Qbix

635 posts

Qbix banner
Qbix

Qbix

@QbixPlatform

Empowering People. Uniting Communities. Expanding Local Businesses. Millions of people around the world use our apps.

New York City Katılım Aralık 2011
1.9K Takip Edilen254 Takipçiler
Sabitlenmiş Tweet
Qbix
Qbix@QbixPlatform·
Why do we trust giant, centralized platforms with our everyday personal conversations, data and identity? Because until recently, there hasn't been really good open-source software to power social communities, like Wordpress does for blogs. Now there is: m.youtube.com/watch?v=pZ1O_g…
English
0
3
5
0
Qbix retweetledi
Shaun Maguire
Shaun Maguire@shaunmmaguire·
We misunderstood the TikTok threat model The algorithm is *not* manipulated TikTok itself is banned in China They use a different app (Douyin) Douyin’s algorithm is: maximize harmony TikTok’s algorithm is: maximize division It’s not manipulated on specific topics (it was occasionally in the past) Why do we allow an app that’s banned in China? I’m open to Douyin in America but TikTok must be banned This is reverse Opium Wars; or digital covid
Drew Pavlou 🇦🇺🇺🇸🇺🇦🇹🇼@DrewPavlou

One of the San Diego shooters. Get the kids off TikTok, for the love of God

English
75
191
1.4K
127.8K
Qbix retweetledi
Ben Sigman
Ben Sigman@bensig·
I've run servers for 20 years. Never seen anything like this. rpow2 went viral. 36K+ verified email signups... then the AI-powered bots came. 200K+ unique IPs. 60K req/sec. GPU farms from China. Sybil farms registering thousands of fake accounts. Botnets spoofing Chrome across 2,100+ IPs. Bots adapting in real-time - block one vector, they switch within minutes. Chaos: 49-second response times. 86% of emails failing. @dotkrueger tagged me for help. ...then 18 hours later... Order restored: 16 anti-bot measures across 4 layers. Response times down to 20ms. Backend traffic cut 97%. Real users mine normally, bots get 403'd. This is modern warfare.
Ben Sigman tweet media
English
18
15
169
33.5K
Qbix retweetledi
Guri Singh
Guri Singh@heygurisingh·
the fakes won. AI-generated lies are now 56% more viral than human-made lies on X. They take 25% longer to get flagged. And once flagged, the AI itself has to fact-check them because humans can't tell anymore. A new paper just analyzed 150,000 misleading posts on X. The findings are worse than anyone wanted to admit. The study is called CONVEX. Here's what they found. Researchers pulled every Community Note flagged as misleading from 2021 to 2026. Sorted them into three buckets: miscaptioned (real photos with fake context), edited (photoshopped), and AI-generated. Then they tracked how each one spread. AI-generated content was disproportionately viral across both images and videos. 56% more likely to hit the top 1% than its share of the dataset would predict. Edited content was average. Miscaptioned was actually below average. But the way it spreads is what should scare you. Real misinformation gets argued about. People reply. People retweet to dunk. The conversation is loud. AI-generated fakes get likes. People scroll past, hit favorite, and move on. No debate. No friction. No noise. Just quiet acceptance. This is the opposite of what we were told would happen. The fakes aren't getting destroyed in the replies. They're winning the algorithm. Then the researchers tested 6 of the best AI image detectors on the market against real X posts from 2023, 2024, and 2025. Every single one is decaying. RINE dropped from 74.69% accuracy in early 2023 to 39.34% by late 2025. BFree dropped from 70.66% to 41.73%. Gemma 3 dropped from 82% to 62%. Even GPT-5 falls apart on the newest fakes. The fakes are evolving. The detectors aren't. The wildest part is buried in the data. AI-generated fakes take 25% longer than other misinformation to get reported. Real humans can't tell they're fake fast enough. And here's where it loops back on itself. The Community Notes that do eventually flag these AI fakes increasingly cite AI tools as their evidence. Grok appeared in 47% of them. ChatGPT in 16%. Gemini in 14%. We've reached the point where the only thing that can reliably fact-check an AI image is another AI. This was published at CVPR 2026. One of the most prestigious computer vision conferences in the world. The takeaway isn't that AI fakes exist. You knew that. The takeaway is that the entire defense layer we built to catch them, specialized detectors, vision-language models, community moderation, is losing ground every quarter. Every month a new generative model ships, the line between real and fake gets thinner. And the tools we trust to draw that line get duller. You won't be able to tell what's real soon. Neither will the AI checking it for you.
Guri Singh tweet media
English
6
15
32
2.8K
Qbix retweetledi
Garry Kasparov
Garry Kasparov@Kasparov63·
MAGA & conspiracy crap. The monster is a reflection of its creator, or, in this case, its buyer and owner. I hope no one was so foolish as to think Musk bought Twitter for any reason other than to make it over in his image and pursue his political and commercial goals.
Nate Silver@NateSilver538

These are the Twitter/X accounts with the most engagement so far in 2026. I suppose I had some intuition for how bad it was, but jeez, this is what you get when the ecosystem is broken.

