subscriptions to AI systems for $20 a month, while Cohere sells its models to business customers for tasks such as coding or data analysis. The Information admitted its OpenAI calculations were “guesstimates” but if they are right, the company needs to raise cash over the next 12 months. OpenAI declined to comment, although with Microsoft as its main financial backer it could call upon a wealthy benefactor – albeit with regulators circling. But even as tech companies have been fighting for domination at the “frontier” of AI, investing billions in building bigger and better systems, a second front has opened up in a more traditional battle: cost. OpenAI’s most recent release isn’t GPT-4o, the headline-grabbing “multimodal” system that sounded so much like Scarlett Johansson that it sparked a lawsuit. Instead, it’s GPT-4o mini, a stripped-back version of the same AI that the company offers to third-party developers at less than 5% of the cost of its frontier system – undercutting the previous cheapest models, Google’s Gemini 1.5 Flash and Anthropic’s Claude 3 Haiku. On Tuesday, Meta’s release of the latest version of its Llama system undercut OpenAI. Again, while the attention was on the frontier version of the model, dubbed 405B, the company was equally eager to push its smallest LLM, 8B. Unlike its competitors, Llama is available for anyone to download and run on their own systems, with cloud providers such as Groq offering it for a third of the price of OpenAI’s competitor systems. Llama 3.1 8B underperforms those OpenAI equivalents, per benchmarks from Artificial Analysis, but for the price, who can complain? Meta calls Llama “open source”, although critics dispute the claim: it’s possible to download the model and use it with a fairly free licence but the training data remains entirely closed, and the model is not free for anyone to use for any purpose. Some of the restrictions are practical: the copyright status of training data for large language models is, at best, controversial. Meta keeps the specific data it trained Llama 3.1 on a secret, and almost certainly lacks the licence to redistribute it free of charge. Other restrictions have more commercial weight behind them. By keeping restrictions on Llama 3.1’s use in place, Meta places itself at the middle of any future sector that grows around its AI: even if it doesn’t charge directly for access to the model, it gets to control the direction of development, and can always close future avenues if competitors grow too big using its tech. On Tuesday, Zuckerberg said: “I believe the Llama 3.1 release will be an inflection point in the industry… I hope you’ll join us on this journey to bring the benefits of AI to everyone in the world.” For investors and other tech companies, those benefits need to produce a meaningful return.

1 0

Soldi pubblici, codice pubblico: la Svizzera obbliga l’open source nella pubblica amministrazione | DDay.it
by Mauro Alovisio July 27, 2024

July 27, 2024

Gent.mi Segnalo https://www.dday.it/redazione/50029/soldi-pubblici-codice-pubblico-la-svizz… Un caro saluto Mauro

3 3

Rethinking Democracy for the Age of AI
by B. J. July 26, 2024

July 26, 2024

Segnalo questo interessante speech di Bruce Schneier - trascrizione: https://www.schneier.com/blog/archives/2024/06/rethinking-democracy-for-the… - video: https://www.youtube.com/watch?v=Q-IEget44OI > There is a lot written about technology’s threats to democracy. Polarization. Artificial intelligence. The concentration of wealth and power. I have a more general story: The political and economic systems of governance that were created in the mid-18th century are poorly suited for the 21st century. They don’t align incentives well. And they are being hacked too effectively. > > At the same time, the cost of these hacked systems has never been greater, across all human history. We have become too powerful as a species. And our systems cannot keep up with fast-changing disruptive technologies. > > We need to create new systems of governance that align incentives and are resilient against hacking—at every scale. From the individual all the way up to the whole of society. > > For this, I need you to drop your 20th century either/or thinking. This is not about capitalism versus communism. It’s not about democracy versus autocracy. It’s not even about humans versus AI. It’s something new, something we don’t have a name for yet. And it’s “blue sky” thinking, not even remotely considering what’s feasible today. > > Throughout this talk, I want you to think of both democracy and capitalism as information systems. Socio-technical information systems. Protocols for making group decisions. Ones where different players have different incentives. These systems are vulnerable to hacking and need to be secured against those hacks.

