Oggi è il Giorno della memoria dedicato alle vittime del terrorismo interno e internazionale e delle stragi di tale matrice, istituito nel 2007 e celebrato ogni anno il #9maggio, giorno dell’uccisione di Aldo Moro e di Peppino Impastato.
Oggi è il Giorno della memoria dedicato alle vittime del terrorismo interno e internazionale e delle stragi di tale matrice, istituito nel 2007 e celebrato ogni anno il #9maggio, giorno dell’uccisione di Aldo Moro e di Peppino Impastato.
We’ve certainly seen people take a photo of a part, bring it into CAD, and then scale it until some dimension on the screen is the same as a known dimension of the part. We like what [Scale Addition] shows in the video below. In addition to a picture of the part, he also takes a picture of a vernier caliper gripping the part. Now your scale is built into the picture, and you can edit out the caliper later.
He uses SketchUp, but this would work on any software that can import an image. Given the image with the correct scale, it is usually trivial to sketch over the image or even use an automatic tracing function. You still need some measurements, of course. The part in question has a vertical portion that doesn’t show up in a flat photograph. We’ve had good luck using a flatbed scanner before, and there’s no reason you couldn’t scan a part with a caliper for scale.
This is one case where a digital caliper probably isn’t as handy as an old-school one. But it would be possible to do the same trick with any measurement device. You could even take your picture on a grid of known dimensions. This would also allow you to check that the distances at the top and bottom are the same as the distances on the right and left.
@Informatica (Italy e non Italy 😁) Milioni di soggetti, che vanno dagli utenti ai creatori dei prodotti connessi, devono essere consapevoli dell'inclusione di un prodotto nell’ambito di applicazione del regolamento europeo e mettere in atto tutti i presidi previsti dalla normativa, superando il binomio fra Data
@Informatica (Italy e non Italy 😁) Nella guerra di Gaza, Israele ha sfruttato strumenti di intelligenza artificiale che destano preoccupazioni etiche per la sorveglianza, l’errata identificazione e danni involontari ai civili. Ecco perché fa discutere il nuovo uso dell'AI in tempo reale da parte di Israele, con l'hub
@Notizie dall'Italia e dal mondo Escalation militare dopo raid indiani in Kashmir: droni, artiglieria e vittime civili lungo la Linea di Controllo. L'articolo India-Pakistan: accuse reciproche di attacchi lungo il confine proviene da pagineesteri.it/2025/05/09/asi…
Indo-Pacifico. Da qualche anno è questa la parola chiave di ogni agenda strategica, il cuore di ogni esercitazione militare rilevante, il baricentro di ogni riflessione geopolitica con ambizione globale. Ma non si tratta solo di un concetto geografico allargato: è lo specchio dell’era multipolare che
La cifra che metterà a disposizione è pari al 99 per cento del suo patrimonio
Gates, che ha 69 anni, ha motivato il cambio di programmi dicendo che il mondo si trova ad affrontare molti problemi urgenti e che lui vuole finanziare, e incoraggiare altri a farlo, nuovi strumenti per cercare di risolverli: «Ci sono troppi problemi urgenti da risolvere perché io trattenga risorse che potrebbero essere usate per aiutare le persone»
Gates ha indicato tre priorità della sua fondazione per il futuro: la mortalità infantile, le malattie infettive, principalmente poliomielite, morbillo e malaria, e iniziative legate all’istruzione
In un’intervista al Financial Times Gates ha esplicitamente attaccato Musk dicendo che «non è una bella immagine quella dell’uomo più ricco del mondo che sta uccidendo i bambini più poveri del mondo», spiegando che lo smantellamento di USAID potrebbe portare a una nuova ondata di morti e che va «ben oltre qualsiasi eliminazione degli sprechi»
L'articolo proviene da #StartMag e viene ricondiviso sulla comunità Lemmy @Informatica (Italy e non Italy 😁) Starlink aggira le regole Agcom e conquista San Pietro: disparità di trattamento evidente. La lettera di Luigi Pereira.
Si è concluso il Consiglio supremo di Difesa, convocato dal presidente della Repubblica, Sergio Mattarella la scorsa settimana. Il massimo organo costituzionale responsabile per la sicurezza nazionale si è riunito per esaminare i più recenti sviluppi sullo scacchiere
@Notizie dall'Italia e dal mondo Distrutte 9 case, 6 grotte abitate, 10 cisterne d'acqua, 4 stalle, una cabina elettrica, l'impianto solare e il centro comunitario. Restano in piedi solo 3 strutture e la scuola L'articolo MASAFER YATTA. Le ruspe israeliane
Nuvole, sogni e antenne. freezonemagazine.com/articoli/… Il potere della musica. Potere della musica, potere della parola ritmata dalla musica. Potere demonico, maieutico, che può essere, allo stesso tempo, violento e consolatorio, spaventoso e accogliente. Nella musica, i contrasti sembrano talvolta ricomporsi e la verità, le nostra verità, offrirsi in un breve lampo di illuminazione: si accende, e subito, non appena l’eco […] L'articolo Nuvole, sogni Il
The 1970s saw a veritable goldrush to corner the home computer market, with Tandy’s Z80-powered TRS-80 probably one of the most (in)famous entries. Designed from the ground up to be as cheap as possible, the original (Model I) TRS-80 cut all corners management could get away with. The story of the TRS-80 Model I is the subject of a recent video by the [Little Car] YouTube channel.
Having the TRS-80 sold as an assembled computer was not a given, as kits were rather common back then, especially since Tandy’s Radio Shack stores had their roots in selling radio kits and the like, not computer systems. Ultimately the system was built around the lower-end 1.78 MHz Z80 MPU with the rudimentary Level I BASIC (later updated to Level II), though with a memory layout that made running the likes of CP/M impossible. The Model II would be sold later as a dedicated business machine, with the Model III being the actual upgrade to the Model I. You could also absolutely access online services like those of Compuserve on your TRS-80.
While it was appreciated that the TRS-80 (lovingly called the ‘Trash-80’ by some) had a real keyboard instead of a cheap membrane keyboard, the rest of the Model I hardware had plenty of issues, and new FCC regulations meant that the Model III was required as the Model I produced enough EMI to drown out nearby radios. Despite this, the Model I put Tandy on the map of home computers, opened the world of computing to many children and adults, with subsequent Tandy TRS-80 computers being released until 1991 with the Model 4.
Ne sentivamo il bisogno, vero? Mi chiedo perché non esista una big tech che investa seriamente su Linux, che come in questo caso avrebbe potuto essere un'alternativa già pronta (e soprattutto appetibile) a Windows.
Già! Se le aziende statunitensi sono ben lontane da un paradigma più etico, figuriamoci poi una big tech cinese quanto può avere a cuore la democrazia!
Tecnologia, geopolitica e società si intrecciano sempre di più. Questa settimana parliamo della supremazia della Cina, dei problemi di Tesla, e poi NVIDIA e DOGE. La NL è uscita eccezionalmente oggi, e non perderla
Although [Vitor Fróis] is explaining linear regression because it relates to machine learning, the post and, indeed, the topic have wide applications in many things that we do with electronics and computers. It is one way to use independent variables to predict dependent variables, and, in its simplest form, it is based on nothing more than a straight line.
You might remember from school that a straight line can be described by: y=mx+b. Here, m is the slope of the line and b is the y-intercept. Another way to think about it is that m is how fast the line goes up (or down, if m is negative), and b is where the line “starts” at x=0.
[Vitor] starts out with a great example: home prices (the dependent variable) and area (the independent variable). As you would guess, bigger houses tend to sell for more than smaller houses. But it isn’t an exact formula, because there are a lot of reasons a house might sell for more or less. If you plot it, you don’t get a nice line; you get a cloud of points that sort of group around some imaginary line.
