Infomaniac 釈情 Godparticle

Ex-milspook, psycholinguist.
100s of browser tabs open
retweets are not endorsements. I leave breadcrumbs. Draw your own conclusions.
godparticle@substack

[Infomaniac 釈情 Godparticle @cainamofni]

It was only a matter of time. Nothing but conspicuous avoidance and name-calling.

[Infomaniac 釈情 Godparticle @cainamofni]

@I_Am_JohnCullen @BretWeinstein We know that you are not a moron, John. So why use multiple logical fallacies in one breath? You are an astroturf limited hangout.
No plane hit the Pentagon. It was a missile.
Thousands of cameras at UVU captured several mini drones but not one frame of a large one. Right.

[Infomaniac 釈情 Godparticle @cainamofni]

@XMoney please stop doing stupid shit like this. My x account was created while I was physically in the USA. You know this. You literally know that I am a US customer through my validated bank accounts. You SHOULD know that many US citizens live abroad and must use VPNs to conduct essential business in the US.
1. you invited me to xmoney
2. you assume I am not a US citizen based on IP, then
3. when I use a VPN, you reject your offer

There is nothing intelligent about any of these policies. You have all this technology and can't do better than this?

[Infomaniac 釈情 Godparticle @cainamofni]

The researchers—Korbak, Mikita Balesni and Jasmine Wang—dispute [OpenAI's] account. In an interview and a statement, the three said they never received a written explanation for what sensitive information they allegedly mishandled, and spent their final months at the company pushing for more industry collaboration on safety.

In particular, the fired researchers said they had raised concerns inside OpenAI about potential new AI designs, or “architectures,” that could accelerate capabilities, but make the systems even harder to understand or control.

[Max Zeff @ZeffMax]

New: The three researchers OpenAI fired last month spent the summer fighting to keep AI's chain-of-thought process visible.

Then, OpenAI asked to see their Signal chats. A week later, they were let go.

More below from @keachhagey, @berber_jin1, and me.

https://t.co/7klEDa8R0M

[Robert Pondiscio @rpondiscio]

This idea--build reading instruction around the child's interest--is how I was trained as a teacher, and what we did in the low-performing South Bronx public school where I taught. It failed and turned bright, capable kids into non-readers. I have spent the last 20 years working against it.

[Kyronis @kyronis_talks]

For 70 years, we were told plastic was safe because it doesn't react with anything.

Nature just published proof that it does. And the trigger is water.

Researchers at Columbia University and UC Berkeley took six of the most common plastics on Earth.

Polyethylene. Polypropylene. Polystyrene. PVC. PET. Acrylic.

The stuff in your water bottles, food containers, pipes and packaging.

They ground them into microplastics and dropped them into plain water.

Nothing else. No sunlight. No chemicals. Just plastic and water.

And the water started changing.

At the surface of every single plastic, it began producing hydroxyl radicals. One of the most reactive and destructive molecules in chemistry.

All six plastics did it. Every one.

To prove it, they added a yellow dye to water with polystyrene. Over 10 days, nearly 90% of the dye was destroyed. The water turned almost clear.

In other liquids, the dye survived. Only water made the plastic active.

In plain English:

The plastic isn't leaking a toxic chemical.

It is turning the water around it into something that attacks.

Then they tested what that does to living things.

DNA and proteins got pulled onto the plastic surface like magnets.

Proteins started clumping together.

And red blood cells that touched the particles were torn into broken fragments.

But here is the twist.

The most dangerous plastics were the ones we trusted most.

Polyethylene and polypropylene, the "safe" plastics in bottles and food containers, were the most reactive of all.

The effect was weaker in blood serum. But it was still there.

And microplastics have already been found in human blood, brains, placentas and arteries.

The lead researcher said it plainly: "Water is a key active player here, rather than just a spectator."

We spent decades asking what chemicals leak out of plastic.

Nobody thought to ask what plastic does to the water inside us.

How many times have you sipped from a plastic bottle today?

[Infomaniac 釈情 Godparticle @cainamofni]

@HowToPrompt__ Devs have been lifting code from stack exchange for years.

[Infomaniac 釈情 Godparticle @cainamofni]

@HowToPrompt__ It should create opportunities for people to become generalists instead of specialists. Net positive.

[Infomaniac 釈情 Godparticle @cainamofni]

@DrJackKruse What are the chances that P. gingivalis enters the brain via root canals?

[John Ʌ Konrad V @johnkonrad]

There’s always an Alex photo.

[Infomaniac 釈情 Godparticle @cainamofni]

@thesupermannx A LLM wrote this.

