🇺🇸 AI imposter targets Trump inner circle with fake Susie Wiles messages

The FBI is investigating a string of spoofed texts and calls from someone pretending to be White House Chief of Staff Susie Wiles. The imposter used AI to mimic her voice and reportedly asked for cash and pardon lists. Lawmakers flagged the messages for bad grammar, odd tone, and incorrect details.

The FBI ruled out foreign involvement and warned in May about a broader “smishing and vishing” campaign. Wiles suspects her personal contacts were hacked. Her email was previously targeted by Iran’s IRGC during Trump’s 2024 campaign. Investigation ongoing.

@aipost 🪙 | Our X 🥇

⚡️ When AI burns too much power... You send it to space?!

Former Google CEO Eric Schmidt wants to launch AI data centers into orbit because Earth’s grid can’t keep up. He says AI is on track to eat 99% of our electricity, and some companies are already dreaming of 10-gigawatt server farms—10 times the output of a nuclear plant.

His bet? Solar-powered data centers floating in space, built by a rocket startup that hasn’t even reached orbit yet. Because when AI melts the grid, you go solar—or you go orbital.

@aipost 🪙 | Our X 🥇

🤖 Ready or not, AI is starting to replace people

In a recent discussion, Anthropic CEO Dario Amodei expressed concerns that artificial intelligence could eventually replace many jobs, including those of CEOs. He emphasized that as AI technology advances, it may outperform humans in decision-making roles.

Amodei highlighted the need for careful consideration of the implications of AI on the workforce and called for proactive measures to ensure a balanced transition. The conversation underscores the broader debate about the future of work in an increasingly automated world.

@aipost 🪙 | Our X 🥇

🇺🇸 The future of AI depends on nuclear power... And the clock’s ticking

AI eats electricity. A lot of it. By 2030, the U.S. will need 100 gigawatts more power just to keep up with AI demand. That’s like adding the entire power output of California, twice. China is already building for 32 gigawatts. The U.S.? Still debating it. At Idaho National Lab, engineers say nuclear reactors are the only serious option. Big Tech—Amazon, Meta, Microsoft—is pouring money into new reactors, but the bottleneck is fuel: testing it, certifying it, scaling it. So INL is turning to AI—not just to power AI, but to run the reactors too.

In one test, AI operated a 5-watt digital reactor hooked up to the cloud. It predicted student behavior and reactor output with 99.7% accuracy. Another trial had AI managing a heat pipe reactor. Error margin: 0.3%. Less than most humans, and it doesn’t take coffee breaks. Old-school nuclear needs weeks of analysis. AI can test materials 24/7 and do the same job in hours.

No power, no processing. No reactors, no AI future.

@aipost 🪙 | Our X 🥇

🇺🇸 AI hunger grows: California’s power grid is getting absolutely cooked

PG&E, California's largest electric utility, says data center requests for electricity jumped 40%, and these aren’t your grandma’s server farms. Last year, companies wanted 50 to 100 megawatts per site. Now they’re asking for 500 to 1,000 megawatts. That’s like powering a whole city just to teach ChatGPT how to write better essays than you.

New AI megacenters are ditching Silicon Valley and heading inland to cheaper spots like Fresno, because apparently the robots don’t care about beachfront views. PG&E’s problem? Everyone wants to plug in right now, and there aren’t enough parts or patience to go around.

@aipost 🪙 | Our X 🥇

👟 Your sneakers just got smarter than you

Researchers made insoles that use AI to track how your feet hit the ground — with scary accuracy. They predict ground reaction forces (fancy term for how your body smacks the floor) with just 4.1% error.

Athletes, doctors, and even people with diabetes can now track movement outside the lab — like at the gym, on a run, or in recovery.

Congrats, your shoes are now biomechanics experts.

@aipost 🪙 | Our X 🥇

🇳🇴AI-work pressure turns up the heat on European bankers

Imagine every European banker being told to either master AI or watch the door. At Norway’s giant $2 trillion fund, Norges, the CEO has said those who skip AI can forget about promotions. Banks like HSBC are already pushing out analysts to make room for machines that never sleep and never need coffee.

Some financiers see AI as the next steam engine, set to power a new industrial age. This is not just about gadgets. It is about who runs the future of money. Those who cling to the old ways might soon be left counting paper clips instead of billions.

@aipost 🪙 | Our X 🥇

🧬 AI just found 200 million mystery molecules

Mass spectrometry sounds boring, but it's like CSI for molecules... only nobody could read the clues. Enter DreaMS, an AI trained on 700 million chemical "fingerprints" from plants, dirt, food, and your insides.

