Please try the URL privacy information feature enabled by clicking the flashlight icon above. This will reveal two icons after each link the body of the digest. The shield takes you to a breakdown of Terms of Service for the site - however only a small number of sites are covered at the moment. The flashlight take you to an analysis of the various trackers etc. that the linked site delivers. Please let the website maintainer know if you find this useful or not. As a RISKS reader, you will probably not be surprised by what is revealed…
Pranshu Verma and Cat Zakrzewski, The Washington Post
Elections from India to Europe have been assailed by AI deepfakes that spread quickly and are no longer easy to debunk—leaving voters vulnerable.
https://www.washingtonpost.com/technology/2024/04/23/ai-deepfake-election-2024-us-india/
In March 2023, a North Carolina student was stepping off a school bus when he was struck by a Tesla Model Y traveling at “highway speeds,” according to a federal investigation that published today. The Tesla driver was using Autopilot, the automaker’s advanced driver-assist feature that Elon Musk insists will eventually lead to fully autonomous cars.
The 17-year-old student who was struck was transported to a hospital by helicopter with life-threatening injuries. But what the investigation found after examining hundreds of similar crashes was a pattern of driver inattention, combined with the shortcomings of Tesla’s technology, resulting in hundreds of injuries and dozens of deaths.
Drivers using Autopilot or the system’s more advanced sibling, Full Self-Driving, “were not sufficiently engaged in the driving task,” and Tesla’s technology “did not adequately ensure that drivers maintained their attention on the driving task,” NHTSA concluded.
Raphael Satter, Reuters, 24 Apr 2024, via ACN TechNews
Cisco Systems on Wednesday said that hackers have subverted some of its digital security devices to break into government networks globally. In a blog post, Cisco said its Adaptive Security Appliances had previously unknown vulnerabilities that had been exploited by a group of hackers they dubbed “UAT4356.” The company described the group as a “sophisticated state-sponsored actor.” Cisco said the vulnerabilities have been patched.
Ellesheva Kissin, Financial Times, 24 Apr 2024, via ACN TechNews
Cybersecurity firm Performanta reported that businesses in Africa, Asia, and South America increasingly are being used by hackers as testing grounds for their latest ransomware before they turn to higher-value targets in North America and Europe. Recent dry runs in developing countries focused on a Senegalese bank, a Chilean financial services company, a Colombian tax firm, and a government economic agency in Argentina.
https://gizmodo.com/airbnb-hidden-cameras-shocking-stories-bedroom-night-1851433108
Airbnb announced in March that all indoor security cameras would be banned at its properties worldwide starting April 30. And if you read through online complaints about cameras that were discovered during Airbnb stays over the years, it’s easy to understand why it’s been such a controversial issue.
Gizmodo filed a Freedom of Information Act request with the FTC for any consumer complaints filed about Airbnb that involved cameras. Some of the complaints are fairly mundane, and simply mention how cameras may have been used to prove things that break the rules at Airbnb properties. But others are pretty horrifying and involve hidden cameras in places where people expect privacy.
https://arstechnica.com/?p=2020055
Kate Kelly and Julian E. Barnes. The New York Times, 25 Apr 2024, via ACN TechNews
The Countering CCP Drones Act, under consideration by the U.S. Congress, would threaten the commercial business of DJI, a Chinese drone manufacturer that dominates sales among U.S. law enforcement agencies and hobbyists. The legislation would put the company on a Federal Communications Commission roster that would prevent it from running on U.S. communications infrastructure. Researchers found vulnerabilities in an app that controls DJI's drones could be used to access personal data (a U.S. official said all known vulnerabilities currently have been patched).
Last week, the 9th Circuit Court of Appeals in California released a ruling that concluded state highway police were acting lawfully when they forcibly unlocked a suspect's phone using their fingerprint. You probably didn’t hear about it. The case didn’t get a lot of coverage, especially because the courts weren’t giving a blanket green light for every cop to shove your thumb to your screen during an arrest. But it’s another toll of the warning bell that reminds you to not trust biometrics to keep your phone’s sensitive info private. In many cases, especially if you think you might interact with the police (at a protest, for example), you should seriously consider turning off biometrics on your phone entirely. https://gizmodo.com/stop-using-your-face-or-thumb-to-unlock-your-phone-1851438205
Google's LLM AI SGE (“Search Generative Experience”) could effectively destroy the Internet for all but the largest sites—the same Internet that #Google so effectively helped to build.
