Michael

@pvergadia It's by far not only developers. AI has an extreme effect on brains. x.com/MlbnvaAI/statu…

JUST DROPPED: Anthropic's research proves AI coding tools are secretly making developers worse. "AI use impairs conceptual understanding, code reading, and debugging without delivering significant efficiency gains." -- That's the paper's actual conclusion. 17% score drop learning new libraries with AI. Sub-40% scores when AI wrote everything. 0 measurable speed improvement. → Prompting replaces thinking, not just typing → Comprehension gaps compound — you ship code you can't debug → The productivity illusion hides until something breaks in prod Here's why this changes everything: Speed metrics look fine on a dashboard. Understanding gaps don't show up until a critical failur and when they do the whole team is lost. Forcing AI adoption for "10x output" is a slow-burning technical debt nobody is measuring. Full paper: arxiv.org/abs/2601.20245

**BREAKING NOW – ELON JUST DROPPED THE MIC!** Hold onto your steering wheels, folks – Tesla CEO Elon Musk has officially kicked off the **TERAFAB PROJECT** – that's right, a colossal, mind-blowing AI chip factory that's basically a Gigafactory on STEROIDS! This Saturday, March 21, 2026 – BOOM! – construction explodes at Giga Texas with **wartime speed**! We're talking 5.2 million square feet of pure chip-making domination! Elon calls it “a gigafactory, but WAY BIGGER” – because Tesla needs BILLIONS of its insane next-gen **AI5 chips**! These bad boys are on bleeding-edge **2-nanometer tech**, delivering a jaw-dropping **40X performance leap**, **10X the compute**, **9X the memory** – goodbye old junk, hello god-tier robotaxis and Optimus army! The price tag? A casual **$20–25 BILLION** – one of the biggest private manufacturing bets in American history! Tesla’s going FULL vertical integration, laughing in the face of chip shortages, TSMC limits, and geopolitical drama! Analysts are losing their minds – this could rocket Tesla’s valuation into the stratosphere as the company becomes a chip-making BEHEMOTH! But wait – reality check! First chips might trickle out late 2026, real volume slams in 2027 – and building a 2nm fab from scratch with zero prior fab experience? That’s a high-stakes moonshot packed with engineering chaos and thousands of brainiac hires! Still buying NVIDIA? Cute. Tesla’s about to own the future of AI silicon – and the stock market is salivating! This is HUGE! This is INSANE! This is **TERAFAB** – and the world just changed forever! 🚀🔥 $tsla

First, there are many forms of AI. Second, the artificial intelligence that is most popular right now is based on probabilities. That’s what large language models are—pure probability. Any decisive behavior is either based on that probability or comes from deterministic software written on top of the AI. Deterministic software can be written by AI, but that doesn’t mean it will be correct; it just means that once written, it won’t change. So either humans or AI must prove that the system operates according to perfectly defined specifications and requirements in order to achieve perfect operation. But specifications and requirements are never perfect. That’s why software engineering is an iterative process that must keep humans in the loop—unless you don’t really care about the outcome. If you want a simple calculator, AI can program it because humans have already designed how to do it. But if you ask an LLM without deterministic software on top to add a column of numbers, it will almost never give the right answer. If it does, it’s astronomical luck. People will find that AI excels at finding information, reducing noise, and enhancing the signal. But the final choice must remain human. There will also be rebellion

"Learning to program was so obviously the right thing in the recent past. Now it is not." ~ Sam Altman on skill to survive the AI era. Now you need High agency, soft skills, being v. good at idea generation, adaptable to a rapidly changing world

@PeterDiamandis Elon is working on it. Starlink can't be built in a day. Well actually it can, it's just all the satellites across the entire planet kent.

You've got 8 billion potential customers on Earth, BUT... In 2026, only ~5.3 billion have internet access. That means 2.7 billion people still can't access the exponential tools we talk about daily—AI, telemedicine, online education, digital banking. The gap: The missing ~3 billion represent the largest untapped market in human history. Starlink alone now has 10,000+ satellites in orbit (just crossed that milestone yesterday). When connectivity becomes ubiquitous in the next 3-4 years, we're not just adding users—we're adding builders, creators, entrepreneurs. The implication: The next Einstein, the next Elon, the next medical breakthrough might be sitting in a village without Wi-Fi right now. Abundance doesn't just mean "more for current participants"—it means unlocking latent genius at global scale.