There are mathematical ways to figure out what line you should imagine, but you can often eyeball it, too. The real trick is evaluating the quality of that imaginary line.
To do that, you need an error measure. If you didn’t know better, you’d probably think expressing the error in terms of absolute value would be best. You know, “this is 10 off” or whatever. But, as [Vitor] explains, the standard way to do this is with a squared error term R2. Why? Read the post and find out.
For electronics, linear regression has many applications, including interpreting sensor data. You might also use it to generalize a batch of unknown components, for example. Think of a batch of transistors with different Beta values at different frequencies. A linear regression will help you predict the Beta and the error term will tell you if it is worth using the prediction or not. Or, maybe you just want to make the perfect cup of coffee.
In the driving simulator community, setups can quickly grow ever more complicated and expensive, all in the quest for fidelity. For [CNCDan], rather than buy pedals off the shelf, he opted to build his own.
[Dan] has been using some commercial pedals alongside his own DIY steering wheel and the experience is rather lackluster in comparison. The build starts with some custom brackets. To save on cost, they are flat with tabs to let you know where to bend it in a vise. Additionally, rather than three sets of unique brackets, [Dan] made them all the same to save on cost. The clutch and throttle are a simple hall effect sensor with a spring to provide feedback. However, each bracket provides a set of spring mounting holes to adjust the curve. Change up the angle of the spring and you have a different curve. The brake pedal is different as rather than measure position, it measures force. A load cell is perfect for this. The HX711 load cell sensor board that [Dan] bought was only polling at 10hz. Lifting a pin from ground and bodging it to VDD puts the chip in 80hz, which is much more usable for a driving sim setup.
[Dan] also cleverly uses a 3d printed bushing without any walls as resistance for the pedal. Since the bushing is just the infill, the bushing stiffness is controlled by the infill percentage. Aluminum extrusion forms the base so [Dan] can adjust the exact pedal positions. To finish it off, a bog standard Arduino communicates to the PC as a game controller.
You might be old enough to remember record platters, but you probably aren’t old enough to remember when records were cylinders. The Edison Blue Amberol records came out in 1912 and were far superior to the earlier wax cylinders. If you had one today, how could you play it? Easy. Just build [Palingenesis’] record player. You can even hear it do its thing in the video below.
The cylinders are made of plaster with a celluloid wrapper tinted with the namesake blue color. They were more durable than the old wax records and could hold well over four minutes of sound.
The player is mostly made from wood cut with a mill or a laser. There are some bearings, fasteners, and — of course — electronics. The stylus requires some care. Conventional records use a lateral-cut groove, but these old records use a vertical-cut. That means the pickup moves up and down and has a rounder tip than a conventional needle.
Rather than try to control the motor to an exact speed, you get to set the speed with a potentiometer and see the resulting RPM on a small display. Overall, an involved but worthwhile project.
We recently looked at some players that would have been new about the same time as the blue record in the video. We don’t think you could modify one of these to play stereo, but if you do, let us know immediately!
If you have seen Star Wars, you know what is being referenced here. Holochess appeared as a diversion built into the Millennium Falcon in the very first movie, way back in 1977. While not quite as iconic a use of simulated holograms as tiny Princess Leia begging for hope, it evidently struck a chord with [Maker Mac70], given the impressive effort he’s evidently gone through to re-create the game table from the film.
The key component of this unit is a plate from Japanese firm ASKA3D that scatters light from displays inside the table in just such a way that the diverging rays are focused at a point above its surface, creating the illusion of an image hovering in space. Or in this case, hovering at the surface of a acrylic chessboard. Granted, this technique only works from one viewing angle, and so is not a perfect recreation of a sci-fi holoprojector. But from the right angle, it looks really good, as you can see in the video below.
There are actually six SPI displays, driven by an Arduino GIGA, positioned and angled to project each character in the game. Placing two of the displays on 3D printed gantries allows them to move, allowing two creatures to battle in the center of the table. As [Maker Mac70] admits, this is quite a bit simpler than the Holochess game seen in the film, but it’s quite impressive for real world hardware.
If this all seems a little bit familiar, we covered an earlier floating display by [Maker Mac70] last year. This works on similar principles, but uses more common components which makes the technique more accessible. If chess isn’t your forte, why not a volumetric display that plays DOOM? If you’re interested in real holograms, not Sci-Fi, our own [Maya Posch] did a deep dive you may find interesting.
Uno studio su larga scala condotto da Rubrik Zero Labs ha confermato che quasi il 90% dei responsabili IT e della sicurezza informatica in tutto il mondo dovrà affrontare attacchi informatici nel 2024. Il rapporto, intitolato “The State of Data Security 2025: A Distributed Crisis”, documenta una tendenza preoccupante: gli ambienti IT ibridi, diventati la norma per le aziende, stanno creando nuove vulnerabilità per le quali le aziende non sono preparate.
Gli esperti sottolineano che il passaggio al cloud è spesso accompagnato da un falso senso di sicurezza. Come ha sottolineato Joe Hladik, CEO di Rubrik Zero Labs, molti si affidano ai fornitori di servizi cloud per la sicurezza. In realtà, gli aggressori sfruttano attivamente le falle nelle architetture ibride e continuano ad attaccare utilizzandoransomware e furto di credenziali.
Gli attacchi stanno diventando sempre più frequenti. I principali vettori sono le fughe di dati, l’infezione dei dispositivi con malware, la compromissione delle piattaforme cloud, il phishing e gli attacchi dall’interno. Le conseguenze sono tangibili: aumento dei costi per la sicurezza informatica (40%), perdite di reputazione (37%) e persino un cambio di management in un terzo dei casi.
La situazione è complicata dal forte aumento dei volumi di dati e dalla diffusione dei sistemi di intelligenza artificiale. Il 90% degli intervistati ha dichiarato di gestire un ambiente cloud ibrido e che oltre la metà dei propri carichi di lavoro si sta già spostando sul cloud. Un terzo degli intervistati ha indicato la protezione dei dati come un problema importante in tali condizioni, mentre un terzo ha sottolineato la mancanza di gestione centralizzata e di trasparenza.
Secondo la telemetria di Rubrik, il 36% dei file sensibili nel cloud sono dati ad alto rischio. Tra questi rientrano dati personali, codici sorgente, numeri di previdenza sociale, numeri di telefono, nonché chiavi API e account. Si tratta del tipo di informazioni che interessa agli aggressori che cercano di rubare identità e penetrare nei sistemi critici.
Particolarmente preoccupante è lo stato dei backup: tra coloro che sono sopravvissuti a un attacco ransomware, l’86% ha pagato il riscatto e il 74% ha ammesso una parziale compromissione dei propri sistemi di ripristino. Nel 35% dei casi i sistemi di backup erano completamente compromessi.
La flessibilità dell’architettura ibrida è diventata anche il suo tallone d’Achille: il 92% delle aziende utilizza da due a cinque piattaforme cloud o SaaS, il che crea difficoltà nella gestione degli accessi. Secondo il 28% degli intervistati, gli attacchi che sfruttano account rubati e privilegi interni stanno diventando sempre più comuni.
In definitiva, Rubrik auspica un passaggio a un modello di sicurezza incentrato sui dati anziché sulle infrastrutture. Visibilità, controllo, classificazione e capacità di recupero rapido diventano priorità. Senza questo, i paesaggi ibridi continueranno a rappresentare un bersaglio appetibile per gli attacchi.
Il 1° maggio 2025, una fotografia scattata durante una riunione di gabinetto ha mostrato il Consigliere per la Sicurezza Nazionale, Mike Waltz, mentre utilizzava un’applicazione di messaggistica sconosciuta ai più.
