Artemis2 vs Apollo

This is HUGE: AI just learns 100× faster using logic instead of brute force

Scientists have developed a new kind of AI that combines neural networks with symbolic reasoning.

Instead of blindly guessing millions of times, this system actually reasons before acting.

In their neuro-symbolic design, the AI uses structured rules like shape, order, and planning to guide decisions, rather than relying only on pattern recognition.

Results were shocking 👀!

It achieved 95% success, compared to just 34% for traditional models and still solved harder unseen tasks with 78% accuracy, where others completely failed.

And Training took only 34 minutes, instead of over 36 hours, while using 100× less energy.

This could change everything, because modern AI systems already consume massive amounts of electricity, and that demand is rapidly growing.

By making AI think instead of guess, this approach could unlock faster, cheaper, and far more efficient intelligence.

Larry Ellison: “When WhatsApp sold for $19B, a lot of us were shocked… until we thought about it”

In 2014, Mark Zuckerberg bought WhatsApp and their 450 million MAUs for $19 billion.

“A lot of us [in Silicon Valley] were shocked — until we thought about it and understood the value of having access to that many consumers,” Oracle founder Larry Ellison says in this interview that took place a few months after the acquisition closed.

Ellison points out that you have to take into consideration how much traditional businesses (e.g. cable TV companies) pay to acquire customers:

“This is not the place to debate how much you can get through WhatsApp — $1 per year per customer is not going to do it — but I believe there will be opportunity to sell these same customers other things as they join your ecosystem.”

He continues:

“There was quite a battle between Facebook and Google over that property. It wasn’t that one guy did something really silly with a lot of money. Mark Zuckerberg won an auction over Google and paid a high price but got a very valuable asset in the 21st century.”

As of 2025, WhatsApp has approximately 3.1 billion MAUs globally.

$NVDA is building an AI empire across the stack, and the investments tell the story.

You’ve got names like $INTC and $SNPS on the semiconductor side, $LITE, $COHR, and $MRVL in optical and networking, $CRWV in GPU cloud, $NOK in telecom infrastructure, and even $LLY and $RXRX in AI-driven drug discovery.

That’s what makes this bigger than one product cycle.

Nvidia isn’t just selling picks and shovels. It’s buying exposure to the roads, rails, and power grid around the AI gold rush.

That’s how real leaders extend.

They stop acting like a company with one product and start acting like a platform with stakes in the whole ecosystem.

$RXRX/USD H2 #Indicator #Signal #Trading #GlobalOneAI

$MRVL/USD H2 #Indicator #Signal #Trading #GlobalOneAI

$PLTR Palantir CEO Alex Karp with Nvidia CEO Jensen Huang: AI = American Intelligence 🇺🇸

“AI makes America the dominant country in the world.”

“We have the right chips, software, engineers, culture…”

“It’s combining into a juggernaut.”

$PLTR/USD H2 #Indicator #Signal #Trading #GlobalOneAI

CIA Director John Ratcliffe:

We deployed both human assets and exquisite technologies that no other intelligence service in the world possesses.

Reporter: Deliberate attacks on civilian infrastructure are against the Geneva Conventions.

Trump: Who are you with?

Reporter: NYT.

Trump: You are failing.

🇮🇷🚀🔥💣 "The video footage shows a fire at the South Pars petrochemical facility in southern Iran, which was bombed today by Israeli Air Force fighter jets…

🏭 This plant accounts for 85% of Iran's petrochemical exports."

🇮🇱🇮🇷 The IDF says it attacked dozens of Iranian aircraft and helicopters overnight across three airports in Tehran, as well as military infrastructure at the sites.

🚨 Do you understand what's happening right now..

Trump: "make a deal in 48 hours or we're blowing up the whole country".. made "expletive-filled threats" about Hormuz..

