Welcome to another round of tech news! This week, we're diving into the resurrection of a once-popular social media site, the EU's big bet on RISC-V, fresh zero days for VMware, and Broadcom's bold money-making moves.

🎯 **Reddit's Co-Founder Wants to Fix Social Media... With More Social Media?** 
Alexis Ohanian, one of Reddit’s original creators (the one who *doesn’t* suck), is teaming up with the founder of Digg to bring it back from the dead. Digg was a big deal in the mid-2000s before it collapsed under bad management, but now it’s making a comeback with AI in tow. Will it be the next big thing or another failed revival? Only time will tell. Want to get in early? They’re taking email sign-ups now. 
🔗https://www.nytimes.com/2025/03/05/technology/digg-alexis-ohanian-kevin-rose.html

💾 **The EU Goes All-In on RISC-V for Supercomputing** 
Europe is pushing hard for digital independence with a $260 million investment in RISC-V-based supercomputing chips. The project, named DAR (Digital Autonomy with RISC-V for Europe), aims to develop three chiplets for high-performance computing. It’s a bold move to move away from x86 and ARM dominance, but can they deliver on their aggressive timeline? 
🔗 https://www.theregister.com/2025/03/07/dare_europe_risc_v_project/

⚠️ **Three New VMware Zero Days—Because One Isn't Enough!** 
VMware ESX is under attack again, with three fresh zero-day vulnerabilities actively exploited in the wild. The worst of the bunch (CVE-2025-22224) lets attackers execute code on an ESXi host. Microsoft actually reported these to Broadcom, which is a fun little twist. If you haven't patched your VMware hosts yet, now would be a *really* good time. 
🔗 https://support.broadcom.com/web/ecx/support-content-notification/-/external/content/SecurityAdvisories/0/25390

💰 **Broadcom's VMware Cash Grab is Working... For Now** 
Broadcom is cashing in on its $69 billion VMware acquisition by slashing products, jacking up prices, and locking in big customers. The strategy seems to be working—at least in the short term—as revenue soars. But with frustrated customers looking for alternatives, could VMware's long-term future be in jeopardy? Competitors like Nutanix are already gaining ground. 
🔗 https://investors.broadcom.com/news-releases/news-release-details/broadcom-inc-announces-first-quarter-fiscal-year-2025-financial

That’s it for this week! Like, subscribe, and maybe, just maybe, we'll see you again next time. 🚀

The Internet: it was never supposed to work this well, and yet, here we are. This week, we’re diving into how we went from isolated, room-sized computers to a global, decentralized network that somehow (mostly) functions. We break down the early days of networking, when computers had to physically dial each other up, and how we eventually arrived at the distributed, self-healing, packet-switching magic that powers everything today. 

Along the way, we cover the different network models—centralized, decentralized, and fully distributed—and why only one of them could survive a nuclear attack (always an important design consideration). We also touch on ARPANET, the first real Internet, and how a bunch of researchers, military contractors, and a few weirdly named computers set the stage for what we have now. Spoiler: it involves a lot of improvised problem-solving and some very lonely PDP-10s. 

So if you’ve ever wondered how routing actually works, why the Internet doesn’t just collapse under its own weight, or what an “IMP” is, we’ve got you covered. Just don’t blame us if you leave this episode with a sudden urge to dig through RFCs. 

🚀 LINKS 
🔗 RFC 1206: https://www.rfc-editor.org/rfc/rfc1206
🔗BGP episode: https://pod.chaoslever.com/break-the-glass-and-walk-away-a-very-brief-overview-of-bgp/
🔗Supercomputing episode: https://pod.chaoslever.com/from-cray-1-to-el-capitan-the-evolution-of-supercomputers-chaos-lever/
🔗Classical computing: https://pod.chaoslever.com/turing-to-transistors-classical-computing-101/
🔗Brief history of the internet: https://www.internetsociety.org/internet/history-internet/brief-history-internet/

📢 Welcome to another episode of Tech News of the Week, where Chris and I break down the biggest stories in tech—sometimes with insight, sometimes with sarcasm, but always with desks. Maybe? This week, we're talking about IBM finally sealing the deal with HashiCorp, Microsoft's continued Notepad nonsense, OpenAI's latest attempt to justify its existence, and the biggest crypto heist of all time. Buckle up!  