[Infomaniac 釈情 Godparticle @cainamofni]

It was only a matter of time. Nothing but conspicuous avoidance and name-calling.

[Infomaniac 釈情 Godparticle @cainamofni]

@I_Am_JohnCullen @BretWeinstein We know that you are not a moron, John. So why use multiple logical fallacies in one breath? You are an astroturf limited hangout.
No plane hit the Pentagon. It was a missile.
Thousands of cameras at UVU captured several mini drones but not one frame of a large one. Right.

[Infomaniac 釈情 Godparticle @cainamofni]

The paper studies distilling from token-based teachers to byte-level students via new logit mapping methods, not ditching tokenizers or ending the LLM era. Byte-level models predate this work.

https://t.co/j2rS5qYNdk
https://t.co/vcnrPaz1CY

[Superman @thesupermannx]

Meta published a paper that might end the current LLM era.

They proved that we can kill the "tokenizer" forever and it actually makes the models stronger.

Every AI you use today (ChatGPT, Claude, Gemini) reads text the exact same way. It chops your words into arbitrary chunks called tokens.

It relies on a massive, bloated vocabulary of around 100,000 specific chunks just to understand a sentence.

We assumed this was the only way to build AI. We treated it as the fundamental speed limit of language models.

Meta proved we were completely wrong.

Their new paper, "Breaking the Token Ceiling," ditches tokens entirely.

Instead, they forced the AI to process raw bytes. A vocabulary of just 256 values.

The results rewrite the rules of model efficiency.

By training on pure bytes and distilling the knowledge, these tiny models didn't just survive. They dominated.

Meta's distilled 1B-parameter byte models asymptotically surpassed state-of-the-art token models like Llama 3.2-1B and Gemma-3-1B across downstream tasks.

They use a fraction of the vocabulary. They require less data. They take up less storage.

It forces the model to actually learn the fundamental relationships between raw data instead of relying on pre-packaged text chunks.

This is a massive shift for anyone building, training, or deploying AI.

We’ve spent the last five years obsessively optimizing how we chop up text to feed these models.

It turns out, the bottleneck wasn't the intelligence of the model.

It was the tokenizer.

[Nicolas Hulscher, MPH @NicHulscher]

🚨BILL GATES/BOURLA TRIAL: “These injections are a BIOLOGICAL WEAPON.”

“COVID-19 injections are LETHAL.”

“The global deployment of this biological weapon can only take place through the UNLAWFUL deception of the public on a global scale.”

“The defendants [including Bill Gates and Albert Bourla] are implementers of the ‘COVID-19: The Great Reset’ project and are GUILTY of this deception; they are therefore being sued individually in these proceedings.”

[Sockfeet Research @sockfeetrsrch]

Let’s look at some of the economics gold production via nuclear transmutation : what are the minimum theoretical thresholds we could potentially deal with here? Or, borrowing Elon Musk’s parlance, what is the molecular idiot index of gold?

As I see it, the economics of the gold production via fusion will critically depend on whether the gold is produced as byproduct of electricity generation or as the primary product of a dedicated reactor.

The mercury transmutation reaction is as follows:
Hg-198 + n -> Hg-197 + 2n (endothermic, -8.49 MeV/atom)

Followed by:
Hg-197 + e- -> Au-197 + nu_e (exothermic)

The first step removes a neutron from mercury-198 which requires energy. The second step then releases energy through radioactive decay.

The minimum kinetic energy required for a neutron to displace the additional neutron from the mercury nucleus can be taken as 8.528 MeV/atom.

1 kg of gold has (1000 g/kg) / (196.967 g/mol) * N_A = 3.06 * 10^24 atoms/kg Au

Using the minimum incident-neutron threshold, the minimum energy required per kg of gold is then:

(3.06 * 10^24 atoms / kg) * (8.528 MeV / atom) * (1.60218 * 10^-13 J / MeV) / (3.6 J/MWh) = 1,160 MWh/kg Au

Let’s assume you get a perfect conversion of electricity into useful incident-neutron energy and using an bulk electric price of 10 cents/kWh ($100/MWh), you get a theoretical minimum economic threshold of $116,040/kg or $3,610/oz.

Where this begins to get really interesting is considering the impact of fusion on the overall energy accounting.
The deuterium-tritium fusion reaction is as follows:

H-2 + H-3 -> He-4 + n (+17.6 MeV)

Assuming one fusion reaction per gold atom, fusion could theoretically give you up to 2,395 MWh/kg Au where 1,160 kWh/kg Au would be consumed by the transmutation of mercury, leaving the overall process exothermic. Strictly speaking, the theoretical minimum external electricity consumption for the mercury transmutation could be zero in a self-sustaining fusion reactor.

