sudo_ml @sudo_ml

Good and bad

Frank Braun @thefrankbraun

11 hours ago

If humans don't get the economic value of private money, AI agents will.

2 6 44 2K 2

Young people actively hate AI. My two younger daughters are 19 and 18, both at universities - one in the UK the other in Italy. Through them, I happen to know quite a few kids their age. And ALL OF THEM seem to universally hate AI. To them it's "something old people invented that we now have to deal with... as if climate change wasnt enough" AI is cringe. When they introduce me to their friends, they literally say "heyya, meet my dad, he's a software engineer but nonono, NOT IN AI, dont worry, hahah". They hate it for flooding their social feeds with slop, ruining music, taking phone scams to the next level, also destroying careers & devaluing their hobbies - why become an architect, this skill is about to become worthless? Why learn a musical instrument if session musicians are already obsolete? The outcome is interesting, though - they're spending less and less time on public social, meeting up IRL, moving their online lives into private group chats with actual verified friends. TikTok and Reels are increasingly cringe-something for old dudes. Which, I guess, might be a net positive after all...

Introducing Safe Rollup. Ethereum-anchored, operator-controlled. Safe as in Safe multisig, not safe funds. Safe Rollup gives institutions the credibility of Ethereum and the control of a database. Live on Sepolia today. saferollup.fyi

Just another Sunday afternoon in SF: I paid $60 for 2 humanoid robots to clean for an hour. Here's how it went: (1) What happened: Sergey and Chelsea (the names of the two robots) finished about 80% of one floor in an hour: tidying, wiping down surfaces, vacuuming, mopping, cleaning the toilet. - They came with a staff engineer at the company, who dropped them off and picked them up after an hour. - They were controlled by 1 human operator each, in the Mission where the company office is at. - The engineer gave me a number to text them, I texted a request to wipe the mirror. They did it. (2) Performance and dexterity: Chelsea dropped a toy, struggled with a pillow, and had trouble balancing on the playmat. But overall, for all the tasks they did, grippers were enough. 5 fingers were not needed. - The engineer told me the hardest task for them was making bed. - They were quite loud, stomping around the house. Took a bit of time to get used to. They both wore gripper socks which protected the floors and helped them not slip. - Interaction: I didn't know they can respond via voice. In the end they were conversing with us in voice which was nice. (3) Unexpected observation: Their pants and shirts have holes in them. It's because they were dragged on and off the vans that transported them. (4) Other user feedback: I shared my experience with some friends afterwards. A user told me that unloading dishwasher was an "advanced task". They didn't attempt at mine since it was running at the time they were working. Another user said they scratched their hardwood floor. That's probably why they now wear gripper socks. (5) Did they do as well as human cleaners? In terms of the tasks they did, I think so. In terms of speed, they were ~80% speed of humans when it comes to the more delicate tasks, but I would say they are quite smooth. They didn't finish the whole place, so it wasn't a full service. (6) How does the math work out? Each booking was only for 1 hour, since their batteries need to be swapped after 1 hour. So they didn't finish the place. For a user: In SF, it's ~$200-220 to clean a full home around my neighborhood, with two human cleaners. Usually it takes around 2.5 hours. So around $40/manhour. This is $30/robot hour. Slightly cheaper, but again it's not fully finished with the 1 hour they were available. For the company: It took 2 robots, 2 human operators, 1 staff member to dropoff/pickup, and 1 driver of the van to transport the robot. Definitely labor intensive. But they also retain the data they collect for training purposes. (7) My biggest takeaway as a user: interaction / hospitality mattered. One of the humanoids was more friendly (Sergey), he waved at us when he came in. It'd have been great if they both introduced themselves in the beginning. Beyond that, being able to do 2 hour slots (instead of one) would be great to finish the cleaning. Overall, I appreciated their work and the fact that they got the robots out in people's homes. Deployment is not easy and so much learning from it each time

> "The Secure Element chip is designed to detect physical tampering and will refuse to execute unsigned code or pass the genuine check if extra hardware attempts to intercept signals." Not only is this misleading consumers into believing the Secure Element can detect "extra hardware," but Ledger's own documentation explicitly states: > "Note that a Genuine check cannot detect unauthorized physical modifications to the hardware, such as spying implants, if the original Secure Element remains intact." @Ledger_Support's statement directly contradict Ledger's own technical documentation, and less than two months later, we're seeing reports of over $90 million stolen through precisely the kind of "extra hardware" intercepting signals that Ledger assured customers would be detected. And here's the worst part: Ledger had already published photographs of actual spying implants in its March 2026 Ledger Academy article, including sophisticated, mass-produced PCBs installed in multiple Ledger models. This wasn't some hypothetical attack they hadn't anticipated. They had already documented the threat, yet months later, their official Support account was assuring customers that Genuine Check would detect precisely this kind of hardware modification. How is this not misleading advertising? x.com/Ledger_Support…

Ledger Support @Ledger_Support

2 months ago

@HeyTaiZen Hi there, instead of relying on a sticker, Ledger uses Cryptographic Attestation. Every time you connect a device to Ledger Wallet, the software sends a cryptographic challenge directly to the Secure Element chip. If the hardware, chip, or internal firmware has been modified or

24 0 14 46K 13

Acoustic side channel?!