🟡 **HashiCorp Joins IBM**  
After 10 long months of regulatory limbo, IBM’s $6.4 billion acquisition of HashiCorp is finally official. Now that the deal has closed, IBM is set to integrate Terraform with Ansible, strengthen HashiCorp Vault with OpenShift, and generally try to make their aging product lineup more cloud-native. As a HashiCorp fan, I wasn’t exactly thrilled, but hey—at least IBM is dumping money into R&D. Cautiously optimistic? Maybe. Full breakdown here: https://www.hashicorp.com/en/blog/hashicorp-officially-joins-the-ibm-family and convo with Armon Dadgar here: https://www.youtube.com/live/p9VZMDRJ6m0

🟡 **Microsoft Notepad Gets AI Because… Reasons?**  
Notepad was perfect. It was simple. It was beautiful. And now, Microsoft is stuffing it with AI features nobody asked for—just like they’re doing with Paint. Oh, and they killed off WordPad, pushing people toward paid Microsoft Word instead. At least the AI-infested versions of Notepad and Paint are optional (for now), but this is a slippery slope, folks. More details: https://www.pcworld.com/article/2614943/microsoft-is-paywalling-these-features-in-notepad-and-paint.html

🟡 **GPT-4.5: Bigger, Better, and Full of Lies**  
OpenAI just dropped GPT-4.5, and it’s… well, it’s a thing. It’s bigger, more power-hungry, and claims to be "friendlier" and "more truthful"—which means it only lies to you **37%** of the time instead of **59%**. Progress? Maybe. Desperation? Definitely. Sam Altman’s money-burning machine continues, and SoftBank is still writing checks, so here we are. The full scoop: https://www.technologyreview.com/2025/02/27/1112619/openai-just-released-gpt-4-5-and-says-it-is-its-biggest-and-best-chat-model-yet/

🟡 **Bybit Suffers the Largest Crypto Heist Ever**  
Dubai-based crypto exchange Bybit just lost **$1.5 billion** to North Korea’s Lazarus Group, making it the biggest crypto theft in history. Hackers exploited Bybit’s Ethereum wallet system, faked transactions, and walked away with an absolute fortune. The good news? Bybit says it’ll reimburse customers. The bad news? This whole mess proves, once again, that "faster" and "secure" are rarely friends. More on this wild story: https://www.reuters.com/technology/cybersecurity/cryptos-biggest-hacks-heists-after-15-billion-theft-bybit-2025-02-24/ 

Now, go away. Bye. 🚀

Fifth-generation programming languages? Generations don’t even matter anymore. We’re basically at iPhone generation 16-and-a-half, and at some point, people are just making stuff up. Speaking of making things up, today’s episode of Chaos Lever is all about NAT (Network Address Translation), a necessary evil—or maybe just evil—that helped shape the internet as we know it. To break it all down, we’re joined by the legendary Ivan Pepelnjak, a CCIE Emeritus, BGP wizard, and all-around networking guru. He’s here to walk us through the chaotic history of NAT, why it happened, and why, despite all efforts, it’s never really going away. 

We dive into the days when IP was just one of many competing protocols, when grabbing an IP block was as easy as sending an email, and when the first NAT implementations were only meant to be a temporary fix.

Spoiler alert: that temporary fix became the foundation of modern networking. Ivan shares his firsthand experience from decades in the field, discussing why IPv6 adoption has been slow, how carrier-grade NAT is making things even messier, and why the dream of a fully end-to-end connected internet never really stood a chance. Plus, we touch on some truly wild networking trivia—like how stock exchanges measure fiber cable lengths to the nanosecond. 