English
312
2.1K
7.8K
350.3K
Qbix
Qbix@QbixPlatform·
Happy April 1st. This is not a joke! (Well, it is, but the underlying events all really happened) qbix.com/blog/2026/04/0…
English
0
0
0
6
Qbix retweetledi
Cheng Lou
Cheng Lou@_chenglou·
My dear front-end developers (and anyone who’s interested in the future of interfaces): I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept): Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow
English
1.3K
8.3K
65.4K
23.9M
Qbix retweetledi
SciTech Era
SciTech Era@SciTechera·
🚨 BREAKING: Intel unveils “Heracles” chip that makes encrypted computing thousands of times faster. Heracles is a new processor designed to dramatically accelerate Fully Homomorphic Encryption (FHE), a technology that allows computers to perform calculations directly on encrypted data. It's presented at the IEEE International Solid-State Circuits Conference, the chip can verify encrypted queries in about 14 microseconds, compared with roughly 15 milliseconds on a standard Intel Xeon server CPU. That’s a speed improvement of up to 5,000×. Heracles includes 64 compute cores and high-bandwidth memory, specifically optimized for encrypted workloads used in cloud computing and AI. Why this matters? Fully Homomorphic Encryption has long been considered one of the “holy grails” of cybersecurity. Normally, data must be decrypted before it can be processed, which creates a moment where sensitive information can potentially be exposed. FHE changes that 👀! With FHE, data can remain encrypted the entire time while still being analyzed or processed. If hardware like Heracles makes this practical at scale, it could enable: > Cloud services that analyze data without ever seeing it > AI models trained on encrypted medical or financial data > Secure collaboration between organizations without sharing raw data In simple terms, Computers could use your data without ever actually seeing it.
SciTech Era tweet media
English
15
93
299
15.8K
Qbix retweetledi
Car
Car@CarOnPolymarket·
How can a random Polymarket account correctly predict: - US striking Venezuela - US kidnapping Maduro - US forces in Venezuela - US striking Iran - Israel striking Iran - US striking Iraq - Israel striking Khamenei - US anti cartel operations And his next bet is public for everyone: US forces entering Iran in the near future. Insane!
PredictFolio@PredictFolio

BREAKING: A suspected military insider won $90K correctly predicting 9 separate military events! This guy is now betting big on US forces entering Iran! He will win nearly $100K if he is correct.

English
304
2.8K
22.2K
4.3M
Qbix retweetledi
HustleBitch
HustleBitch@HustleBitch_·
🚨 WHAT THE HELL ARE THEY PUSHING THROUGH WHILE EVERYONE WATCHES THE WAR? This woman just dropped a theory that’s sending people down a rabbit hole. She claims that while the world is glued to the war, something enormous is quietly moving in the background: The largest media merger in American history. And when it closes… one family will control: CNN. CBS. HBO. Warner Bros. Paramount. According to her breakdown, the entire system behind it could run on Oracle infrastructure. The same Oracle that just landed federal contracts tied to Medicare and Medicaid data for 150+ million Americans… and AI operations for the U.S. Air Force. Then she connects the dots in the video: Government contracts send billions into Oracle. That revenue supports Oracle’s stock. That stock is reportedly being used as collateral tied to the merger. And Oracle ends up as the infrastructure underneath the new media empire. Infrastructure. Collateral. Beneficiary. Every layer of the same machine. Then she points out the timing. The war timeline that just got mentioned by Trump About four weeks. The shareholder vote on this merger? March 20. Coincidence? Her argument: While everyone is staring at missiles and war coverage... something massive could be sliding through the system behind the curtain. Is the war the only story right now… or is something huge being pushed through while nobody’s looking?
English
727
11K
25.3K
1.1M
Qbix
Qbix@QbixPlatform·
Totally not surprised somehow
Shanaka Anslem Perera ⚡@shanaka86