L’app in questione era TM SGNL, una versione modificata di Signal sviluppata da TeleMessage, azienda israeliana specializzata in soluzioni di archiviazione per comunicazioni sicure. L’immagine ha sollevato interrogativi sulla sicurezza delle comunicazioni ai massimi livelli del governo statunitense. TM SGNL è progettata per archiviare messaggi da piattaforme come Signal, WhatsApp e Telegram, permettendo alle agenzie governative di conservare comunicazioni per scopi normativi e di conformità. Tuttavia, l’implementazione di queste funzionalità ha introdotto vulnerabilità significative nel sistema.
Il 4 maggio, il sito 404 Media ha riportato che un hacker ha violato i sistemi di TeleMessage, ottenendo accesso a dati sensibili, tra cui contenuti di chat, informazioni di contatto di funzionari governativi e credenziali di accesso al backend del servizio. L’attacco è stato descritto come sorprendentemente semplice: “Mi è bastata una ventina di minuti,” ha dichiarato l’hacker, sottolineando la facilità con cui ha penetrato i sistemi dell’azienda.
Sebbene i messaggi di Waltz e di altri membri del gabinetto non siano stati compromessi, l’incidente ha evidenziato gravi lacune nella sicurezza delle comunicazioni governative. I dati sottratti includevano informazioni su funzionari della Customs and Border Protection (CBP), dipendenti di Coinbase e altri enti sensibili.
In seguito alla divulgazione dell’attacco, TeleMessage ha sospeso temporaneamente tutti i suoi servizi per indagare sull’incidente, come confermato dalla società madre Smarsh. Un portavoce ha dichiarato: “Abbiamo agito rapidamente per contenere l’incidente e abbiamo coinvolto una società di cybersicurezza esterna per supportare le nostre indagini.”
L’incidente ha sollevato preoccupazioni sulla pratica di modificare applicazioni di messaggistica cifrate per scopi di archiviazione, poiché tali modifiche possono compromettere la sicurezza intrinseca delle comunicazioni. Signal, l’applicazione originale su cui si basa TM SGNL, ha dichiarato di non poter garantire la sicurezza di versioni non ufficiali del suo software.
L’uso di TM SGNL da parte di alti funzionari governativi, combinato con le vulnerabilità evidenziate dall’attacco, ha sollevato interrogativi sulla sicurezza delle comunicazioni ai vertici del potere. L’incidente, soprannominato “SignalGate”, ha portato alla rimozione di Mike Waltz dal suo incarico, sebbene sia stato successivamente nominato come prossimo ambasciatore degli Stati Uniti presso le Nazioni Unite.
Questo episodio mette in luce la necessità di valutare attentamente le soluzioni tecnologiche adottate per la gestione delle comunicazioni sensibili, bilanciando le esigenze di conformità normativa con la protezione della sicurezza nazionale.
A hacker has breached and stolen customer data from TeleMessage, an obscure Israeli company that sells modified versions of Signal and other messaging apps to the U.S. government to archive messages, 404 Media has learned. The data stolen by the hacker contains the contents of some direct messages and group chats sent using its Signal clone, as well as modified versions of WhatsApp, Telegram, and WeChat. TeleMessage was recently the center of a wave of media coverage after Mike Waltz accidentally revealed he used the tool in a cabinet meeting with President Trump.
The hack shows that an app gathering messages of the highest ranking officials in the government—Waltz’s chats on the app include recipients that appear to be Marco Rubio, Tulsi Gabbard, and JD Vance—contained serious vulnerabilities that allowed a hacker to trivially access the archived chats of some people who used the same tool. The hacker has not obtained the messages of cabinet members, Waltz, and people he spoke to, but the hack shows that the archived chat logs are not end-to-end encrypted between the modified version of the messaging app and the ultimate archive destination controlled by the TeleMessage customer.
Data related to Customs and Border Protection (CBP), the cryptocurrency giant Coinbase, and other financial institutions are included in the hacked material, according to screenshots of messages and backend systems obtained by 404 Media.
💡 Do you know anything else about TeleMessage? I would love to hear from you. Using a non-work device, you can message me securely on Signal at signalaccount.05 or send me an email at joseph@404media.co.
The breach is hugely significant not just for those individual customers, but also for the U.S. government more widely. On Thursday, 404 Media was first to report that at the time U.S. National Security Advisor Waltz accidentally revealed he was using TeleMessage’s modified version of Signal during the cabinet meeting. The use of that tool raised questions about what classification of information was being discussed across the app and how that data was being secured, and came after revelations top U.S. officials were using Signal to discuss active combat operations.
The hacker did not access all messages stored or collected by TeleMessage, but could have likely accessed more data if they decided to, underscoring the extreme risk posed by taking ordinarily secure end-to-end encrypted messaging apps such as Signal and adding an extra archiving feature to them.
“I would say the whole process took about 15-20 minutes,” the hacker said, describing how they broke into TeleMessage’s systems. “It wasn’t much effort at all.” 404 Media does not know the identity of the hacker, but has verified aspects of the material they have anonymously provided. A screenshot provided by the hacker. Redactions by 404 Media. The data includes apparent message contents; the names and contact information for government officials; usernames and passwords for TeleMessage’s backend panel; and indications of what agencies and companies might be TeleMessage customers. The data is not representative of all of TeleMessage’s customers or the sorts of messages it covers; instead, it is snapshots of data passing through TeleMessage’s servers at a point in time. The hacker was able to login to the TeleMessage backend panel using the usernames and passwords found in these snapshots.
A message sent to a group chat called “Upstanding Citizens Brigade” included in the hacked data says its “source type” is “Signal,” indicating it came from TeleMessage’s modified version of the messaging app. The message itself was a link to this tweet posted on Sunday which is a clip of an NBC Meet the Press interview with President Trump about his memecoin. The hacked data includes phone numbers that were part of the group chat.
One hacked message was sent to a group chat apparently associated with the crypto firm Galaxy Digital. One message said, “need 7 dems to get to 60.. would be very close” to the “GD Macro” group. Another message said, “Just spoke to a D staffer on the senate side - 2 cosponsors (Alsobrooks and gillibrand) did not sign the opposition letter so they think the bill still has a good chance of passage the senate with 5 more Ds supporting it.” playlist.megaphone.fm?p=TBIEA2… This means a hacker was able to steal what appears to be active, timely discussion about the efforts behind passing a hugely important and controversial cryptocurrency bill; Saturday, Democratic lawmakers published a letter explaining they would oppose it. Bill cosponsors Maryland Sen. Angela Alsobrooks and New York Sen. Kirsten Gillibrand did not sign that letter.
One screenshot of the hacker’s access to a TeleMessage panel lists the names, phone numbers, and email addresses of CBP officials. The screenshot says “select 0 of 747,” indicating that there may be that many CBP officials included in the data. A similar screenshot shows the contact information of current and former Coinbase employees.
Another screenshot obtained by 404 Media mentions Scotiabank. Financial institutions might turn to a tool like TeleMessage to comply with regulations around keeping copies of business communications. Governments have legal requirements to preserve messages in a similar way.
Another screenshot indicates that the Intelligence Branch of the Washington D.C. Metropolitan Police may be using the tool. A screenshot provided by the hacker. Redactions by 404 Media. The hacker was able to access data that the app captured intermittently for debugging purposes, and would not have been able to capture every single message or piece of data that passes through TeleMessage’s service. However, the sample data they captured did contain fragments of live, unencrypted data passing through TeleMessage’s production server on their way to getting archived.
404 Media verified the hacked data in various ways. First, 404 Media phoned some of the numbers listed as belonging to CBP officials. In one case, a person who answered said their name was the same as the one included in the hacked data, then confirmed their affiliation with CBP when asked. The voicemail message for another number included the name of an alleged CBP official included in the data.