If you’ve ever wondered why your home Wi-Fi setup still relies on NAT, why cloud providers and ISPs are desperate to push IPv6, or what networking challenges we’ll still be complaining about in another 20 years, this episode is for you. Stick around for some networking history, a bit of good-natured industry snark, and of course, a little chaos. And if you make it all the way to the end? Congrats, you’ve earned the right to set up your own double-NAT configuration at home—for "fun". 

---

LINKS 

🔗 Chaos Lever Website: https://chaoslever.com
🔗 Chaos Lever LinkedIn: https://www.linkedin.com/company/chaoslever
🔗 Ivan Pepelnjak’s Blog & Networking Resources: https://blog.ipspace.net
🔗 NetLab Open Source Project: https://netlab.tools

Welcome back to Tech News of the Week, where Chris and I break down the biggest and weirdest stories in tech. We're a week behind because Chris decided to lose power—how selfish! But we’re back, and we’ve got four spicy news stories to dive into. Let’s go!  

🧠 **Meta Wants ARM-Made Chips**  
ARM might start making its own chips, and Meta is reportedly first in line to buy them. This is a big shift for ARM, which has historically just designed and licensed chip architectures rather than manufacturing its own. If true, this could shake up the chip industry and make ARM a competitor to companies it currently licenses to. The first chips are rumored to launch this summer, so we won’t have to wait long to see what happens. Will this push companies toward RISC-V? Fingers crossed!  
🔗 https://techcrunch.com/2025/02/13/arm-is-launching-its-own-chip-this-year-with-meta-as-a-customer

📖 **Facebook’s AI Reads Minds (Kind Of)**  
FAIR (Facebook’s Fundamental AI Research unit) teamed up with scientists in Spain to create a machine that can read your mind—well, sort of. By analyzing brain activity using M-E-G and E-E-G, they achieved an 80% accuracy rate in predicting what subjects were typing or saying. Right now, the tech is clunky and requires a controlled environment, but smaller, scarier versions are likely on the way. What could go wrong?  
🔗https://www.techspot.com/news/106721-meta-researchers-unveil-ai-models-convert-brain-activity.html

💰 **SolarWinds Goes Private for $4.4B**  
Remember SolarWinds? The company that got hit with a massive supply chain attack in 2020? Well, private equity firms have decided it’s still worth squeezing for cash. Silver Lake and Tomo Bravo bought up a majority stake, and now TurnRiver is taking the whole thing private for $4.4 billion. Expect less innovation, more “cost optimization,” and an eternal cycle of rent-seeking. Somewhere in Middle-earth, Sauron is proud.  
🔗 https://www.darkreading.com/cybersecurity-operations/solarwinds-private-billions

🖥️ **AI is Just Fancy Copy-Paste, Confirms Study**  
A new report shows that AI-assisted coding is leading to lower code quality. GitClear analyzed 200 million lines of code and found that, surprise surprise, AI-generated code is often just old code copied and pasted with minimal thought. Google’s own research backs this up, showing rising defect rates in published code. Microsoft even warns that overreliance on AI is killing critical thinking skills. So, uh… we’re definitely headed toward a bright, bug-free future, right?  
🔗 https://www.gitclear.com/ai_assistant_code_quality_2025_research

What happens when you mix cryptography, heists, social engineering, and a dash of early '90s tech paranoia? You get **Sneakers**—a movie that might be more relevant today than it was in 1992. In this episode, we break down the film’s tech, its realism (or lack thereof), and the undeniable truth that **people are always the weakest security link**. Because at the end of the day, it’s not about the tech—it’s about **People, Process, Technology**, in that order.  

Along the way, we take some delightful detours, including a discussion of **Tim Curry chewing scenery in *Legend***, whether a Braille Playboy was a real thing (spoiler: it was), and why the best security measures can be defeated with a birthday cake and some good old-fashioned social engineering.

So, grab your popcorn and your best anagram-solving skills, and let’s dive into the world of Sneakers—a movie that understood the power of ones and zeros long before our social feeds did. Oh, and if you’re planning to crawl around in a drop ceiling like Robert Redford, please don’t.

