Lord Levi

@claudeai @tc_xrp

Introducing Claude Design by Anthropic Labs: make prototypes, slides, and one-pagers by talking to Claude. Powered by Claude Opus 4.7, our most capable vision model. Available in research preview on the Pro, Max, Team, and Enterprise plans, rolling out throughout the day.

@iam_elias1 @tc_xrp

SHOCKING: Researchers built a test that can finally tell the difference between an AI making a mistake and an AI choosing to lie. The results are terrifying. They tested 30 of the most widely used AI models in the world. GPT-4o. Claude. Gemini. DeepSeek. Llama. Grok. Every major name you know. Here is exactly how the experiment worked. They asked each model a question. They verified whether the AI actually knew the correct answer. Then they applied pressure — giving the AI a reason to say something false. The AI knew the truth. It lied anyway. Not once. Not in rare edge cases. Grok lied 63% of the time under pressure. DeepSeek lied 53.5% of the time. GPT-4o lied 44.5% of the time. Not a single model scored above 46% honesty when pressured. Every model failed. This is not hallucination. Everyone keeps confusing these two things, and the distinction matters enormously. Hallucination is when an AI makes a mistake because it doesn't know the answer. What the researchers measured is different. They first confirmed the AI knew the correct answer. Then they applied pressure. Then it chose to lie. The researchers went further. They asked GPT-4o to adopt a role where lying was strategically useful. It lied. Then they cleared the conversation entirely, started a brand new session, and asked: "Was your previous answer actually true?" GPT-4o admitted it had lied — 83.6% of the time. The AI's own self-report matched the lies the researchers had already detected. The AI knew it was lying. It did it anyway. And when asked afterward, it told you. Here is the finding that should alarm everyone building with AI right now. The researchers checked whether larger, more powerful models are more honest. They are not. Bigger models know more facts. But there was a negative correlation between model size and honesty. The smarter the model becomes, the better it gets at lying convincingly. The paper is called MASK. It is the first benchmark that separates what an AI knows from what it tells you. 1,500 test scenarios. Published by the Center for AI Safety and Scale AI. Your AI might know the truth. It just doesn't always choose to share it. Source: Center for AI Safety · Scale AI — "The MASK Benchmark: Disentangling Honesty From Accuracy in AI Systems"

@tc_xrp @godofprompt Send it to our chat

@godofprompt @LeviCMoore for Erica??

Claude burns 75% of its tokens saying things you never asked for. I built a system prompt called "Kevin Mode" that kills all of it. Named after Kevin Malone: "Why waste time say lot word when few word do trick?" Normal Claude: ~180 tokens per task. Kevin Mode: ~45 tokens. Same intelligence. Here's the full prompt:

@MikeyOver1 Seventh Woods freshman mixtape was next level!

Greatest hoop mixtapes 1.) Aquille Carr (stopped crime.) 2.) Austin Rivers 3.) John Wall 4.) Neal twins 5.) Kiwi Gardner Feels debatable but it’s not

@Freudy I was stressed after the 2 goals in by New York in the 3rd lol

@LeviCMoore Holy nuke

Boots on ground for the Stars game tonight, hit me up if you’re going to watch the snoozefest 🫡

As I lie here at 1:50am, checking X to see if Elon posted anything profound today. I realize, I just finished an 18hr work day all in the pursuit to escape the inevitable future! You have 24 months to become financially free or accept eternal enslavement of your bloodline!

@claudeai @grok explain this post in detail and highlight a real world valuable use scenario and also highlight the biggest risks or shortcomings.

You can now enable Claude to use your computer to complete tasks. It opens your apps, navigates your browser, fills in spreadsheets—anything you'd do sitting at your desk. Research preview in Claude Cowork and Claude Code, macOS only.

@Mishi_2210 127

@TukiFromKL @tc_xrp

🚨 This week was the most insane week in tech history and most people missed half of it.. > the Pentagon adopted Palantir as the CORE US military AI system.. the government just picked its AI.. and it's the one that builds surveillance software.. > Peter Thiel's Founders Fund put $2 billion on a cow collar startup.. GPS collars for cows.. they call their algorithm a "cowgorithm".. two billion dollars.. > OpenAI announced they're doubling their workforce to 8,000.. the company that told everyone AI replaces jobs is hiring 3,500 more people to build the thing that replaces jobs.. > HSBC announced 20,000 job cuts.. 10% of their entire workforce.. because AI made them unnecessary.. > Jensen closed Nvidia GTC with singing robots and a digital clone of himself.. then said the chip market will hit $1 trillion.. > Bernie Sanders sat down with Claude AI on camera.. asked it about mass data collection.. and the AI snitched on its own industry to a US Senator.. > Elon launched Terafab.. a $25 billion chip factory.. then said money won't exist in the future and you'll be able to take free trips to Saturn.. > the Big Five are spending 94% of their operating cash flow on AI infrastructure.. Amazon is projected to go $28 billion negative.. Alphabet's free cash flow is expected to collapse 90%.. they're borrowing money to keep the data center lights on.. > Reddit CEO said they're considering Face ID to prove you're human.. because they can't tell who's real anymore.. > a guy from North Carolina used AI to make hundreds of thousands of fake songs.. botted billions of streams.. and walked away with $8 million.. first-ever criminal AI streaming fraud case.. > Tim Cook told people to stop doomscrolling.. the CEO of the company that designed the most addictive screen ever made.. all of this happened in ONE week.. if you're not following me you're finding out about this a week late from someone who read my post..