404 Media ran several phone numbers that appeared to be associated with employees at crypto firms Coinbase and Galaxy through a search tool called OSINT Industries, which confirmed that these phone numbers belonged to people who worked for these companies.
The server that the hacker compromised is hosted on Amazon AWS’s cloud infrastructure in Northern Virginia. By reviewing the source code of TeleMessage’s modified Signal app for Android, 404 Media confirmed that the app sends message data to this endpoint. 404 Media also made an HTTP request to this server to confirm that it is online.
TeleMessage came to the fore after a Reuters photographer took a photo in which Waltz was using his mobile phone. Zooming in on that photo revealed he was using a modified version of Signal made by TeleMessage. The photograph came around a month after The Atlantic reported that top U.S. officials were using Signal to message one another about military operations. As part of that, Waltz accidentally added the editor-in-chief of the publication to the Signal group chat.
TeleMessage offers governments and companies a way to archive messages from end-to-end encrypted messaging apps such as Signal and WhatsApp. TeleMessage does this by making modified versions of those apps that send copies of messages to a remote server. A video from TeleMessage posted to YouTube claims that its app keeps “intact the Signal security and end-to-end encryption when communicating with other Signal users.”
“The only difference is the TeleMessage version captures all incoming and outgoing Signal messages for archiving purposes,” the video continues.
It is not true that an archiving solution properly preserves the security offered by an end-to-end encrypted messaging app such as Signal. Ordinarily, only someone sending a Signal message and their intended recipient will be able to read the contexts of the message. TeleMessage essentially adds a third party to that conversation by sending copies of those messages somewhere else for storage. If not stored securely, those copies could in turn be susceptible to monitoring or falling into the wrong hands.
That theoretical risk has now become very real.
A Signal spokesperson previously told 404 Media in email “We cannot guarantee the privacy or security properties of unofficial versions of Signal.”
White House deputy press secretary Anna Kelly previously told NBC News in an email: “As we have said many times, Signal is an approved app for government use and is loaded on government phones.”
The hacker told 404 Media that they targeted TeleMessage because they were “just curious how secure it was.” They did not want to disclose the issue to the company directly because they believed the company might “try their best to cover it up.”
“If I could have found this in less than 30 minutes then anybody else could too. And who knows how long it’s been vulnerable?” the hacker said.
404 Media is not explaining in detail how the hacker managed to obtain this data in case others may try to exploit the same vulnerability.
According to public procurement records, TeleMessage has contracts with a range of U.S. government agencies, including the State Department and Centers for Disease Control and Prevention.
Guy Levit, CEO of TeleMessage, directed a request for comment to a press representative of Smarsh, TeleMessage’s parent company. That representative did not immediately respond to an email or voicemail.
Recently, after the wave of media coverage about Waltz’s use of the tool, TeleMessage wiped its website. Before then it contained details on the services it offers, what its apps were capable of, and in some cases direct downloads for the archiving apps themselves.
Neither CBP, Coinbase, Scotiabank, Galaxy Digital, nor Washington D.C. Metropolitan Police responded to a request for comment.
Former national security adviser Mike Waltz was photographed using a Signal-like messaging app during Wednesday’s Cabinet meeting, more than a month after he first came under intense scrutiny for accidentally including a journalist in a group chat th…
Independent ATProto infrastructure has been rapidly expanding recently, experiments with games on ATProto, and Graze offers developer grants.
I also run a weekly newsletter, where you get all the articles I published this week directly in your inbox, as well as additional analysis. You can sign up right here, and get the next edition tomorrow!
Independent Infrastructure news
Over the last week, the effort towards decentralisation and running independent pieces of ATProto infrastructure has sped up significantly. There are now multiple relays that are publicly accessible. Other people also have made alternate AppViews that are Bluesky-compatible. Combined, this makes it now possible to fully use Bluesky without using any infrastructure owned by Bluesky PBC, and the first people have done so. To do so means using a separate PDS, relay, AppView and client.
Some of the updates regarding relays:
Blacksky has built their own relay, using their own custom implementation. This relay is publicly accessible, meaning that other people can use this relay instead of the relay that Bluesky PBC uses.
A writeup on how to set up your own relay by Bluesky engineer Bryan Newbold, for some 34 USD/month.
Making relays cheaper has been due to the Sync 1.1 update, Bluesky PBC goes into more detail in a blog post what this entails.
And the updates regarding clients and AppViews:
Two clients now support the ability for users to set their own AppView, Deer and TOKIMEKI.
AppViewLite is another AppView for Bluesky that has been around for a while, that focuses on being cheap to run. It also heavily optimises for network data storage, with creator Alnkq running AppViewLite that contains full network data on a cheap 10 year old machine. So far, AppViewLite only worked with a custom frontend. An update this week now make it possible to use AppViewLite in combination with other clients.
Some further thoughts:
The way ATProto works, is that it takes the software that runs a social network and splits it up into separate components, with each of those components being able to be run independently. This has made self-hosting any component possible since the beginning of the network opening up. But to tak advantage of this, and get to a state of full independence, it means running multiple pieces of software. This has created a bit of a catch-22 in the ecosystem: you could run your own relay, but without another independent AppView to take advantage of this, it is not super useful. You could run your own (focused on the Bluesky lexicon) AppView, but without a client that allows you to set your own AppView it is not particularly useful either. What happened now in the last weeks is that all these individual pieces are starting to come together. With Deer allowing you to set your own custom AppView, there is now a use to actually run your own AppView. Which in turn also gives more purpose to running your own relay.
For building features in a Bluesky client that Bluesky itself does not have, a different AppView is needed. Now that these are starting to become available, there is new space to experiment with clients that have features that Bluesky does not have. Deer has already started going in this direction by allowing people to set any account as a trusted verifier, for example.
There has been skepticism around Bluesky PBC’s claims regarding decentralisation, especially from people within the ActivityPub community. Part of this distrust has come from people applying a mental framework of how ActivityPub works to how ATProto works. In this framework, Bluesky being decentralised would mean that there are other software platforms that are interoperable with the Bluesky lexicon. I’ll be writing more about those different mental frameworks, and how that relates to decentralisation later. But for now these developments strengthen the claims of Bluesky PBC around decentralisation and building a network that is ‘billionaire-proof’.
In Other News
at://2048 is the game of 2048, integrated with ATProto. 2048 is a sliding tile puzzle game where players combine numbered tiles to reach the 2048 tile, that has gotten popularity years ago and has been reimplemented a number of times. What makes the at://2048 version stand out is that the scores of the game are stored on your ATProto PDS. This creates new features and challenges: it gives the game a more social element, with features like leaderboards. It also creates a new challenge, of how to verify that a score on someone’s PDS is actually legit. at://2048 is experimenting with verified badges to authenticate if a score is legit. Integrating games with ATProto is one of the areas that is under-explored, and this reimplementation of 2048 is worth watching to get a sense of how the integration of games with ATProto will further develop.
Bluesky differs from other social networks in one significant way, namely that users blocking each other is public information. This creates new dynamics, from people being able to see who have blocked them, to leaderboards of the most blocked accounts on the network. A new paper, ‘Self-moderation in the decentralized era: decoding blocking behavior on Bluesky‘, takes advantages of data on blocks being public to study user behaviour. Some of their findings: “users who receive a high number of blocks exhibit distinctive behavioral traits that set them apart from the general user population. These patterns are not necessarily linked to toxicity or misinformation, indicating that block-worthy behavior is more nuanced and complex than traditional moderation markers might suggest. Second, these distinctive traits can be effectively encoded and leveraged by machine learning models, suggesting the feasibility of early-warning or flagging systems able to assist moderation teams by surfacing potentially problematic users even before issues escalate.”