1 0

Re: [nexa] sul disastro informatico
by graffio July 23, 2024

July 23, 2024

Cara lista, segnalo questo mio breve commento su Il Manifesto a proposito del disastro dell'altro giorno. https://ilmanifesto.it/oligopoli-privati-decidono-la-nostra-esistenza-digit… E' molto breve, per comodità lo incollo anche qui. L’Occidente ieri si è svegliato con la notizia che migliaia di server equipaggiati con Windows non funzionavano, con conseguenze globali catastrofiche. Il responsabile di un down di tali enormi proporzioni risiedeva in un errore nell’aggiornamento dell’antivirus di Crowdstrike. Si tratta di un software usato da grandi aziende che utilizzano il sistema operativo Windows. La compagnia non è molto nota al pubblico, ma è molto importante per i servizi sulla sicurezza che offre alle grandi aziende. La domanda è: come è possibile che una compagnia che si occupa di sicurezza abbia potuto causare una interruzione di un numero così grande di servizi? La prima risposta risiede probabilmente nel modo con cui è stato distribuito l’aggiornamento del software. Nella maggior parte dei casi si tratta di sistemi automatizzati per installare gli aggiornamenti in tutti i software che sono stati configurati per ricevere tali aggiornamenti, senza la possibilità per i sistemisti “locali” di accettare o meno gli upgrade. Ciò comporta, oltre a evitare di installare gli aggiornamenti “a mano”, risparmiando risorse, anche l’inevitabilità dell’aggiornamento di tutti i sistemi software. In un sistema completamente automatizzato su larga scala, senza un piano B, se qualcosa va storto il problema che ne deriva è serio! E qui veniamo alla seconda possibile risposta. Siamo in situazione di oligopolio. I sistemi operativi della maggior parte di questi sistemi (ospedali, finanza, controllo trasporti etc) sono per lo più basati su Microsoft Windows e usano l’antivirus di Crowdstrike. Probabilmente una sana biodiversità tecnologica avrebbe garantito una mitigazione del problema causato da un errore in un software. Se le aziende colpite dal problema avessero usato antivirus e sistemi operativi diversi, l’impatto dell’aggiornamento con errore sarebbe stato minore Inoltre, dopo pochissimo tempo, Crowdstrike ha fornito i dettagli dell’errore. Un qualsiasi sistemista con le conoscenze giuste sarebbe potuto intervenire “manualmente”, ma con la centralizzazione sempre più spinta di questo genere di sistemi le competenze necessarie sono sempre più scarse. Infine, e questo è un problema che affligge tutta l’informatica commerciale, i test dei software vengono fatti con una precisione ancora inadeguata all’impatto che questi software hanno nella vita reale. Costituiscono un costo per le aziende e in ottica di profitto abbassare i costi significa maggiori ricavi. Pochissime aziende, per lo più statunitensi, hanno in mano le sorti dell’informatica mondiale che ormai è fondamentale nella vita degli umani. Forse è arrivato il momento di sottrarre alle grandi compagnie private il potere di decidere come devono funzionare le tecnologie digitali e restituire questo potere alla sfera pubblica. * gruppo di ricerca circex.org <https://circex.org/it> Un caro saluto m Il 20/07/24 23:08, nexa-request(a)server-nexa.polito.it ha scritto: > 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. sul disastro informatico (Angelo Raffaele Meo) > 2. Re: sul disastro informatico (Benedetto Ponti) > 3. The Observer view on the global IT crash: lessons must be > learned from CrowdStrike fiasco | Observer editorial | The > Guardian (Alberto Cammozzo) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Sat, 20 Jul 2024 17:28:39 +0000 > From: "Angelo Raffaele Meo"<angelo.meo(a)polito.it> > To:"nexa(a)server-nexa.polito.it" <nexa(a)server-nexa.polito.it> > Subject: [nexa] sul disastro informatico > Message-ID: > <AM6PR0502MB3621782BCD350A30731975858FAE2(a)AM6PR0502MB3621.eurprd05.prod.outlook.com> > > Content-Type: text/plain; charset="iso-8859-1" > > carissimi, > su La Stampa di oggi, Riccardo Luna, commentando il crollo di ieri del sistema mondiale delle Telecomunicazioni, > si pone la domanda: "Saremmo più sicuri con un cloud di Stato?" e risponde immediatamente: "lo escludo". > Io invece non solo non lo escludo, ma lo voglio. Non solo per ragioni ideologiche, ma concrete. Non è possibile che un piccolo baco di una piccola azienda di cui sino ad oggi ignoravamo l'esistenza provochi un disastro di quelle dimensioni. > Cosa ne pensate? > Raf Meo >