This is becoming clear as SGE rolls out to most users, with SGE “answers” now appearing on a vast number of Google queries. Leaving aside the serious questions around the accuracy of such responses and everything associated with that, the mere presence of the responses could be devastating to most sites.
These SGE answers are frequently verbose and can take up much of the entire first screen—or more—of the results pages. This means you may have to scroll down to even FIND the first organic “blue link” results. Devastating.
To be clear, many of the SGE responses are themselves showing links to the answers' source materials (e.g., in colored boxes)—but the obvious question is, why the hell would most users bother to click on those links once they already have the answers that Google's LLM has provided, based on the information that Google sucked without compensation into their LLM from those sites? It's impossible to imagine that click through rates to those sites won't be crushed.
Google executives appear to be thrilled with how well this is going — FOR THEM. For the sites providing the data that is now powering Google's SGE encroaching, destructive storm, it's likely going to be a disaster, unless Google and other AI firms make major changes in their deployment models—whether voluntarily or under the force of new regulatory models. -L
The Washington Post, 26 Apr 2026 https://www.washingtonpost.com/technology/2024/04/26/amazon-ftc-messages-de=ted-bezos/
The Federal Trade Commission is accusing Amazon founder Jeff Bezos and other top company executives of using disappearing messaging apps such as Signal to conceal potential evidence in the agency's ongoing antitrust case against the e-commerce behemoth.
“For years, Amazon's top executives, including founder and former CEO Jeff Bezos, discuss[ed] sensitive business matters, including antitrust, over the Signal encrypted-messaging app instead of email,” the FTC alleged in a document filed Thursday evening. “These executives turned on Signal's disappearing message feature, which irrevocably destroys messages, even after Amazon was on notice that Plaintiffs were investigating its conduct.”
The agency, which first accused Amazon of intentionally deleting messages in its original antitrust complaint last fall, is now asking a U.S. District Court judge to order the company to turn over documents related to its handling of data. It's the latest salvo in a landmark case in which the FTC is arguing that Amazon abused its dominance of e-commerce to squeeze merchants and bury rivals, leading to higher prices for Bezos owns The Washington Post.
“The FTC's contentions are baseless,” Amazon spokesman Tim Doyle said in a statement, responding to the filing alleging destruction of evidence. “Amazon voluntarily disclosed employees' limited Signal use to the FTC years ago, thoroughly collected Signal conversations from its employees' phones, and allowed agency staff to inspect those conversations even when they had nothing to do with the FTC's investigation. The FTC has a complete picture of Amazon's decision-making in this case, including 1.7 million documents from sources like email, internal messaging applications, and laptops (among other sources), and over 100 terabytes of data.” [..]
https://www.theverge.com/2024/4/26/24141801/ftc-amazon-antitrust-signal-ephemeral-messaging-evidence
https://arstechnica.com/gadgets/2024/04/ai-marketing-hype-is-coming-for-your-favorite-gadgets
https://techcrunch.com/2024/04/25/kaiser-permanente-health-plan-million s-data-breach/
https://gizmodo.com/chaturbate-porn-age-verification-law-ken-paxton-pornhub-1851439770
https://arstechnica.com/?p=2020252
https://techcrunch.com/2024/04/25/kaiser-permanente-health-plan-million s-data-breach/
The first few minutes of the Pearl Harbor attack which caused the U.S. entry into WWII sadly proved Billy Mitchell 100% correct. The good news re Dec. 7th—if there was any—was that no U.S. aircraft carriers were in Pearl Harbor that day.
Ditto with the ‘Millennium Challenge 2002’ wargames, in which essentially the entire U.S. Mediterranean fleet was ‘virtually’ sunk within days using 'asymmetric warfare'.