Custom feed builder Graze is giving out 5 grants of 1k USD for other projects in the ATProto ecosystem. Explaining why the startup is giving out grants, Graze says: “First, we want to help accelerate growth in the ATProto / Bluesky ecosystem. Projects that help *others* are vital. Second, we want to empower communities to sustain themselves. Third, we want to help give people & orgs direct access to their audiences. Broadly, those are *our* goals as an org.”
Bluesky in the media
Time Magazine talks with Bluesky CEO Jay Graber and COO Rose Wang after they both got recognised as rising leaders in the Asian Pacific Community by Gold House. On monetisation, Graber says “she’s considering subscription models or monetizing Bluesky’s marketplaces of custom tools, but no concrete plans have been set in motion.”
Wired published an article on how digital archivists are racing to save Black History while the Trump administration is trying to erase it. Wired talks with Blacksky’s Rudy Fraser, who describes “Blacksky as a living archive. Currently its database holds 17 million posts from Black users over the last two years”.
The two developers behind Git collaboration platform Tangled, the brothers Anirudh and Akshay Oppiliappan, gave an interview on the devtools.fm podcast about Tangled. The platform also got various feature updates this week, and customisable profiles.
WhiteBreeze is a self-hostable frontend for WhiteWind, allowing people to build their own blog on ATProto.
ATProto Migrator is a tool to migrate your ATProto account to a different PDS. It does so via a web application, without people having to touch the Command Line Interface (CLI). This makes account migration more accessible, as other tools until now (such as goat by Bluesky engineer Bryan Newbold) require people to use the CLI.
Flashes is a Bluesky client focused on images, and they are experimenting with some new ways to deal with the limitations that come from using Bluesky’s data. A Bluesky post can contain a maximum of 4 images and 300 characters. Flashes has upgraded that limit to 900 characters and 12 images. It works by actually creating 3 separate Bluesky posts in a thread, and displaying this as a single post in the Flashes app.
That’s all for this week, thanks for reading! If you want more analysis, you can subscribe to my newsletter. Every week you get an update with all this week’s articles, as well as extra analysis not published anywhere else. You can subscribe below, and follow this blog @fediversereport.com and my personal account @laurenshof.online on Bluesky.
These days, if you are in the market for a capable digital voltmeter, you might as well consider getting one with an oscilloscope built-in. One choice is the Owon HDS160, which [Kerry Wong] covers in the video below. The model is very similar to the HDS120, but the multimeter in the HDS160 has more counts–60,000 vs 20,000 as you might expect from the model number.
The internal chip is an HY3131, which is rated at 50,000 counts which is odd since the meter is 60,000 counts, but presumably the meter uses some capability of the chip, possibly putting it out of spec. The oscilloscope is the same between the two models. Almost everything else works the same, other than the capacitance measuring feature, as the video shows.
The difference in cost between the two units isn’t much, so if you are shopping, the small extra cost is probably worth it. Not that a 20,000 count meter isn’t perfectly fine for most normal uses.
The word “Schlieren” is German, and translates roughly to “streaks”. What is streaky photography, and why might you want to use it in a project? And where did this funny term come from?
Think of the heat shimmer you can see on a hot day. From the ideal gas law, we know that hot air is less dense than cold air. Because of that density difference, it has a slightly lower refractive index. A light ray passing through a density gradient faces a gradient of refractive index, so is bent, hence the shimmer.
Heat shimmer: the refractive index of the air is all over the place. Image: “Livestock crossing the road in Queensland, Australia” by [AlphaLemur]German lens-makers started talking about “Schelieren” sometime in the 19th century, if not before. Put yourself in the shoes of an early lensmaker: you’ve spent countless hours laboriously grinding away at a glass blank until it achieves the perfect curvature. Washing it clean of grit, you hold it to the light and you see aberration — maybe spatial, maybe chromatic. Schliere is the least colourful word you might say, but a schliere is at fault. Any wonder lens makers started to develop techniques to detect the invisible flaws they called schlieren?
When we talk of schlieren imagery today, we generally aren’t talking about inspecting glass blanks. Most of the time, we’re talking about a family of fluid-visualization techniques. We owe that nomenclature to German physicist August Toepler, who applied these optical techniques to visualizing fluid flow in the middle of the 19th century. There is now a whole family of schlieren imaging techniques, but at the core, they all rely on one simple fact: in a fluid like air, refractive index varies by density.
Toepler’s pioneering setup is the one we usually see in hacks nowadays. It is based on the Foucault Knife Edge Test for telescope mirrors. In Foucault’s test, a point source shines upon a concave mirror, and a razor blade is placed where the rays focus down to a point. The sensor, or Foucault’s eye, is behind the knife edge such that the returning light from the pinhole is interrupted. This has the effect of magnifying any flaws in the lens, because rays that deviate from the perfect return path will be blocked by the knife-edge and miss the eye.
[Toepler]’s single-mirror layout is quick and easy.Toepler’s photographic setup worked the same way, save for the replacement of the eye with a photographic camera, and the use of a known-good mirror. Any density changes in the air will refract the returning rays, and cause the characteristic light and dark patterns of a schlieren photograph. That’s the “classic” schlieren we’ve covered before, but it’s not the only game in town.
Fun Schlieren Tricks
A little color can make a big difference for any kind of visualization. (Image: “Colored schlieren image“ by [Settles1])For example, a small tweak that makes a big aesthetic difference is to replace the knife edge with a colour filter. The refracted rays then take on the colour of the filter. Indeed, with a couple of colour filters you can colour-code density variations: light that passes through high-density areas can be diverted through two different colored filters on either side, and the unbent rays can pass through a third. Not only is it very pretty, the human eye has an easier time picking up on variations in colour than value. Alternatively, the light from the point source can be passed through a prism. The linear spread of the frequencies from the prism has a similar effect to a line of colour filters: distortion gets color-coded.
A bigger tweak uses two convex mirrors, in two-mirror or Z-path schlieren. This has two main advantages: one, the parallel rays between the mirrors mean the test area can be behind glass, useful for keeping sensitive optics outside of a high-speed wind tunnel. (This is the technique NASA used to use.) Parallel rays also ensure that the shadow of both any objects and the fluid flow are no issue; having the light source off-centre in the classic schrilien can cause artifacts from shadows. Of course you pay for these advantages: literally, in the sense that you have to buy two mirrors, and figuratively in that alignment is twice as tricky. The same colour tricks work just as well, though, and was in often use at NASA. The z-fold allows for parallel rays in the test area. There’s absolutely no reason that you could not substitute lenses for mirrors, in either the Z-path or classical version, and people have to good effect in both cases. Indeed, Robert Hooke’s first experiment involved visualizing the flow of air above a candle using a converging lens, which was optically equivalent to Toepler’s classic single-mirror setup. Generally speaking, mirrors are preferred for the same reason you never see an 8” refracting telescope at a star party: big mirrors are way easier to make than large lenses. T-34s captured in flight with NASA’s AirBOS technique. Image credit : NASA. What if you want to visualize something that doesn’t fit in front of a mirror? There are actually several options. One is background-oriented schrilien, which we’ve covered here. With a known background, deviations from it can be extracted using digital signal processing techniques. We showed it working with a smart phone and a printed page, but you can use any non-uniform background. NASA uses the ground: by looking down, Airborn Background Oriented Schlieren (AirBOS) can provide flow visualization of shockwaves and vortices around an airplane in flight.