2 1

AI advances could lead to more child sexual abuse videos, watchdog warns | Child protection | The Guardian
by Alberto Cammozzo July 22, 2024

July 22, 2024

<https://www.theguardian.com/society/article/2024/jul/22/ai-child-sexual-abu…> Advances in artificial intelligence are being used by paedophiles to produce AI-generated videos of child sexual abuse that could increase in volume as the technology improves, according to a safety watchdog. The majority of such cases seen by the Internet Watch Foundation involve manipulation of existing child sexual abuse material (CSAM) or adult pornography, with a child’s face transplanted on to the footage. A handful of examples involve entirely AI-made videos lasting about 20 seconds, the IWF said. The organisation, which monitors CSAM around the world, said it was concerned that more AI-made CSAM videos could emerge as the tools behind them become more widespread and easier to use. Dan Sexton, chief technology officer at the IWF, said if the use of AI video tools followed the same trend as AI-made still images, which have increased in volume as the technology has improved and become more widely available, more CSAM videos could emerge. “I would tentatively say that if it follows the same trends, then we will see more videos,” he said, adding that future videos could also be of “higher quality and realism”. IWF analysts said the majority of videos seen by the organisation on a dark web forum used by paedophiles were partial deepfakes, where AI models freely available online are used to impose a child’s face, including images of known CSAM victims, on existing CSAM videos or adult pornography. The IWF said it found nine such videos. A smaller number of wholly AI-made videos were of a more basic quality, according to the analysts, but they said this would be the “worst” that fully synthetic video would be. The IWF added that AI-made CSAM images have become more photo-realistic this year compared with 2023, when it first started seeing such content. Its snapshot study this year of a single dark web forum – which anonymises users and shields them from tracking – found 12,000 new AI-generated images posted over a month-long period. Nine out of 10 of those images were so realistic they could be prosecuted under the same UK laws covering real CSAM, the IWF said. The organisation, which operates a hotline for the public to report abuse, said it had found examples of AI-made CSAM images being sold online by offenders in place of non-AI made CSAM. The IWF’s chief executive, Susie Hargreaves, said: “Without proper controls, generative AI tools provide a playground for online predators to realise their most perverse and sickening fantasies. Even now, the IWF is starting to see more of this type of material being shared and sold on commercial child sexual abuse websites on the internet.” The IWF is pushing for law changes that will criminalise making guides to generate AI-made CSAM as well as making “fine-tuned” AI models that can produce such material. The cross-bench peer and child safety campaigner Baroness Kidron tabled an amendment to the proposed data protection and digital information bill this year that would have criminalised creating and distributing such models. The bill fell by the wayside after Rishi Sunak called the general election in May. Last week the Guardian reported that AI-made CSAM was overwhelming US law enforcement’s ability to identify and rescue real-life victims.