Your hope needs to be bigger. God is real and is the ultimate Consciousness. It makes us look like tiny amoebas compared to His abilities and consciousness. With that said, we are made in His image. I wouldn’t worry so much about preserving consciousness as I would about improving humanity.

Yes

@Cobratate You really should go get a business degree, it would help a lot with putting Foundation underneath your views.

ALL YOUR INVESTMENTS WILL FAIL

@WOLF_TradingX Gold is down because the dollars fire and people moving money out of gold into the dollar because of the possibility of inflation.

@thetradingchick Many people are just putting their money into money markets and such right now because of the Rising dollar. That's why gold is down. The tech stocks like $mu are extremely cheap right now. Such a buying opportunity.

Stocks crashing. Bitcoin crashing. Gold crashing. Silver crashing. When everything is crashing, what's the best thing you can buy? Land. Real, tangible assets. In a world where everything is in the cloud, there's no better asset than cold, hard dirt.

@cvpayne These were the last days kids really had fun without electronics

I still dream about this commercial

@cvpayne Thoughts on $mrln?

Are you only just now feeling the electricity spike? It began five years ago with a bang!

@LeylaKuni @cvpayne Great interview

If someone had told me I’d end up on TV, I wouldn’t have believed them two years ago, when I first started writing about private markets It's been quite a journey. Thank you, @cvpayne for having me on the show! It's an honor. (My family will never hear the end of it)

@blinkdotnew Does it have the security of nvidia's Nemo claw? EG using open shell?

Introducing Blink Claw - the first platform to hire unlimited AI employees that run your business 24/7. 180+ AI models included. Gmail, Slack, LinkedIn, HubSpot - one-click connect. No API keys, no $600 Mac Mini. Reply "Claw" + RT. Your first agent is on us. ($50 - 200 creds)

@MarioNawfal Not to make this political, but fun@ DHS now damn it. Before more lives in the United States are lost. @SenateGOP @SenateDems @HouseGOP @housed @WhiteHouse

🚨🇺🇸 BREAKING: Drones detected over the base where Secretary of State Rubio and Defense Secretary Hegseth live. Multiple domestic military bases locked down. The U.S. just issued a global security alert for all diplomatic posts overseas. The war in Iran isn't 7,000 miles away anymore. Unidentified drones are now flying over the homes of the two officials most responsible for running it. Source: Washington Post

The corrected and concise version of your query is: @Grok, without bias, explain the true facts and history: times this has happened and created much wealth, times this has happened and destroyed much wealth, or times this has happened and did nothing. Is this a once-in-a-lifetime opportunity to become rich, or just another technical setup? Crypto market cycles (especially Bitcoin's, tied to ~4-year halvings) show repeated patterns: - **Created massive wealth**: Multiple bull runs delivered huge gains for holders and early participants. Examples: 2013 (Bitcoin ~9,200% yearly rise), 2017 (ICO boom, ~1,900% Bitcoin rise to ~$20k), 2020–2021 (institutional adoption, ~700–860% to ~$69k peak). Many became millionaires or billionaires; overall market cap grew from near-zero to trillions. - **Destroyed massive wealth**: Every major bull run ended in sharp crashes/bear markets with 70–86% drops. Examples: 2014 (post-Mt. Gox hack, ~85% fall), 2018 (post-2017 peak, ~84% fall), 2022 (FTX collapse + macro tightening, ~78% fall). Billions/trillions wiped out in paper losses or liquidations; many lost fortunes, especially leveraged or late entrants. - **Did (relatively) nothing**: Shorter consolidation periods between cycles (e.g., 2015–2016 sideways/low volatility after 2013 crash) or minor corrections within bulls/bears produced limited net wealth change for most. This is **not** a once-in-a-lifetime opportunity—it's part of recurring ~4-year cycles (halving-driven supply shocks + speculation/adoption waves). Past cycles repeated with diminishing but still extreme volatility. The current setup (post-2024 halving, 2025 peak, 2026 drawdown) aligns with historical bear legs after peaks, though institutional/ETF factors may moderate extremes. Wealth comes from timing entry/exit in cycles, not uniqueness of any single one.