In the days before we all had supercomputers in our pockets, large-scale flow-visualization was still possible; it just needed an optical trick. A pair of matching grids is needed: one before the lamp, creating a projection of light and dark, and a second one before the lens. Rays deflected by density variations will run into the camera grid. This was used to good effect by Gary S. Styles to visualize HVAC airflows in 1997 Can’t find a big mirror? Try a grid. Which gets us to another application, separate from aerospace. Wind tunnel photos are very cool, but let’s be honest: most of us are not working on supersonic drones or rocket nozzles. Of course air flow does not have to be supersonic to create density variations; subsonic wind tunnels can be equipped with schlieren optics as well. HVAC as you’ve never seen it before. Imagine those were ABS fumes? (Image from Styles, 1997.) Or maybe you are more concerned with airflow around components? To ID a hotspot on a board, IR photography is much easier. On the other hand, if your hotspot is due to insufficient cooling rather than component failure? Schlieren imagery can help you visualize the flow of air around the board, letting you optimize the cooling paths.
That’s probably going to be easiest with the background-oriented version: you can just stick the background on one side of your project’s enclosure and go to work. I think that if any of you start using schlieren imaging in your projects, this might be the killer app that will inspire you to do so.
Another place we use air? In the maker space. I have yet to see someone use schlieren photography to tweak the cooling ducts on their 3D printer, but you certainly could. (It has been used to see shielding gasses in welding, for example.) For that matter, depending what you print, proper exhaust of the fumes is a major health concern. Those fumes will show up easily, given the temperature difference, and possibly even the chemical composition changing the density of the air.
Remember that the key thing being imaged isn’t temperature difference, but density difference. Sound waves are density waves, can they be imaged in this way? Yes! The standing waves in ultrasonic levitation rigs are a popular target. Stroboscopic effects can be used for non-standing waves, though keep in mind that the sound pressure level is the inverse of frequency, so audible frequencies may not be practical if you like your eardrums. Schlieren photography isn’t limited to air. Density variations in liquids and solids are game, too. Want to see how multiple solutions of varying density or tempeature are mixing? Schlieren imaging has you covered. Watch convection in a water tank? Or, if you happen to be making lenses, you could go right back to basics and use one of the schlieren techniques discussed here to help you make them perfect.
The real reason I’m writing about these techniques aren’t the varied applications I hope you hackers can put them to: it’s an excuse to collect all the pretty pictures of flow visualization I can cram into this article. So if you read this and thought “I have no practical reason to use this technique, but it does seem cool” – great! We’re in the same boat. Let’s make some pretty pictures. It still counts as a hack.
The iMac G3 is an absolute icon of industrial design, as (or perhaps more) era-defining than the Mac Classic before it. In the modern day, if your old iMac even boots, well, you can’t do much with it. [Rick Norcross] got a hold of a dead (hopefully irreparable) specimen, and stuffed a modern PC inside of it.
From the outside, it’s suprizingly hard to tell. Of course the CRT had to go, replaced with a 15″ ELO panel that fits well after being de-bezeled. (If its resolution is only 1024 x 768, well, it’s also only 15″, and that pixel density matches the case.) An M-ATX motherboard squeezes right in, above a modular PSU. Cooling comes from a 140 mm case fan placed under the original handle. Of course you can’t have an old Mac without a startup chime, and [Rick] obliges by including an Adafruit FX board wired to the internal speakers, set to chime on power-up while the PC components are booting.
These sorts of mods have proven controversial in the past– certainly there’s good reason to want to preserve aging hardware–but perhaps with this generation of iMac it won’t raise the same ire as when someone guts a Mac Classic. We’ve seen the same treatment given to a G4 iMac, but somehow the lamp doesn’t quite have the same place in our hearts as the redoubtable jellybean.
Detecting a signal pulse is usually basic electronics, but you start to find more complications when you need to time the signal’s arrival in the picoseconds domain. These include the time-walk effect: if your circuit compares the input with a set threshold, a stronger signal will cross the threshold faster than a weaker signal arriving at the same time, so stronger signals seem to arrive faster. A constant-fraction discriminator solves this by triggering at a constant fraction of the signal pulse, and [Michael Wiebusch] recently presented a hacker-friendly implementation of the design (open-access paper).
A constant-fraction discriminator splits the input signal into two components, inverts one component and attenuates it, and delays the other component by a predetermined amount. The sum of these components always crosses zero at a fixed fraction of the original pulse. Instead of checking for a voltage threshold, the processing circuitry detects this zero-crossing. Unfortunately, these circuits tend to require very fast (read “expensive”) operational amplifiers.
This is where [Michael]’s design shines: it uses only a few cheap integrated circuits and transistors, some resistors and capacitors, a length of coaxial line as a delay, and absolutely no op-amps. This circuit has remarkable precision, with a timing standard deviation of 60 picoseconds. The only downside is that the circuit has to be designed to work with a particular signal pulse length, but the basic design should be widely adaptable for different pulses.
[Michael] designed this circuit for a gamma-ray spectrometer, of which we’ve seen a few examples before. In a spectrometer, the discriminator would process signals from photomultiplier tubes or scintillators, such as we’ve covered before.
Of all FDM filament types, flexible ones such as TPU invite a whole new way of thinking, as well as applications. Case in point the TPU-based bellows that the [Functional Part Friday] channel on YouTube recently demonstrated.
The idea is quite straightforward: you print TPU and PLA in alternating layers, making sure that the TPU is connected to its previous layer in an alternating fashion. After printing, you peel the PLA and TPU apart, remove the PLA layers and presto, you got yourself bellows.
There were some issues along the way, of course. Case in point the differences between TPU from different brands (Sainsmart, Sunlu) that caused some headaches, and most of all the incompatibility between the Bambu Lab AMS and TPU that led to incredibly brittle TPU prints. This required bypassing the feed mechanism in the AMS, which subsequently went down a rabbit hole of preventing the PTFE tube from getting sucked into the AMS. Being able to print TPU & PLA at the same time also requires a printer with two independent extruders like the Bambu Lab H2D used here, as both materials do not mix in any way. Great news for H2D and IDEX printer owners, of course.
As for practical applications for bellows, beyond printing your own 1900s-era camera, accordion or hand air bellows, you can also create lathe way covers and so on.
It might seem strange to people like us, but normal people hate wires. Really hate wires. A lot. So it makes sense that with so many wireless technologies, there should be a way to do USB over wireless. There is, but it really hasn’t caught on outside of a few small pockets. [Cameron Kaiser] wants to share why he thinks the technology never went anywhere.
Wireless USB makes sense. We have high-speed wireless networking. Bluetooth doesn’t handle that kind of speed, but forms a workable wireless network. In the background, of course, would be competing standards.
Texas Instruments and Intel wanted to use multiband orthogonal frequency-division multiplexing (MB-OFDM) to carry data using a large number of subcarriers. Motorola (later Freescale), HP, and others were backing the competing direct sequence ultra-wideband or DS-UWB. Attempts to come up with a common system degenerated.
This led to two systems W-USB (later CF-USB) and CW-USB. CF-USB looked just like regular USB to the computer and software. It was essentially a hub that had wireless connections. CW-USB, on the other hand, had cool special features, but required changes at the driver and operating system level.
Check out the post to see a bewildering array of orphaned and incompatible products that just never caught on. As [Cameron] points out, WiFi and Bluetooth have improved to the point that these devices are now largely obsolete.
Of course, you can transport USB over WiFi, and maybe that’s the best answer, today. That is, if you really hate wires.
In a study published in Physical Review Letters that was co-authored by [Babak Seradjeh], a Professor of Physics at Indiana University Bloomington, and theoretical physicists [Rekha Kumari] and [Arijit Kundu], from the Indian Institute of Technology Kanpur, the scientists validate their theory using numerical simulations.
In the absence of room-temperature superconductors — the Holy Grail of superconductivity, everybody put your thinking caps on! — the low temperatures required lead to expense (for cooling) and errors (due to decoherence) which need to be managed. Using the techniques proposed by the study, quantum information may be modeled non-locally and be spread out spatially in a material, making it more stable and less error prone, immune to local noise and fluctuations.