JUST IN: Meta sold 7 million Ray-Ban smart glasses in 2025 alone. Workers in Kenya are watching the footage. Not metadata. Not anonymized clips. The actual videos. People undressing. People in bathrooms. People having sex. Bank cards. Medical documents. The blurring is supposed to protect privacy. It fails constantly. The contractors see everything. Here is the part that should stop you cold: You did not buy the glasses. You did not agree to the terms of service. You did not consent to anything. But if someone wearing Meta glasses walks into your bedroom, your bathroom, your doctor's office, your home, a contractor on the other side of the world may be watching you right now. The person wearing the glasses consented. Everyone else in the room did not. Meta's defense is that this is all disclosed in the privacy policy. They are technically correct. Buried in language so dense that 99% of users never read it. And even if they did, it would not matter, because the terms govern the wearer's data. Not yours. You are not a party to the contract. You are the product being annotated. Millions of AI-enabled cameras walking around in public. Recording constantly. Uploading to servers. Reviewed by humans earning a few dollars an hour to label your most intimate moments so the algorithm gets smarter. This is not a bug. This is the business model. The EU is already asking questions. MEPs submitted formal inquiries to the Commission this week demanding answers on GDPR compliance. The problem is obvious: European data protection law requires consent from data subjects. Bystanders are data subjects. Bystanders never consented. The entire architecture violates the regulation by design. Meta's response has been silence and a reference to terms of service that do not apply to the people actually being filmed. Google Glass died because people called the wearers "Glassholes" and banned them from bars. Meta solved the social problem by making the glasses look normal. They did not solve the privacy problem. They hid it. Seven million units sold in 2025. The installed base is accelerating. Every unit is a potential surveillance node operated by someone who may not understand what they are feeding into the system and reviewed by contractors who see everything the algorithm cannot process. The question is not whether this becomes a scandal. The question is whether the scandal arrives before or after the glasses are on 50 million faces. Watch the EU. If Brussels moves on GDPR enforcement, Meta faces a choice: disable human review in Europe and cripple the AI training pipeline, or accept fines that could reach billions. Neither outcome is priced into the stock. The glasses are selling faster than ever. The contractors keep watching. And somewhere right now, someone you have never met is looking at footage of you that you never knew existed.

English
0
0
0
7
Qbix retweetledi
AnthroVet50
AnthroVet50@spooked75·
Nature just published peer-reviewed evidence that X's algorithm shifts users toward conservative political opinions. Not engagement—actual beliefs. But we've had the receipts since 2023. When Twitter open-sourced its algorithm, researchers found author_is_elon as a literal boost factor, users labeled "Republican" or "Democrat," and hand-coded heuristics amplifying specific accounts. Musk called it "weird" his name was in the code. The Nature study ran a 7-week randomized experiment—independent, no X cooperation. Result: algorithmic feed users became more conservative on policy issues. This was before Musk endorsed Trump. Now in January 2026, xAI releases their new "clean" Rust-based algorithm. The pitch: "We eliminated every single hand-engineered feature." Except the new system uses a Grok transformer trained on engagement data. What shaped that engagement? Years of the biased algorithm. They didn't remove the bias. They laundered it through ML. Old system: "Boost Republicans" (findable in code) New system: "Predict engagement" (trained on biased data) Same output. Plausible deniability. The algorithm isn't neutral. It never was. And the new version is designed so you can't prove it anymore. nature.com/articles/s4158…
English
21
175
366
16.9K
Qbix retweetledi
rat king 🐀
rat king 🐀@MikeIsaac·
amazon's internal A.I. coding assistant decided the engineers' existing code was inadequate so the bot deleted it to start from scratch that resulted in taking down a part of AWS for 13 hours and was not the first time it had happened incredible ft.com/content/00c282…
rat king 🐀 tweet media
English
384
3.2K
23K
6.5M
Qbix retweetledi
sui ☄️
sui ☄️@birdabo·
discord’s age verification vendor Persona just got hacked. openAI + US government + persona have been running identity surveillance since nov 2023 anyone who verified age on discord is now traceable. nothing is private anymore. vmfunc.re/blog/persona
sui ☄️ tweet mediasui ☄️ tweet media
English
494
8K
51.5K
4M
Qbix retweetledi
The Rage
The Rage@theragetech·
🚨NEW: HACKERS EXPOSE AGE-VERIFICATION SOFTWARE POWERING SURVEILLANCE WEB Three hacktivists tried to find a workaround to Discord’s age-verification software. Instead, they found its frontend exposed to the open internet, bundling financial reporting with facial recognition tech. For the first time, they lay bare what is really behind a software serving age-verification laws – and why such technology is far from harmless. We spoke with the researchers that exposed the software, who hope that their findings will serve as a wakeup call.
The Rage tweet media
English
106
2.7K
6.8K
414.5K
Qbix
Qbix@QbixPlatform·
We need to decentralize. Soon.
English
0
1
0
14