Zoë @zoecyber001

2 days ago

Everyone swears their phone is secretly recording their voice. You talk about buying hiking boots, and twenty minutes later Instagram serves you an ad for boots, even though your microphone permissions are turned completely off. Most of the time, ad networks don't need audio

41 157 1K 104K 1K

🤔 food for thought

Ari Paul @AriDavidPaul

2 days ago

Fairly soon we’ll likely be recording (almost) all conversations by default. Listening devices everywhere makes it happen passively, and then we’ll want the benefits of automated note taking and follow up by our AI assistants. How does this change life and society? I think it

46 8 128 40K 36

did you know someone can track your exact location just by sending you a PDF? no malware. no link to click. no app to install. just a regular looking PDF let me explain how this works because most people have no idea when someone sends you a PDF they can embed a tiny invisible image inside it that loads from their server the moment you open the PDF the image loads. their server records your IP address your IP address tells them your city, your internet provider, and your approximate location you didn’t click anything suspicious. you just opened a document this is how people get doxxed. someone sends a “job offer” or a “partnership proposal” as a PDF you open it thinking it’s legit but it might not be and this isn’t just PDFs. word documents and HTML emails do the same thing the most innocent looking file on your screen might be quietly telling someone where you live right now

‼️ UPDATE: A researcher says the Broadcom Wi-Fi router bug affects way more countries. The vulnerable code is also in ZTE's F680 v4, which ISPs from Spain to Vietnam, Egypt and Moldova offer. Three Movistar Spain routers appear affected. Broadcom still won't assign a CVE.

keyboard_arrow_left Previous keyboard_arrow_right Next

International Cyber Digest @IntCyberDigest

3 days ago

🚨 BREAKING: A Broadcom Wi-Fi bug lets anyone in range run code on routers without the password. Quentin Kaiser showed it on Proximus, Orange Belgium and VOO routers and estimates ~1.5M in Belgium are exposed, none patched. The bug dates to 2010. Broadcom won't assign a CVE.

keyboard_arrow_left Previous keyboard_arrow_right Next

68 529 3K 368K 1K

🚨 CIA Casually Admits Bitcoin Is Used As A Surveillance Tool CIA General Counsel: “Bitcoin isn’t truly anonymous… it’s a TOOL we use for intel gathering.” Now imagine a stablecoin economy…

Time is your greatest asset. "If you had a thousand dollars in your pocket and someone took out five and ran away, you wouldn’t then say, “Hey, hey, here’s the rest, right?” @RyanSerhant told me that's often what we do with our time. "You have a bad phone call for five minutes, someone pisses you off, and you give them the rest of that money. You're like, "You know what? Here you go. You should take it."

Shane Parrish @shaneparrish

a week ago

My guest today is @RyanSerhant, founder of SERHANT and star of Netflix’s Owning Manhattan. We talk about the psychology of getting noticed, earning trust, and becoming the person people choose. 00:00 Intro 02:19 Get Chosen Before You’re Established 10:03 Why Sellers Should

21 46 365 403K 706

So apparently it's common now to embed passive RF chips and clothing labels..... I had no idea till my 7 year old pointed it out in his shirt. Said " look at this invisible picture in my tag" Read up on it a little bit and although it's obviously not for nefarious purposes it could absolutely be abused. I was able to scan for it across my house... Spooky stuff. You guys seen this before? I had no idea it was a thing @Sad_HAMs_ @TheAngeryAmeric @PeteRobertucci @DolioJ @BigTMountainRed @TheLaserHam