Majorana fermions are named after Italian physicist [Ettore Majorana] who proposed them in 1937. Unlike most particles, Majorana fermions are their own antiparticles. In the year 2000 mathematical physicist [Alexei Kitaev] realized Majorana fermions can exist not only as elementary particles but also as quantum excitations in certain materials known as topological superconductors. Topological superconductors differ from regular superconductors in that they have unique, stable quantum states on their surface or edges that are protected by the material’s underlying topology.
Superconductivity is such an interesting phenomenon, where electrical resistance all but vanishes in certain materials when they are very cold. Usually to induce a current in a material you apply a voltage, or potential difference, in order to create the electrical pressure that results in the current. But in a superconductor currents can flow in the absence of an applied voltage. This is because of a peculiar quantum tunneling process known as the “Josephson effect”. It is hoped that by tuning the Josephson current using a superconductor’s “chemical potential” that we discover a new level of control over quantum materials.
Ettore Majorana picture: Mondadori Collection, Public domain.
È stata individuata una campagna di phishing mirata agli utenti SPID dal gruppo del CERT-AgID, che sfrutta indebitamente il nome e il logo della stessa AgID, insieme al dominio recentemente registrato agidgov[.]com, non riconducibile all’Agenzia. Il messaggio fraudolento, con oggetto “Sospensione imminente SPID: azione obbligatoria“, invita l’utente ad aggiornare la propria documentazione, inducendolo a cliccare su un pulsante etichettato “Aggiorna la Documentazione“, che rimanda al sito malevolo. L’obiettivo della campagna è sottrarre le credenziali SPID delle vittime, insieme a copie di documenti di identità e a video registrati secondo istruzioni specifiche per la procedura di riconoscimento, come: “Guarda verso la telecamera. Rimani serio, poi sorridi“.
Azioni di contrasto
È stata richiesta la disattivazione del dominio malevolo al fine di prevenire ulteriori compromissioni. Gli IoC relativi alla campagna sono stati diramati attraverso il Feed IoC del CERT-AGID verso le strutture accreditate.
Si raccomanda di prestare sempre la massima attenzione a questo tipo di comunicazioni, in particolare quando contengono collegamenti ritenuti sospetti. Nel dubbio, è sempre possibile inoltrare le email ritenute
Phishing sempre più sofisticato: l’AI al servizio della truffa
Il fenomeno del phishing si è evoluto drasticamente negli ultimi anni, grazie anche all’impiego dell’intelligenza artificiale per generare siti web contraffatti quasi indistinguibili dagli originali. Questi portali imitano in modo sorprendentemente accurato la grafica, il linguaggio e il comportamento dei siti ufficiali di enti pubblici o aziende private, rendendo estremamente difficile per l’utente medio accorgersi del raggiro.
Nel caso specifico segnalato dal CERT-AGID, il dominio fraudolento agidgov[.]com riproduceva fedelmente il layout e i contenuti del sito dell’Agenzia per l’Italia Digitale, inducendo l’utente a inserire le proprie credenziali SPID in un ambiente che appariva del tutto legittimo.
Fidarsi… ma verificare
Per difendersi, non basta più fare attenzione ai soli errori grammaticali o ai loghi sgranati. È fondamentale conoscere le normali modalità operative delle agenzie e delle aziende:
Le istituzioni pubbliche non richiedono mai via email o SMS l’inserimento diretto delle credenziali SPID.
Diffidare da messaggi che inducono urgenza o paura per costringere l’utente ad agire impulsivamente.
Se un messaggio o una pagina sembra sospetta, è sempre bene controllare l’indirizzo web (URL) e confrontarlo con quello ufficiale o, in caso di dubbio, contattare direttamente l’ente coinvolto tramite i canali ufficiali.
La scorsa notte, il gruppo ransomwareLockBit ha subito un grave attacco informatico che ha compromesso la sua infrastruttura nel dark web. Gli affiliati e gli amministratori del gruppo hanno trovato i loro pannelli di controllo compromessi e le home modificate con il messaggio: “Don’t do crime CRIME IS BAD xoxo from Prague”, accompagnato da un link per scaricare un file denominato “paneldb_dump.zip” contenente un dump del database MySQL del gruppo.
L’archivio trapelato include informazioni altamente sensibili, tra cui:
59.975 indirizzi Bitcoin utilizzati per le transazioni del gruppo.
4.442 messaggi di negoziazione tra LockBit e le sue vittime, datati tra dicembre 2024 e aprile 2025.
Configurazioni dei ransomware utilizzati negli attacchi, inclusi dettagli su quali file o sistemi evitare.
Elenco di 75 affiliati e amministratori, con password in chiaro come “Weekendlover69” e “Lockbitproud231”.
Il leader del gruppo, noto come “LockBitSupp”, ha confermato la violazione, affermando che non sono state compromesse chiavi private o dati critici. Da un’analisi veloce del dump SQL abbiamo notato che il database è stato esfiltrato il 29 Aprile, quindi possiamo ragionevolmente supporre che in quella data xoxo From Prague (o chiunque ci sia dietro a questo data leak) abbia dumpato il database e solo nella notte fra il 7 e l’8 Maggio sia stato eseguito il deface dei siti.
Questo attacco rappresenta un duro colpo per LockBit, già indebolito da precedenti operazioni delle forze dell’ordine, come “Operation Cronos”, che aveva portato al sequestro di server, arresti e sanzioni internazionali.
La fuga di dati offre agli esperti di sicurezza e alle autorità un’opportunità unica per analizzare le operazioni interne di LockBit e potrebbe accelerare ulteriori azioni legali contro i suoi membri.
Per adesso chiudiamo l’articolo con una dichiarazione di LockBitSUP
!!! LockBitSupp statement: “It’s Not Scary to Fall – It’s Scary Not to Get Up”
Gli autori della minaccia collegati all’operazione ransomware Play hanno sfruttato una vulnerabilitàzero-day in Microsoft Windows prima della sua correzione, avvenuta l’8 aprile 2025.
Il Threat Intelligence Center (MSTIC) e il Security Response Center (MSRC) di Microsoft ha scoperto che l’attività di sfruttamento la quale è stata attribuita a un gruppo di minacce denominato Storm-2460, che distribuisce il malwarePipeMagic in campagne ransomware.
“Gli autori di minacce ransomware apprezzano gli exploit di elevazione dei privilegi post-compromissione perché potrebbero consentire loro di trasformare l’accesso iniziale in un accesso privilegiato”, ha affermato Microsoft nel suo avviso di sicurezza.
Si tratta del CVE-2025-29824, un bug del driver CLFS (Common Log File System) di Windows e consente agli aggressori di elevare i propri privilegi da utente standard ad accesso completo al sistema.
L’analisi tecnica ha rivelato che lo sfruttamento prevedeva una sofisticata catena di attacchi. La vulnerabilità risiede nel driver del kernel CLFS e consente agli aggressori di sfruttare una condizione di “use-after-free”. Durante l’esecuzione dell’exploit, gli aggressori hanno creato file nel percorso C:\ProgramData\SkyPDF, inclusa una DLL che è stata iniettata nel processo winlogon.exe.
Ciò ha consentito loro di estrarre le credenziali dalla memoria LSASS utilizzando strumenti come Sysinternals procdump.exe, creare nuovi utenti amministratori e stabilire la persistenza.
Il Symantec Threat Hunter Team ha segnalato che gli aggressori affiliati al gruppo ransomware Play (noto anche come Balloonfly o PlayCrypt) hanno preso di mira un’organizzazione non identificata negli Stati Uniti, probabilmente utilizzando un Cisco Adaptive Security Appliance (ASA) accessibile al pubblico come punto di ingresso.