1 0

[humor]
by 380° July 21, 2024

July 21, 2024

--8<---------------cut here---------------start------------->8--- Guaglio (NA) – Sta suscitando molte polemiche la presentazione di un’Intelligenza Artificiale (AI) realizzata all’Università Federico II di Napoli dal team del Prof. Goffredo Besozzi, direttore del Dipartimento di Informatica dell’ateneo campano. Il ricercatore coordina da mesi un progetto, finanziato con i fondi del PNRR e il contributo dei parcheggiatori abusivi che operano vicino alla Facoltà, che ha portato allo sviluppo di un software innovativo denominato VAIASSA (Vomero AI Advanced Super Self Aware). “Il nostro lavoro dimostra che il Sud Italia può eccellere nelle frontiere più avanzate della tecnologia, e si pone l’obiettivo ambizioso di superare molti ingiusti stereotipi di cui spesso i napoletani sono vittime”, ha dichiarato il Prof. Besozzi vestito da Pulcinella mentre, accompagnato da un mandolino, cantava “O sole mio” e portava un vassoio di pizza ca pummarola n’coppa da servire ai giornalisti invitati alla presentazione. Una della preoccupazioni ricorrenti quando si parla dello sviluppo straordinario delle AI è l’impatto che questi nuovi strumenti avranno sull’occupazione, in particolare la paura che le macchine possano rubare il lavoro agli umani. Questo per fortuna non sembra essere un problema per il progetto VAIASSA: “Qui di lavoro non ce n’è proprio, nessuno ha paura che gli rubino una cosa che non ha, qualcuno per caso ha paura che gli rubino un unicorno?”. La AI partenopea nel suo processo di apprendimento è stata istruita con input locali, spingendola ad assumere comportamenti tipici della zona in cui opera e portando a risultati inaspettati: sembra infatti che questa non sia interessata al lavoro, quanto al sussidio di disoccupazione. In poche settimane di addestramento è riuscita a ottenere il reddito di cittadinanza (anche se cancellato dal Governo Meloni, NdR), l’assegno di inclusione, il supporto per la formazione e lavoro e il riconoscimento di invalidità, diventando la prima macchina falsa invalida al mondo. Il vero timore dei cittadini campani quindi è che VAIASSA possa rubare loro non tanto il lavoro, quanto il sussidio di disoccupazione. Nel frattempo, il processo di addestramento continua tra le polemiche, visto che sta portando a comportamenti sempre più discutibili. Negli ultimi giorni una AI ha perso tre dita sparando botti di Capodanno ed è notizia dell’ultim’ora che sono state avvistate tre AI senza casco sullo stesso computer. Andrea Michielotto (Quest’articolo è stato scritto anche grazie al contributo di Box) --8<---------------cut here---------------end--------------->8--- -- 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 2

Obsolescenza programmata... *E DICHIARATA* [...che si tratti di "scarpe", è un dettaglio]
by Damiano Verzulli July 21, 2024

July 21, 2024

Ignoravo l'esistenza di scarpe "smart" (i miei figli mi avevano introdotto al mondo delle scarpe "customizzate"/"sponsorizzate"... ma non a quelle smart). Leggo oggi, pero', che "/...Nike’s self-lacing Adapt BB sneakers are losing their remote control mobile app..../". https://www.theverge.com/2024/7/4/24192184/nike-adapt-bb-shoes-sneakers-app… Sembra, quindi, che delle (costose) scarpe-smart di qualche anno fa... stiano per perdere la connettivita' con la companion-app che ne favoriva la fruibilita'. E' --a mio avviso-- una storia interessante... perché la "chiusura" sia del firmware del microcontrollore, sia del protocollo di comunicazione con l'app (oltre che l'app, ovviamente)... impedisce comunque che un qualche studente di ingegneria o di informatica (ma, forse, anche di scuola superiore), possa passare del tempo a... ridare nuova vita al tutto (riscrivendo APP equivalenti o sostituendo il firmware nel microcontrollore). E se qualche non-tecnico pensasse che si tratti di attivita' complesse, non alla portata di tali giovani figure, invito a leggere la storia di Sayan [1], un mio amico indiano poco piu' che diciottenne, grazie al quale il mio smartphone... funziona in modo "libero" :-) Tornando al lock-in ed all'obsolescenza programmata, secondo me, siamo solo agli inizi.... considerando che di "smart", ormai, abbiamo un sacco di cose.. Un saluto, DV [1] https://dvblog.soabit.com/sayan-sarkar-a-sparkos-android-guru-you-should-kn… P.S.: il caso vuole che proprio in questo momento, qui di fianco a me, ci sia uno smartwatch... ma con sopra AsteroidOS :-) -- 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