keyboard_arrow_left Previous keyboard_arrow_right Next

OpenAI just told federal judges there may soon be no scientific way to tell a real video from a fake. So the guy your doorbell cam caught stealing your packages can stand up in court and say "that's AI." Only 15 of 931 federal judges surveyed have had someone try that. That number is going to go up fast. Two law professors, Bobby Chesney and Danielle Citron, saw this coming back in 2018 and called it the liar's dividend. Once everybody knows a perfect fake is possible, real footage stops ending the argument, and a jury that hears "could be AI" suddenly has something to chew on. And detection was always going to lose this fight. The first image generators that actually worked, GANs back in 2014, got good by training a forger against a detective over and over until the detective couldn't catch it anymore. So any detector somebody publishes basically becomes practice for the next model. Meta's product director told the same panel that regular people already can't tell recent AI video from the real thing, and making it costs next to nothing. Now think about what the judges have spent three years drafting. Under a draft rule called 901(c), if you want to call an exhibit fake, you first have to bring some evidence it was fabricated, and then the other side has to prove it's real. Both of those steps assume somebody can actually run a test, which is the exact thing OpenAI just told them might never exist. So courts are going to lean on people again, kind of like before anyone owned a camera. It'll come down to who shot the video and whether they're willing to swear to it. The tech industry's best idea is to stamp footage the second it's recorded so anything without a stamp looks suspicious. Every photo from a Pixel 10's camera app already gets a cryptographic signature baked in, and Google's camera lead said the whole point is getting people suspicious of pictures that show up without one. But OpenAI's lawyer also told the judges that every company runs its own system and nothing reads all of them. A stamp only helps if the camera that shot the video added one in the first place, and your doorbell probably didn't. Until that changes, every real video of somebody doing something awful comes with a free alibi.

Polymarket @Polymarket

5 days ago

JUST IN: OpenAI warns U.S. judiciary that there may be “no scientific way” to reliably detect AI-generated images & videos in the future.

274 659 5K 637K 739

Why tf am I starting to get ads on the most random, obviously ai-garbage product ads, read by a bot, and video generated by another bot Anyone seeing these too?

‼️ Japan is facing a storm of data breaches, and it keeps getting worse. Delivery giant Yamato Transport now says hackers took 5.31 million records, from names and addresses to what people bought. Cosmetics firms Fancl and Attenir say 834,947 more may have leaked. Ransomware gang Qilin claims it hit rail operator Keio and posted what it says are Keio files. In the past month, JR East, travel site Skyticket and software firm Sourcenext have each disclosed possible leaks of millions of records. Japan's National Police Agency says it is not yet clear whether the attacks are linked.

keyboard_arrow_left Previous keyboard_arrow_right Next

!

Lionel Mora @lionel_mora

4 days ago

In July we open-sourced a dynamic graph of everything a child learns in primary school. 1,590 concepts, 3,221 dependencies. Now we’re turning every concept into interactive experiments kids can play and learn from. First one, age 7: heat and water. Everything that used to be

208 832 8K 423K 10K

Would you like privacy with frontier AI models? Yes? Well, right now it’s impossible, because we don’t know how. But we can try to get closer, which is what @ethereumfndn and @OpenAnonProject have been trying to do by co-developing zkAPI! zkAPI is now listed on our Privacy page. It is a service that offers short-lived OpenRouter keys for ETH through an intermediary called Open Anonymity. You deposit into a shared pool, but your API key request to them contains only a proof that you have deposited, unlinking your specific deposit from the key. x.com/ethereumfndn/s…

Ethereum Foundation @ethereumfndn

3 weeks ago

Today the EF and OA announce zkAPI—a new means for private AI Picture this: deposit ether into a vault, sign a zk proof, get a fixed amount of private inference. Advances like these help us move past middlemen's requirements that we link our data and identities. Read more below

100 307 2K 388K 441

Unfortunately this is true. I've had an SSD die on me recently because 600gb of worktrees were being rapidly created and deleted by Codex. Huge fan of Codex - but SSDs are a bit pricey right now. I've submitted a few feedbacks about this but have not heard back. Sometimes my feedbacks don't seem to go through. Overall though you have to admit that normal users likely won't experience this - I think I only hit it because I am processing several Tb of data a day

Mark Kretschmann @mark_k

4 days ago

🚨 ChatGPT desktop users beware: A bug may be causing unnecessary wear on your SSD. @OpenAI I noticed constant SSD activity whenever the app was running. Tracing the writes revealed that the animated desktop Pet was triggering a rewrite of the entire ~1 MB app state file AND its

81 69 930 121K 375

🤖

Polymarket @Polymarket

4 days ago

JUST IN: Chinese robotics company unveils a massive 13-foot-tall, 2,400-pound humanoid robot capable of lifting more than 1,100 pounds.

keyboard_arrow_left Previous keyboard_arrow_right Next

1K 2K 14K 895K 2K

Hijack one employee's WhatsApp, clone the client group, ask for a transfer. HK$4M gone from a Hong Kong OTC desk. Your chat app is not a treasury approval system. Big transfers need a callback to a number you already had, and a second person who wasn't in the chat.

Coincamps @coincamps

4 days ago

Hong Kong police said a local crypto OTC company lost more than HK$4 million (about US$514,000) after scammers hijacked an employee’s WhatsApp account, blocked genuine clients and colleagues, and cloned the company’s work group. The fraudsters then impersonated clients and

0 2 11 15K 12