Sebbene nell’intrusione scoperta non sia stato impiegato alcun payload ransomware, gli aggressori hanno utilizzato uno strumento personalizzato per il furto di informazioni chiamato Grixba, precedentemente associato all’operazione ransomware Play.
La vulnerabilità, che ha ricevuto un punteggio CVSS di 7,8 (Alto), è stata risolta nell’ambito degli aggiornamenti Patch Tuesday di aprile 2025 di Microsoft , che ha corretto un totale di 121 vulnerabilità.
Il gruppo ransomware Play, attivo da giugno 2022, è noto per l’impiego di tattiche di doppia estorsione, in cui i dati sensibili vengono esfiltrati prima della crittografia.
Questa iniziativa, della durata di tre mesi, punta a standardizzare l’utilizzo delle applicazioni AI, promuovere uno sviluppo etico del settore e proteggere i diritti legittimi dei cittadini cinesi. L’operazione è divisa in due fasi: la prima mira a controllare le fonti tecnologiche e rettificare gli strumenti IA illegali, migliorando la capacità delle piattaforme di identificare e bloccare contenuti sintetici pericolosi; la seconda, invece, si concentrerà sulla rimozione di contenuti falsi, pornografici, impersonificazioni digitali e attività online manipolatorie, punendo account e organizzazioni coinvolte.
Nella prima fase, sei problemi principali sono nel mirino delle autorità: dalla diffusione di software illegali come tool di “svestizione in un clic” e sintetizzatori vocali non autorizzati, alla mancanza di gestione dei dataset usati per l’addestramento dei modelli IA, spesso tratti da fonti non verificate o illegali.
A questi si aggiungono gravi carenze nei sistemi di sicurezza delle piattaforme, l’assenza di identificazione chiara dei contenuti generati e i rischi in settori sensibili come medicina, finanza ed educazione, dove l’AI viene impiegata senza adeguati controlli, generando “prescrizioni AI” o “investimenti fittizi” con effetti potenzialmente disastrosi.
La seconda fase si occupa invece di ripulire il web da sette categorie di contenuti pericolosi: tra questi, la generazione e diffusione di voci false su politica e attualità, la manipolazione di notizie tramite deepfake, la pubblicazione di contenuti pseudoscientifici e superstiziosi, e la produzione di materiale pornografico e violento tramite AI.
Altre pratiche sotto osservazione sono l’utilizzo di intelligenza artificiale per impersonare personaggi famosi o defunti a fini fraudolenti, il controllo delle conversazioni online tramite bot e content farm automatizzate, e l’uso di IA per creare piattaforme contraffatte o servizi seducenti rivolti a minori, con conseguente rischio di dipendenza e danni psicologici.
L’intervento dell’autorità cinese non si limita alla rimozione: tutte le piattaforme sono obbligate a implementare strumenti di rilevamento dei contenuti AI, effettuare revisioni regolari e rafforzare i controlli su API e chatbot. Gli amministratori locali del cyberspazio hanno ricevuto l’ordine di vigilare e intervenire attivamente, promuovendo al contempo la divulgazione scientifica e l’alfabetizzazione tecnologica della popolazione.
Con questa campagna, la Cina mira a consolidare un modello di governance dell’intelligenza artificiale che non solo freni gli abusi, ma educhi gli utenti, responsabilizzi gli sviluppatori e imponga limiti chiari a una tecnologia che, se lasciata incontrollata, può trasformarsi da risorsa a minaccia.
Il messaggio è chiaro: l’era dell’IA selvaggia sta finendo, almeno in Cina.
Il 6 maggio, l’Università Sun Yat-sen ha ospitato una conferenza dedicata allo sviluppo dell’intelligenza artificiale, durante la quale è stato ufficialmente inaugurato l’Istituto di Ricerca sull’Intelligenza Artificiale e annunciata la creazione di 17 nuovi centri di ricerca. Qian Depei, accademico dell’Accademia Cinese delle Scienze e primo preside della Facoltà di Informatica dell’università, presiederà il Comitato Accademico dell’Istituto.
Chen Hongbo, vicepresidente esecutivo dell’Istituto, ha spiegato che l’iniziativa integra le competenze scientifiche presenti nei vari dipartimenti dell’ateneo, articolandosi su tre livelli: “materia”, “fondamento” e “applicazione” dell’intelligenza artificiale.
L’obiettivo è affrontare le sfide strategiche nazionali, valorizzare i vantaggi industriali della Greater Bay Area e promuovere ambiti tecnologici chiave come i modelli multimodali di grandi dimensioni, i chip neuromorfici a basso consumo, i sistemi autonomi, l’economia a bassa quota e altri settori emergenti. Il fine ultimo è creare un ecosistema su larga scala che unisca industria, ricerca, accademia e applicazione. La Cina sembra oramai oggi arrivata al pareggio con gli Stati Uniti, tanto che il NYT ha riportato questi traguardi raggiunti in appena 19 mesi da parte della Cina con un articolo che riporta “La posta in gioco di questa competizione è alta. Le principali aziende statunitensi hanno in gran parte sviluppato modelli di intelligenza artificiale proprietari e addebitato royalties per il loro utilizzo, in parte perché addestrare i loro modelli costa centinaia di milioni di dollari. Le aziende cinesi di intelligenza artificiale stanno espandendo la loro influenza rendendo disponibili gratuitamente i loro modelli al pubblico, che può utilizzarli, scaricarli e modificarli, rendendoli così più accessibili a ricercatori e sviluppatori di tutto il mondo.”
I 17 centri di ricerca copriranno una vasta gamma di settori interdisciplinari tra arti, scienze, medicina e ingegneria. Tra le aree di interesse figurano: calcolo scientifico ad alte prestazioni, fondamenti matematici dell’IA, chip e sistemi intelligenti, dispositivi di rilevamento ispirati al cervello umano, software intelligenti, modelli multi-agente e intelligenza incarnata, IA applicata ai big data medici e intelligenza collettiva.
Gao Song, presidente dell’Università e anch’egli accademico dell’Accademia Cinese delle Scienze, ha sottolineato il duplice approccio dell’ateneo: da un lato, rafforzare la ricerca teorica e lo sviluppo di tecnologie chiave come chip avanzati e software di base; dall’altro, utilizzare l’intelligenza artificiale per guidare un cambiamento di paradigma nella ricerca scientifica, promuovendo innovazioni tecnologiche rivoluzionarie in più settori.
Nel corso dell’evento è stato presentato anche il Piano di Lavoro per la Promozione dell’Intelligenza Artificiale, che include 15 iniziative suddivise in tre ambiti: formazione dei talenti, innovazione scientifica e tecnologica, e governance. L’università prevede di consolidare le risorse informatiche, migliorare i meccanismi di supporto e creare un ambiente favorevole per lo sviluppo dell’IA e la valorizzazione dei talenti.
Zhu Kongjun, segretario del comitato di partito dell’ateneo, ha dichiarato che, in quanto istituzione di riferimento della Greater Bay Area del Guangdong-Hong Kong-Macao, la Sun Yat-sen University si assume la responsabilità di guidare lo sviluppo strategico dell’IA, con un focus sull’autosufficienza tecnologica, sull’innovazione di base e sull’applicazione concreta al servizio degli obiettivi nazionali.
Fondato nel giugno 2020, l’Istituto di Ricerca sull’Intelligenza Artificiale ha ampliato ulteriormente le proprie attività nel dicembre 2024, entrando in piena operatività con una sede di oltre 40.000 metri quadrati e numerose piattaforme sperimentali di livello mondiale.
A screenshot provided by the hacker. Redactions by 404 Media.
A screenshot provided by the hacker. Redactions by 404 Media.
eutampieri
in reply to Informa Pirata • • •Eugenio
in reply to Informa Pirata • • •