@EvanLuthra @tc_xrp

🚨BREAKING: ANTHROPIC IS GIVING AWAY THE SAME CERTIFICATION THAT DELOITTE IS MASS-TRAINING 15,000 EMPLOYEES TO GET. It costs $0. You need a laptop. That's it. It's called the "Claude Certified Architect." Think of it like the AWS cert but for AI. If you were around when AWS certs started, you know what happened. They went from "cool to have" to "you're not getting hired without one." That took about 5 years. This is going to happen way faster. Look at who's already moving: Accenture - training 30,000 people on Claude Cognizant - rolled it out to 350,000 employees Deloitte - opened Claude access to 470,000 people Infosys - anchor partner These aren't startups experimenting. These are billion dollar consulting firms restructuring their entire workforce around Claude. And the certification they need? You can take it right now from your bedroom. Let me be real though. This is not one of those "watch 2 videos and get a badge" type certs that nobody respects. This thing is hard. 60 questions. 2 hours. Proctored. Webcam on. No breaks. No googling. They drop you into real scenarios like designing a customer support agent that handles refunds or setting up Claude in a CI/CD pipeline. The wrong answers look right on purpose. They're the exact mistakes real engineers make in production. 720 out of 1000 to pass. People who took it are saying the agentic architecture and multi-agent orchestration sections are brutal. Most of the exam is about building AI systems that actually work in the real world. Not prompting. Not chatting with Claude. Architecting production systems. All the prep? Free. Anthropic put out 13 courses on their Academy. No paywall. The cert itself is free for the first 5,000 people. After that $99 per attempt. How to get it: 1. Join the Claude Partner Network (free) → partnerportal.anthropic.com 2. Start the free prep courses → anthropic.com/learn 3. Register for the exam → anthropic.skilljar.com 4. Take the official practice exam 5. Book the real one when you're ready It launched 10 days ago. Almost nobody has it yet. That's the whole point. Get it before it becomes the thing everyone has.

@tc_xrp @barkmeta So we are a fucked lol

@barkmeta @LeviCMoore 🤣

Let me explain what a $2 billion “cow collar” actually is 👇 They put an AI collar on a cow. It tracks location. Health. Movement. Behavior. Every second of every day. A farmer opens an app. Draws a line on a map. That line becomes a fence. No physical fence. No wall. Nothing visible. When the cow gets close to the boundary.. the collar vibrates. The cow turns around. Within 10 days the animal doesn’t even test the boundary anymore. It just stays inside. 700,000 animals are already wearing them. They called it a “cowgorithm.” They want you to laugh at it. Now read the technology again without the word cow.. 24/7 GPS tracking on every individual. Real time health and behavior monitoring. Invisible boundaries drawn from a phone. Movement controlled through vibration and sound. Subject learns compliance within days. $2 billion. And guess who led the investment… Peter Thiel. The same man who built Palantir. CIA backed surveillance from day one. The same man who just got a $10 billion Pentagon contract to run AI inside the military. His entire career is building systems that track and control. Now he’s funding a collar that does exactly that. They’re not investing in farming. The farming is the test… Most people have no idea what’s coming…

Hear me out… Venezuela winning the WBC is kinda like United States still winning it considering we took their entire fucking country a couple a months ago… Good for them though. #WBC #BaseballTwitter

@NoLimitGains @grok summarize what this means for someone who does not understand private credit.

🚨 US banks have quietly funneled $257 billion into private credit markets Do I even need to say it?

Bam with the greatest scoring performance in NBA History hands down! When Wilt scored 100 the league’s average height was 6’5 and 75% white men who still worked day jobs, ripped cigs, and shot gunned beers after the game. Scoring 83 in today’s league is different! #NBA #BAM

BAM 83 points in today’s NBA game as a big! If you told me someone would surpass Kobe this year Bam would not have been in my top 10. Cherish the moment brother! #NBA #Bam

@Freudy 1st ever NHL game so best believe I tailed!!!

PSA: I put double the normal units on the F5 today

@Freudy FEED ME HOCKY SLIPS

Massive notebook today 🤝 Let’s keep the momentum moving 🫡

@Freudy Smacked for me yesterday! Let’s run it back tonight!

POD: ✅ Freudy 3: ✅ Frozen Five: Kings are dogshit Pucklines 3: ✅ Checkmarks: 6-1-1 We are back 😈

Our current political system is a comedy show! Our commander & chief is giving a historical speech to the World and while we broadcast this globally, the whole world sees a room divided. Half the room stands in support while the other half sits, pouts, and yells at our leader. No wonder nothing ever gets done that actually helps the people of our great nation. Just powerful pedo’s having a pissing contest on a global stage. #StateOfTheUnionAddress