It finds hidden links between things like pesticides and skin diseases. Yes, really. Old tools needed humans and hard rules. DreaMS just vibes with the data and figures it out. It even learned to spot fluorine, a sneaky atom in most meds, just by watching molecules do their thing.

@aipost 🪙 | Our X 🥇

Perplexity launched Labs mode.

Not just search anymore, now builds trading bots, dashboards, web apps, and scrapes data automatically. Basically turned Perplexity into a full development environment.

@aipost 🪙 | Our X 🥇

🇳🇴 Norway’s $1.8T fund goes all in on AI to crush the competition

No AI, no future. Nicolai Tangen is putting Norway’s sovereign wealth fund on warp drive, telling all 670 staff to fully adopt AI or be left behind.

The $1.8 trillion powerhouse is seeing faster trades, sharper analysis, and soaring efficiency from tools like Claude, Copilot, and Perplexity. AI now writes code, scans news in 16 languages, and even flags CEO pay votes with 95% accuracy.

Tangen’s vision is simple: dominate or disappear.

@aipost 🪙 | Our X 🥇

🇸🇦 Saudi Arabia wants the AI crown, and they might get it

Saudi Arabia launched a $10 billion fund to buy into the future—and they’re aiming straight at America’s biggest tech brains. Their new AI company, Humain (yep, that’s the real name), wants to control 7% of the world’s AI power by 2030. Think: giant data centers, chip deals, and more GPUs than you can count.

They’re already talking with OpenAI, Elon Musk’s xAI, and VC giant Andreessen Horowitz. Amazon, AMD, Nvidia, and Qualcomm? Already signed. The goal? Outbuild everyone, do it fast, and make sure the U.S. is on their side—not China’s.

@aipost 🪙 | Our X 🥇

Geoffrey Hinton says his greatest fear is that humans become irrelevant, replaced by digital beings more intelligent than us.

Want to know how it feels not to be the smartest species? Ask a chicken

@aipost 🪙 | Our X 🥇

What would you do if this was your doctor? 🤔

@aipost 🪙 | Our X 🥇

Sam Altman says the world must prepare together for AI’s massive impact

OpenAI releases imperfect models early so the world can see, adapt, and help shape regulations. "There are going to be scary times ahead"

But collaboration can turn AI into much more good than bad

@aipost 🪙 | Our X 🥇

DeepSeek updated its R1 model, instantly propelling it to the #3 spot on the Artificial Analysis leaderboard with:

— Enhanced front-end capabilities
— Reduced hallucinations
— JSON output & function calling support

@aipost 🪙 | Our X 🥇

Google quietly launched Edge AI Gallery, a library of open AI models to access on smartphones

Currently on Android, the gallery lets users pick models for their task and run using the available device hardware, entirely locally

iOS version coming soon!

@aipost 🪙 | Our X 🥇

ElevenLabs launched Conversational AI 2.0, its next-gen platform for creating voice agents with:

— Real-time turn-taking model
— Built-in language detection
— Integrated RAG
— Multimodality
— Scalable batch calling

@aipost 🪙 | Our X 🥇

Hume launched EVI 3, a new AI that can speak in any voice or personality via speech-to-speech interaction

It uses a voice-to-voice architecture and comes with a deeper understanding of tune, rhythm, timbre, and speaking style

@aipost 🪙 | Our X 🥇

🇳🇿 MP holds up AI nude of herself in parliament to fight deepfakes

New Zealand politician Laura McClure held up an AI-generated nude of herself in Parliament to push a law against fake explicit images. She made it at home to show how easy it is to create deepfakes that can ruin lives.

McClure: “In the general debate in the House, I brought to the attention of all the other members of Parliament about how easy it is to do this and how much abuse and harm it is causing, particularly for our young Kiwis and more likely to be our young females.”

Yes, it was awkward, but it definitely got attention.

@aipost 🪙 | Our X 🥇

🪐 AI exposed the universe’s secrets

Scientists built an AI called SimBIG that looks at galaxy patterns like it's playing cosmic Sudoku, and it’s crushing it. It figured out the “settings” of the universe (like how much dark matter exists) way more accurately, with way less data.

Old methods needed four times more galaxies to do the same thing. Bonus: This could help settle the universe’s biggest nerd fight: How fast it’s actually expanding. If the numbers still don’t add up? Congrats, we might’ve just found new physics.

@aipost 🪙 | Our X 🥇