In an idealized world, the cost of gold production could only be a function of the raw material inputs into the system, decoupled from any intrinsic energy expenditures. Of course, in the real world there will be things like actual equipment costs but we won’t worry about those pesky things here :)

From a material input perspective, the main constraint here appears to be the need for lithium-6 which is required for the sustainable tritium production:

Li-6 + n -> H-3 + He-4

Tritium on its own cost somewhere in the ballpark of $30,000/g. The idealized reaction would take 28 g of tritium for every kg of gold. If tritium cannot be produced in-situ, then the economics basically fall apart:

$30,000/g * 28 g tritium / kg Au = $840,000 kg Au or $26,130/oz

If, however, tritium can be effective bred from lithium-6, then the overall picture become much more interesting:

Natural Hg
- cost: $50/kg
- requirement: 10 kg / kg Au

deuterium
- cost: $2/g
- requirement: 18.7 g/kg Au

lithium-6
- cost: $2/g
- requirement: 55.0 g/kg Au

total unit cost: $649.2/kg Au

From a raw material perspective, the theoretical minimum floor for the gold production then becomes ~$650/kg or $20/oz!

For an energetically unconstrained production scenario, that would imply a molecular idiot index of 200 using today’s gold prices!

Of course, all of these numbers are idealized theoretical values so you could likely apply an order of magnitude or two on top of this, but its interesting to lay out where this approach could potentially head towards.

Food for thought...

[Infomaniac 釈情 Godparticle @cainamofni]

"Trump threatened the International Order Council on Foreign Relations, Washington, D.C., Brookings Institute types. They were never told their wars were stupid, or their generals were losers, or that NATO didn't step up, or that China would unleash a virus. Let's call it the China virus and hold them to account for it.

All of those things were a calling out of an international order that was failing and is continuing to fail, and now has a great advocate in the regime currently in the White House and in Ukraine. It's basically a fight: Donald Trump by fighting Putin through Zelensky's war. They finally found a war they just absolutely love."

[DataRepublican (small r) @DataRepublican]

🧵 THREAD: The true reason Pete Hegseth is being targeted is because he's standing between President Trump and a coup

@PeteHegseth named the institutions... CFR, Brookings, the general class... in 37 seconds in a video by @Liz_Wheeler . Within 72 hours of his nomination, a color revolution planning document cited him as an insider threat.

They didn't go after him because of drinking. They didn't go after him because of women. They went after him because every color revolution manual ever written says the same thing: you cannot topple a government unless the security forces defect. And a loyal Secretary of Defense is the one person who makes sure they don't.

I have the receipts. Their own documents. Their own training sessions. Their own words on camera.

As always, patience as I pull together the thread. 👇

[Nicolas Hulscher, MPH @NicHulscher]

The DNA of some COVID-vaccinated victims has been permanently modified by Pfizer or Moderna.

A study found that Pfizer mRNA can be reverse-transcribed into DNA in human cells within 6 hours.

We found genetic code from an mRNA "vaccine" fused into chromosome 19 in a turbo bladder cancer victim.

Indefinite spike protein production is wreaking havoc on a large portion of the global population.

There is no known off-switch for those who have experienced an INTEGRATION EVENT.

Those responsible for this global crime must be held accountable.

[Miranda Devine @mirandadevine]

Ahem. Five years ago I wrote a column about DNA test results that showed there was a 99.999998 percent chance that Ilhan Omar and her second husband Ahmed Elmi were siblings. A few hours before we went to press my informant was arrested by Biden's FBI. Funny that.

[Superman @thesupermannx]

Sony built an AI that discovers scientific facts on its own.

And it reportedly discovered actual previously unknown facts about "aging"

It's called "Hakken"

Hakken analyzes millions of research papers, maps how scientific facts evolve over time, and predicts exact, undocumented relationships that humans haven't found yet.

It generates a mathematical confidence score for scientific facts that do not currently exist in any literature.

Researchers pointed it at the biology of aging. The AI generated 1.5 million brand-new hypotheses.

But they didn't stop at the computer screen.

They took the AI's best predictions and physically tested them in a real-world wet-lab.

And the results are historic.

Physical lab confirmed two previously undocumented biological interactions in the human body (involving TP53, BAMBI, RAF1, and TNF).

the AI successfully saw the invisible mechanics of biology that human scientists had missed. And the lab proved it was right.

We are crossing a massive threshold.

We used to build AI to read scientific papers faster. Now, AI is writing the next chapter of human knowledge before we even run the experiments.

The bottleneck for scientific progress is no longer human intelligence.

It is how fast the physical labs can keep up with the machine.