Iran's response.. a 10-clause counter-proposal:

> permanent end to ALL wars — Iran, Lebanon, Gaza
> toll rights on Hormuz
> every sanction lifted
> nuclear enrichment in writing
> reparations
> reconstruction aid
> security guarantees

the US wanted a 45-day pause.. Iran wants a new world order..

and while diplomats exchange paper through Pakistan.. the IRGC is ALREADY collecting informal tolls on ships passing through the strait..

Iran isn't waiting for a deal.. Iran is building the toll booth while negotiations happen..

the system worked for someone today.. and it wasn't the one making threats..

Trump says Iranian civilians want him to bomb their infrastructure.

Q: Why would Iranians want you to blow up their infrastructure, to cut off their power?

"They would be willing to suffer that in order to have freedoms. 'Please keep bombing.'"

🚨 US DATA CENTER BUILDOUT HITTING A MAJOR BOTTLENECK

Nearly half of planned US data centers are now being delayed or cancelled due to power shortages.

Companies like Amazon, Oracle, Meta, Google and OpenAI have committed over $600B to AI infrastructure this year.

Critical electrical components like transformers, switchgear and batteries can’t keep up with demand.

Even with increased imports from China, plus sourcing from Canada, Mexico and South Korea, supply is still falling short.

Trump on NATO:

It all began with Greenland. We want Greenland, and they don’t want to give it to us, and I said, “Bye-bye!”

AI companies are piling into biotech because biology is one of the few markets where better prediction can directly compound into enormous economic value.

The economics alone explain the urgency: Deloitte estimates the average cost to take an asset from discovery to launch at about $2.23 billion in 2024, which means even modest gains in hit-finding, trial design, or regulatory workflow are worth real money.

Biology also fits the machinery of modern AI unusually well, because it is built from sequences, structures, images, papers, assays, and clinical records that can be encoded, searched, and modeled at scale.

AlphaFold made that tangible by expanding structural knowledge from a scarce experimental resource into a database covering roughly 200 million proteins, which is exactly the kind of information compression investors notice.

Then the capital followed: Isomorphic Labs raised $600 million, Nvidia invested in Recursion, Sanofi partnered with OpenAI and Formation Bio, and Lilly just expanded its Insilico collaboration to a deal worth up to $2.75 billion.

Biotech is not attractive because language models can write elegant summaries of papers, but because biology offers something most AI markets do not: expensive decisions, abundant data, and hard external feedback from the lab, the clinic, and the regulator.

That feedback loop is why this vertical matters so much.

A chatbot can sound right and still be useless, but a drug model eventually faces binding data, toxicity data, trial outcomes, and approval standards, which makes biotech a brutal but credible test of whether AI can become an industrial tool.

The caution is that progress will be slower than the money suggests, because even strong models do not erase wet-lab bottlenecks or clinical uncertainty.

---

Biology is a brutal filter for AI hype, because the world pushes back. A model can summarize a pathway beautifully and still fail to predict what a cell will actually do.

Healthcare and biotech punish shallow intelligence more than most software markets do, since the gap between correlation and causation is expensive, regulated, and sometimes dangerous.

That is precisely why AI companies keep investing: if they can survive biology’s reality check, they are no longer selling intelligence as performance, but as production.

So a deal like this (Anthropic's $400mn acquisition of stealth-mode AI biotech startup Coefficient Bio), is less about adding another vertical and more about buying access to a proving ground where language models must be tied to real workflows, domain expertise, and empirical feedback.

H100 GPU rental prices just surged 40% in 6 months.

From $1.70/hour to $2.35/hour. Capacity is effectively sold out across the board. Blackwell lead times are stretching into mid-2026.

Here’s what that means in plain English:

The companies that ALREADY SECURED power, land, and grid connections for AI compute are sitting on scarce infrastructure.

$IREN. $NBIS. $CIFR.

The four largest hyperscalers plan to spend $700 billion on AI infrastructure in 2026.

That’s not a forecast. That’s disclosed capital.

GPU prices going up is the neocloud business model working exactly as designed.

Short contracts. Rising rates. Locked-in capacity.

Still holding all three.

$IREN/USD H2 #Indicator #Signal #Trading #GlobalOneAI