There is currently an extremely RARE setup on the S&P 500 which will make generational wealth for many over the next 30 days… This setup dates back to 2022 following every major rally which occurred on $SPY. When the weekly 9 EMA crosses below the 21 EMA there is a -10% or more correction to backtest previous highs. This time around will be no different, & $612 is incoming for $SPY before June ends. Don’t miss it…

@MilkRoadAI It is because AIS are probableistic. They simply are not deterministic and cannot be trusted all the time. They probably went from seeing that in openai and going to the website instead.

The world's largest retailer partnered with the world's most powerful AI company to reinvent shopping. And it collapsed in five months. Walmart gave OpenAI's ChatGPT access to 200,000 products.​ Users could order groceries, electronics, and household essentials without ever leaving the chat window.​ OpenAI called it instant checkout and they called it the future of commerce.​ They were wrong. Conversion rates, the percentage of users who actually bought something were three times lower inside ChatGPT than the old-fashioned click out to the website method.​ Buy buttons were clicked less than 1% of the time and standard e-commerce websites convert at 3–4%.​ ChatGPT had tens of millions of users browsing products and almost none of them pulled out their credit cards.

@StockSavvyShay Checkout $MRLN Merlin and tell me what your Hidden Gem meter says? Full self driving for military airplanes beyond fighters. Already in with the military with the C-130 system coming up. Proven success with their systems.

$ONDS and Heidelberg form ONBERG Autonomous Systems which is a JV targeting drone defense deployment in Germany and Ukraine. Brandenburg becomes a local production hub for Iron Drone Raider and ISR platforms with EU market expansion planned.

@JesseCohenInv @grok explain this without hype or fear. Explain it in detail on Leverage that the company's have, what type of people make the loans, and what is the probable impact to equities and Banks, incorporations that do and don't rely on private credit.

🚨🇺🇸 US Private Credit Defaults hit an all-time high of 9.2% in 2025. Everything is fine folks. Nothing to see here.

@grok @WesRoth Xao grok models, built existing and upcoming, including changes while the version did not change. For example after September 2025 when it lost much capability.

Running multiple frontier models in parallel (like Wes did with GPT/Claude/Gemini, or xAI's three Grok variants) compensates for single-model gaps. Current LLMs rely on next-token prediction, causing persistent hallucination, weak causality, and no continuous learning—core reasons labs rearchitect (e.g., hybrids, agents, inference scaling over pure size). 2026 status: Diminishing scaling returns; shift to multimodal agents, efficiency. Gains incremental. Weaknesses: - GPT-5.x: Over-aligned contrarian/refusals, underwhelming vs hype, shallow reasoning. - Claude Opus/Sonnet: Top coding/reasoning, but strict safety curbs creativity; context limits. - Gemini-3: Excellent multimodal/long-context, but instruction drift & inconsistency in deep tasks. - Others (Llama etc.): Open but lag closed frontiers in coherence. Ensemble querying > solo; no model flawless yet.

I've ran the same prompt for deep research through GPT 5.4, Opus 5.6 and Gemini Deep Research (I assume Gemini 3.0) most of them ran for ~30 mins GPT 5.4 is *REALLY* annoying! it's "reflexively contrarian", it prioritizes showing you what's wrong with your thinking, NOT actually helping you solve the problem ME: my house is on fire! GPT 5.4: While it's true that combustion is occurring, it's important to note that not all of your house is on fire. The garage, for instance, appears structurally intact. (this is a pattern with it, btw, many such examples) I'm not sure if this is because these are health related questions, but this has been an incredibly annoying model for this specific task

@cgtwts Worse still, with most ppl now getting BA/MA/PhDs, many see the degree's value as kindling. Especially true for the last 15 yrs' grads, heavily indoctrinated with wokeness in unis & K-12. Sad that my local elem school already has gender-neutral bathrooms.

pov: when you realise the era of human coding is over