cyberpunksurvivalguide.com - Profile | ThreadSky | a Reddit-style client for Bluesky

Here’s the thing about #AI. It’s an extremely useful set of tools and we’re just starting to fully realize their capabilities. Which is why the hyperventilating, borderline televangelist-like claims about the capabilities of #LLMs like ChatGPT are getting out of hand, as I try to explain.

submitted 50 minutes ago • 0 comments

The bad news is that the #surveillance state is about to get orders of magnitude worse. The worse news is that given how it's trying to supercharge itself, the end result will be chaos, an ocean of bad data, and terrible decisions that will ruin lives for very stupid reasons.

submitted 20 hours ago • 0 comments

I’m pretty sure that at this point, you could find 70-80% GOP support for setting immigrants on fire at the border as a warning.

submitted 20 hours ago • 0 comments

I remember how effective DARE was at teaching us about drugs and sex and how fun they are…

submitted 1 day ago • 0 comments

Look, as someone who was (just barely) alive during this in Ukraine, all I can say is that this party will be rad(s).

submitted 2 days ago • 0 comments

Even tech executives who know better keep overpromising what #AI can do and how fast it can advance. In this case, the head of a Google project claims that we're on the verge of exploring and colonizing the galaxy.

submitted 2 days ago • 0 comments

WaPo in 2017: “Democracy dies in darkness.” WaPo in 2025: “Mmm… Our favorite, boot! Tasty, tasty boot…”

submitted 4 days ago • 0 comments

This week on Canadian #radio, I decided to keep it nice and light, and talked about war and AI-amplified propaganda based on some eye-opening reports from defense think tanks trying to use #LLMs to decide when to go to war.

submitted 5 days ago • 0 comments

We daydream and fantasize on a daily basis, but we use the same neural pathways to do that as we do to process our environment. So, how does our mind know the difference between reality and imagination? And what does that tell us about why we daydream and how we think? #science #brain #imagination

submitted 5 days ago • 0 comments

The outraged reaction of #AI enthusiasts on social media to Apple's recent paper on the limitations of #LLMs seems way out of proportion given what we know about machine learning and #compsci. But why? Personally, I blame the endless, hyperbolic hype from chatbot startups.

submitted 7 days ago • 0 comments

Since Sam Altman is once again promising that he can deliver a benign #AI overlord with ChatGPT, I'm re-upping this piece on Silicon Valley's cargo cult of #AGI in my new newsletter focused on AI, technology, and humanity's future.

submitted 8 days ago • 0 comments

A few blocks in downtown #LA saw protests. The city is just over 502 square miles. Everything is open. All but a few hundred people are going about their lives with no interruptions. Yet legacy media is desperately trying to paint the city as a war zone because lying to their audience gets views.

submitted 9 days ago • 0 comments

Not only is your brain not like a computer, the idea that we think at the same speed is dead wrong. In fact, we seem to have two processing rates, one for reacting to stimuli and one for deep rumination, planning, and decision-making. And those speeds are wildly different...

submitted 9 days ago • 0 comments

The only Trump EO that I support in any way, shape, or form. Replacing a speed ban with a noise ban for supersonic aircraft will help drive innovation in the aerospace sector while not making anything worse for areas around major airports from which long haul flights routinely depart.

submitted 9 days ago • 0 comments

Meanwhile, in the real world, it was a predictable, boring Sunday for 99.9999% of Angelenos. We aren’t thundering down the 101 or the 405 in war rigs, wearing our best bondage post-apocalypse gear while pursed by low riders blasting aggressive mariachi music. We’re at Trader Joe’s buying groceries.

submitted 10 days ago • 0 comments

Look, it’s really simple. Pee Wee Himmler got cucked by an immigrant so he’s decided to do a little ethnic cleansing in LA as self-care.

submitted 11 days ago • 0 comments

I mean… the guards at Auschwitz were real people with real families too, so he’s not exactly wrong.

submitted 11 days ago • 0 comments

Computer scientists are always interested in how to write and run software in extreme environments. This is why a few of them created an entire ecosystem for running machines after the end of the civilized world. But do we really need it? And how and why would we compute after the apocalypse?

submitted 12 days ago • 0 comments

submitted 13 days ago • 0 comments

Seed oils are not killing you. Grocery store food is not killing you. Corn is not killing you. Fast food is not killing you. Sweets are not killing you. Coffee is not killing you. Nothing that social media influencers jacked on steroids say is killing you is actually killing you. 🧵

submitted 13 days ago • 1 comment

Okay, no. No one is taking Space KKKaren back into the Democratic tent. It’s not happening. The last person to do that was Byrd and he left the Klan as a young man, then had a 30 year long apology tour to atone for his bigoted beliefs.

submitted 14 days ago • 0 comments

So, hey, fun fact. You know what other country believes there should be no foreign imports of anything or international trade, and that self-reliance and isolationism are all that matters? North Korea. Trump and his merry band of morons don't want reciprocity. They want Juche.

submitted 14 days ago • 1 comment

Not everything happening with massive #AI models is necessarily bad. A new 1.3 billion+ parameter model called Aurora may offer us much better weather and storm forecasts, saving lives and tracking climate change more accurately.

submitted 14 days ago • 0 comments

If you don't remember the Iraq War, it was started because the Bush administration wanted to settle old scores and lied to the world. Now, defense think tanks want to use LLMs to do the same thing, but using a smokescreen of technobabble. www.cyberpunksurvivalguide.com/p/pentagon-u...

submitted 15 days ago • 0 comments

“You just need to send a request to our API.” “Oh, you mean your backend Application Programming Interface?” “No, no, no, I mean our Actual People in India.”

submitted 15 days ago • 0 comments

When the person in charge of the government body that makes sure your children are educated thinks AI is actually called A-One, can’t multiply by ten, and doesn’t know her own budget… She makes the pyramid scheme lady in the first Trump term look like a genius by comparison.

submitted 15 days ago • 0 comments

RFK Jr: “I want studies that say vaccines are bad, roadkill is good for you, and science is evil.” ChatGPT: “Here’s a bunch of garbage with imaginary conclusions.” RFK Jr: “Perfect!” Remember, they don’t want good AI. They want a digital yes man to defend their ridiculous fantasies with bullshit.

submitted 19 days ago • 0 comments

In an age of scams, rent-seeking, MLMs, and mass enshittification, is it any wonder that the “creator economy” is turning out to be yet another false promise of escaping the increasingly frustrating rat race? www.cyberpunksurvivalguide.com/p/the-creato...

submitted 19 days ago • 1 comment