Ditto with ‘cheap drones’ in the current Ukraine war; they have rebalanced the battle between infantry—now equipped with cheap drones for surveillance and attack—and tanks—a balance which has existed for a century since the battle tank appeared near the end of WWI.
Cheap drones put into serious question most—if not almost all—of the ‘prevailing wisdom’ re strategy/tactics/weapons of modern warfare. These put big ‘?’ marks next to every ‘big ticket’ asset in modern warfare — from $billion ships/aircraft carriers, to $100m fighters, to $10m battle tanks, to $billion spy satellites. In chess terminology, coordinated pawns beat rare expensive bishops, rooks, and queens.
https://nationalinterest.org/blog/reboot/exact-day-navy-battleships-became-obsolete-clear-209558
“In 1921, General Billy Mitchell, a vocal advocate of airpower, staged a controversial exercise sinking obsolete battleship with bombers. This foreshadowed the dominance of aircraft carriers in World War II despite Mitchell's goal of a separate air force. The Navy initially dismissed his claims, but the sinking of the ‘unsinkable’ German battleship Ostfriesland proved the vulnerability of battleships.”
“Mitchell believed that aviation—which could respond to both air and naval threats—much better suited to protecting the country's coastline than battleships. Mitchell was fond of stating that a thousand bombers could be purchased for the cost of a single battleship, and told a House subcommittee that properly equipped, an Air Service could sink any battleship in existence.”
https://www.msn.com/en-us/news/world/ar-AA1nIxGp
“Cheap Russian drones overwhelm US-made Abrams tanks, taken out of action”
“Ukrainian forces are withdrawing US-provided Abrams M1A1 main battle tanks from the front lines after at least five have been destroyed by cheap Russian drones, according to the AP.”
“The failure of the Abrams to make a difference is a costly miscalculation. The export cost of an Abrams tank can be around $10mn, while Col. Markus Reisner, an Austrian military trainer who follows the weapons being used in Ukraine, told the Euromaidan Press that the Russian suicide drones being used to destroy them can be as cheap as $500 each (a ratio of 20,000:1).”
https://en.wikipedia.org/wiki/Millennium_Challenge_2002
“In a preemptive strike, Red launched a massive salvo of cruise missiles that overwhelmed the Blue forces' electronic sensors and destroyed sixteen warships: one aircraft carrier, ten cruisers and five of Blue's six amphibious ships. An equivalent success in a real conflict would have resulted in the deaths of over 20,000 service personnel. Soon after the cruise missile offensive, another significant portion of Blue's navy was ‘sunk’ by an armada of small Red boats, which carried out both conventional and suicide attacks that capitalized on Blue's inability to detect them as well as expected.”
https://www.theguardian.com/world/2002/sep/06/usa.iraq
“In the first few days of the [Millennium Challenge] exercise, using surprise and unorthodox tactics, the wily 64-year-old Vietnam veteran sank most of the US expeditionary fleet in the Persian Gulf, bringing the US assault to a halt.”
All of the recent discussions of the risks of AI bring to mind the incredibly prescient movie (& radio play) ‘Forbidden Planet’:
https://en.wikipedia.org/wiki/Forbidden_Planet
In addition to being a pretty decent takeoff on Shakespeare's ‘The Tempest’, the movie version of ‘Forbidden Planet’ introduces us to talking robots (now almost passe !), and incredibly power- hungry planet-sized data centers capable of turning human thoughts into reality.
Amazingly, this 1956 movie still holds up for modern viewers, thanks to the supplanting of typical cheesy 1950's scifi effects in favor of laserlike animations and electronic music.
The risks of AI, according to this movie: be very careful what you wish for, because an AI with access to planet-sized energy capabilities can fulfill even your worst nightmare.
Your choice: watch it again (safely) in movie form, or watch it play out in real life.
BTW, I listened as a young boy to a radio serial version of Forbidden Planet during the summer of either 1955 or 1956; but after extensive Google searching, I have been unable to find a reference to this radio play version. I know exactly where I was while listening to it on my grandmother's huge radio with quite decent fidelity; perhaps someone else here also heard it at the same time?
Please report problems with the web pages to the maintainer