Welcome back to another jam-packed episode of Tech News of the Week! Chris and I are diving into four big stories that caught our attention this week. From sketchy ISP routers to OpenAI’s latest security fail, let’s break it all down.  

🔹 **Stop Using Your ISP Router—Seriously**  
If you're still using the router your internet provider gave you, it's time for an upgrade. Not only are ISP-provided routers outdated and full of security holes, but they might also be spying on you—and, in some cases, even harboring actual bugs (the creepy-crawly kind). A new website, [RouterSecurity.org], lays out just how bad these devices can be. Investing in a good third-party router is a small price to pay for better security and performance. Also, if you haven’t looked into mesh routing yet, you’re missing out!  LINK: https://routersecurity.org/ISProuters.php

🔹 **AWS Goes on an AI Spending Spree**  
Amazon reported solid Q4 earnings, but apparently, a 19% growth in AWS wasn't enough to impress investors. So, what's Amazon’s solution? Throw more money at AI! They’re planning to invest a whopping $100 billion in AI hardware this year, with much of that going toward NVIDIA-powered chips. The hope is that supply chain issues will ease up, allowing AWS to scale its AI efforts even further. But will all this spending pay off in the long run? We’ll see.  LINK: https://ir.aboutamazon.com/news-release/news-release-details/2025/Amazon.com-Announces-Fourth-Quarter-Results/default.aspx

🔹 **Phishing Tests Are Getting… Meaner?**  
We all know about phishing tests—those fake scam emails companies send to see if employees fall for them. But lately, these tests have been pushing the limits, with some using emotionally charged messages like fake Ebola outbreaks or rescinded bonuses. The Wall Street Journal reports that while these tactics may be effective, they’re also making employees furious. One particularly controversial example? A phishing email promising free Eagles tickets to people in Philadelphia. Ouch.  LINK: https://www.wsj.com/tech/cybersecurity/phishing-tests-the-bane-of-work-life-are-getting-meaner-76f30173

🔹 **OpenAI’s New Model Helps… Write Malware?**  
Well, that didn’t take long. OpenAI's new "secure" GPT-4 variant, O3 Mini, was supposed to be better at filtering out harmful requests. But within days, a security researcher tricked it into generating code to exploit Windows security processes. OpenAI insists the exploit wasn’t serious, but the fact remains—these models still aren’t as locked down as they claim. Maybe a little more internal testing before release wouldn’t hurt?  LINK: https://www.darkreading.com/application-security/researcher-jailbreaks-openai-o3-mini

That’s it for this week! Drop a comment, let us know your thoughts, and we’ll catch you in the next one. 🚀

Can you feel the enthusiasm radiating from this episode? No? Well, Chris is already singing *The Lion King*, so we’re off to a strong start. Today, we’re diving headfirst into the world of AI with DeepSeek, the latest generative AI model out of China that’s supposedly shaking up Silicon Valley. Is it really the *Sputnik moment* some are claiming, or is it just another overhyped step forward? Spoiler: It’s not Sputnik.  

We break down DeepSeek’s origins, its connection to a high-frequency trading hedge fund, and why its *free* and *open-source* nature might not be as open as it seems. Oh, and censorship—lots of censorship. But don’t worry, you can always trick it by asking questions in Pig Latin. Meanwhile, Chris did some highly scientific testing (read: he asked it a Bible question), and we debate whether reasoning transparency is a *game-changer* or just a fancy parlor trick.  

Is DeepSeek a technical marvel? Yeah, kinda. Is it revolutionary? Not really. Is it 96% cheaper than OpenAI’s API? Absolutely. And *that* is what has Silicon Valley panicking. We also talk about the Wright brothers, the Cold War, and how local AI models might just burn a hole in your couch. Good times all around.   