1 0

Building Artificial Consciousness With GPT-3
by Fabio Alemagna July 21, 2024

July 21, 2024

> I've developed Molly, a conversational agent with an inner monologue, using OpenAI's Large Language Model (LLM). > This experiment gives the illusion of artificial consciousness, and raises questions about the nature of consciousness. > > [...] > > One theory of consciousness called Attention Schema Theory, developed by neuroscientist Michael Graziano at Princeton University, > strikes me as particularly interesting. For one part, this theory introduces a concept called "attention", which is reminiscent > of the famous "Attention is all you need" that triggered the recent spike in Language Model performance. > But most interesting, Attention Schema Theory shares similarities with illusionism, and postulates that consciousness is an illusion. Qui: https://marmelab.com/blog/2023/06/06/artificial-consciousness.html PS: ho replicato l'esperimento con ChatGPT-4. con risultati del tutto analoghi, se non migliori. Fabio

4 6

The Observer view on the global IT crash: lessons must be learned from CrowdStrike fiasco | Observer editorial | The Guardian
by Alberto Cammozzo July 21, 2024

July 21, 2024

<https://www.theguardian.com/commentisfree/article/2024/jul/20/the-observer-…> One bit of good news about the “epic IT crash” that brought the western world to a temporary standstill is that it was a product of human error rather than a Russian cyber-attack like the SolarWinds hack of 2020 that had a similar modus operandi. Last week’s outage was caused by an update that a big US cybersecurity firm, CrowdStrike, pushed to its corporate clients early on Friday morning, which conflicted with Microsoft’s Windows operating system, rendering devices inoperable – with predictable consequences, given that virtually every large organisation in the world is using Microsoft Windows. Fortunately, fixing the problem turned out to be straightforward, though tedious, which will doubtless lead people to think of it as a hiccup rather than as a dry-run for something much worse. After all, if a single error by a single tech company can cause this much disruption, imagine what a determined adversary could do. Just as the pandemic forced us to confront the limitations of the global supply chains that had been created to improve efficiency rather than resilience, this CrowdStrike mistake should trigger a reappraisal of our networked world. One question to be pondered concerns the societal risks of industrial consolidation in the tech industry. CrowdStrike is one of the largest companies in the cybersecurity market. Microsoft has a stranglehold on the business computing marketplace. Every large organisation runs Windows, and most small businesses do, too. Add the pressures that governments, agencies and the National Cyber Security Centre are putting on companies to improve their cybersecurity, which leads them to sign up for tools like CrowdStrike’s Falcon, and we have the potential for the kind of perfect storm we witnessed last week. Most businesses run on Microsoft Windows, so corporate computing is basically a monoculture. This may be good for efficiency, standardisation, training, etc, but it is also bad for resilience if anything goes wrong. Industrial consolidation also highlights the “attack surface” that hackers seek. If there are a handful of large cybersecurity companies supplying, and regularly updating, millions of desktop corporate PCs, then those supply chains constitute a surface with attractive potential for massive disruption. This is what the SolarWinds attack vividly demonstrated: important US government departments (homeland security, state, commerce and treasury) were affected, as well as corporations such as FireEye, Microsoft, Intel, Cisco and Deloitte. There are lessons to be learned from this fiasco. The obvious one is that, while regular automated updates of security software are invaluable, there should always be a phased rollout of each update so that problems surface before they become catastrophic. But what the CrowdStrike error has revealed above everything else is how fragile our networked world has become. We have become utterly dependent on a complex web of technologies that few understand, created by an industry that seems indifferent to the consequences of its creations. We find ourselves in a new world, but it’s not exactly a brave one.

2 2