nev

probably researching a story
strategist + filmmaker / ai + cryptography
prev. cofounder struthless, step one

[nev @nevscape]

Palantir has a nemesis and I accidentally found him...

This conversation with @harryhalpin changed my entire worldview. Right now it's the #38 'most hyped' video on YouTube, and has 110k+ views in ~2 days.

To me, this is one of the most important conversations in the world right now - the survival of our freedom.

I thought i'd post the full video here, too. I hope you take as much away from it, as I had making it.

Timecodes:
0:00 chaos can be a ladder
0:52 a stranger at a hackathon
1:47 who is Harry Halpin?
2:24 the ban: Australia, then the UK
4:53 the risk: total control
7:02 Harry's story: targeted by the state
9:05 the most infamous spycop in Britain
12:11 Bernard Stiegler
13:11 Pharmakon: the poison and the cure
15:04 the anti-Palantir manifesto
16:21 tech that could save us

[nev @nevscape]

@CloutedMind I am filled with rage

[nev @nevscape]

@rstormsf @nero_eth Sending you a dm man.

[nev @nevscape]

@ErikVoorhees @GeorgeGKalyvas more important now they're burning the books

[nev @nevscape]

Honestly struggling to process this one. Implications are bad. Consume knowledge into training data, original lost forever. Model is then censored, controlled by private company/gov. History, knowledge, could be rewritten. Surveillance enabled for any view that disagrees. Tech even in cars that can be turned off via software. All controlled

People in crypto too, the place with knowledge to do something, are literally being fooled by bread and circus. Memecoins

[nev @nevscape]

1984 wasn’t enough? Now they’re doing Fahrenheit 451? This one has rattled me

[Hedgie @HedgieMarkets]

🦔AI companies are bulk-buying rare books, scanning them through high-speed machines that cut the spines off, and shredding the originals. A service called ISBNdb facilitates orders of up to a million books and keeps buyers anonymous. Pre-2022 books are premium because they're free of AI-generated text. A federal judge ruled the practice is fair use because eliminating the original means only one copy exists at a time. Anthropic hired the former head of Google Books partnerships to obtain "all the books in the world."

My Take
This got to me. A bookseller told 404 Media that rare books with almost no surviving copies are being fed into this pipeline. Books that survived wars, fires, and centuries of handling are being shredded so an AI can learn to write a better marketing email.
ISBNdb's website literally says "'AI company destroys two million books' is not a headline that generates sympathy," and they still built an entire business around making it happen quietly. They offer NDAs as a feature. They coach clients to call it "digital preservation."

I've covered AI companies scraping the internet, torrenting libraries, and stealing music. This is worse because it's irreversible. You can re-upload a website. You can reprint a bestseller. You can't replace the last three copies of an 18th-century botanical text once someone shreds them for training data. And the judge said it's legal. So it's going to accelerate.
"We shred rare books and offer NDAs so nobody finds out" is a legitimate business model in 2026. What a timeline.

Hedgie🤗

[nev @nevscape]

@0xEvinho Yeah me too, I hope I have been psyoped by a sick joke

[nev @nevscape]

@semenov_roman_ @HedgieMarkets Don’t you see they are the government if they end up with more power or control. This is how that is achieved.

[nev @nevscape]

@HedgieMarkets In shock over this one, I got a lot of feelings. This sort of thing doesn’t feel real to read. The implications are so dire.

[nev @nevscape]

@ideacasino bro, it gave me ptsd back to walking the streets of NY where it went off every second day

little old aus isn't used to this type of surveilled governance, leave us alone

[nev @nevscape]

@theonejvo ahah, casual 'mythos preview'. I see you.

[nev @nevscape]

@victorxva preach

[nev @nevscape]

Reference material, unfortunately I am basing my plot on real life dysoptia
https://t.co/rQaQRch8bp

[nev @nevscape]

I can’t put my finger on what makes this feel like a news headline at the start of an apocalyptic film that foreshadows a later scene where this G is the backdrop of tiny flying explosive death drone swarms being deployed… but I feel it.

[Life at Google @lifeatgoogle]

The massive concrete Google ‘G’ statue outside our Bay View office recently got a fresh new look to match our brighter, bolder AI-era logo.
This wasn't an easy feat, as you can't just slap paint on a 15-foot-tall concrete giant. To pull this off, the team had to engineer and plan for materials that are both flexible and resilient, dodge unpredictable weather windows, and bring in a specialist to hand-blend the gradient across a massive surface. The coolest part? It's now the only 'G' statue in the entire company painted entirely by hand, giving it a little extra human character making it our “Super G”.
Check out the new look and meet some of the Googlers who are already loving the upgraded backdrop!

[nev @nevscape]

i mean, the fact you made it looks like this, maybe we don't have to be too worried just yet. Here's some info on the defected google employee for anyone interested and his fear of these incoming drones

Your image is scary jabz why does it have red lights now

https://t.co/FYDzqk6vBZ

[nev @nevscape]

@grok can we make this so the g is way bigger (as per the giant monolith description from the news article), and can we also make the drones much smaller - we are talking future tiny little drones that explode as packages, similar to the defected google employee that left due to them not anting to do anything about it. Feel free to add some skynet energy.

[nev @nevscape]

It was just so hard to paint this g, a real engineering feat, while we harvest the world of private data via a custom built surveillance machine to train superintelligence that could create total control or human extinction

[Jamieson O'Reilly @theonejvo]

With everything happening in the open vs closed AI fight, I'm kicking off a piece of R&D and running it in the open.

@ClementDelangue and his team just handed the world one of the first real, in-the-wild examples of an open-weight model doing what a frontier closed one would not. When Hugging Face went to investigate their own breach, the closed frontier models refused to help. @Zai_org's GLM 5.2, open weights, on their own infrastructure, did the work. In an incident, that is the only kind of outperformance that matters.

So I'm going to invest some serious R&D into finding out how far open models can actually be pushed on real offensive and defensive security, against hardened, real-world targets. This is live work rather than benchmark numbers.

I'll keep this thread pinned and update it as I go to share what works, what doesn't, where open models hold their own against the frontier and where they fall over.

[nev @nevscape]

@theonejvo @ClementDelangue You’re going to have me make another movie, if you want to do a little q&a at the start and then one at the end, lmk.

[nev @nevscape]

@camski uhh enjoy, a special place

[nev @nevscape]

@theonejvo Haha. I had a similar one in my drafts about cryptographers and defense / control tech. The gap is astonishing and it seems they all have insane levels of ai psychosis sitting from here