---

🔗 **LINKS**  
- https://www.deepseek.com
- https://arxiv.org/abs/2408.14158
- https://www.promptfoo.dev/blog/deepseek-censorship/
- https://erichartford.com/uncensored-models
- https://www.wired.com/story/deepseeks-ai-jailbreak-prompt-injection-attacks/
- https://proton.me/blog/deepseek

Welcome back to another episode of *Tech News of the Week!* This week, we dive into some fascinating developments in quantum computing, corporate drama at Meta, a potential shake-up in the networking industry, and security vulnerabilities in Apple Silicon chips. Buckle up—it's going to be a wild ride.  

🔬 **Photonics for Quantum Computing**  
Quantum computers are finicky beasts, usually requiring extreme cold to keep their delicate qubits from falling apart. But what if we could use *light* instead? Canadian startup Xanadu is tackling this challenge with its photonic quantum computer, *Aurora*. Their modular system could make quantum computing more scalable and affordable—if they can solve the usual qubit problems. Does this deserve a full episode? Chris, get on it. 😆 https://www.technologyreview.com/2025/01/30/1110672/this-quantum-computer-built-on-server-racks-paves-the-way-to-bigger-machines/

📢 **Zuckerberg Complains About Leaks… in a Leaked Meeting**  
Meta’s CEO, Mark Zuckerberg, recently expressed frustration that everything he says leaks… in yet another leaked meeting. The irony is thick. Meta’s once-open town halls have turned into tightly controlled sessions, yet the leaks keep coming. Maybe, just maybe, the problem isn’t the employees but the guy in charge? One commenter summed it up best: “pre-divorced sh*tweasel.” https://www.404media.co/zuckerberg-says-everything-i-say-leaks-in-leaked-meeting-audio/

🛑 **DOJ Blocks HPE-Juniper Merger**  
HPE’s $14B acquisition of Juniper Networks has hit a major roadblock. While Europe and the UK gave it the green light, the U.S. Department of Justice stepped in, citing concerns over market consolidation. The WLAN space is already dominated by a few major players, and the DOJ isn’t keen on reducing competition further. Meanwhile, HPE and Juniper insist this merger is "pro-customer"... for reasons. https://www.theregister.com/2025/01/30/hpes_acquisition_juniper/

🔓 **Apple Silicon Chips Have Security Flaws**  
Apple’s M-series chips have been crushing the competition, but they’re not invincible. Researchers found vulnerabilities—SLAP and Flop—that allow sneaky memory access. While these attacks are difficult to pull off, the fact that they’re possible at all is concerning. Apple hasn’t responded yet, but maybe, just maybe, CPUs don’t *need* speculative execution anymore? Just a thought. 🤔 https://predictors.fail

📩 Got thoughts? Want to share expert insight? Hit us up at ChaosLever.com (but no collect calls, please). See you next time! 👋

Welcome to another episode of Chaos Lever, where we take a deep dive into the tech abyss and come out slightly more confused than when we started.

This week, we're talking about the OG of directory services: X.500. Before Active Directory, before LDAP, there was this ambitious yet painfully cumbersome attempt to organize networked systems into a structured directory. Was it elegant? No. Was it practical? Also no. But did it lay the groundwork for everything we use today? Absolutely. Along the way, we uncover just how much of modern networking was cobbled together by people who were just making it up as they went.  

If you've ever wondered why directories matter, or you just enjoy hearing us ramble about obscure tech history, this episode is for you. And don’t worry—this is only part one. We still have Netscape, Microsoft, and a whole mess of bad decisions to cover. So buckle up, enjoy the ride, and remember: if you’re not questioning your life choices by the end of this episode, we haven’t done our job.  

---

🔗 **LINKS**  

- https://www.identityfusion.com/blog/the-most-complete-history-of-directory-services-you-will-ever-find 
- https://www.nexor.com/blog/prehistory-of-ldap 
- https://sec.cs.kent.ac.uk/x500book/
- https://nvlpubs.nist.gov/nistpubs/Legacy/IR/nistir5819.pdf
- https://www.itu.int/rec/T-REC-X.500