The data IS inconclusive on whether I give a shit.

Good explanation but if I can TL;DR something here that is basic but a lot of geologists don’t explicitly introduce as context to lay people because they are silly…

Think of earth as a lava lamp where hot goop rises to the surface and cools into new surface area at mid-ocean spreading ridges and is shunted along the conveyor belt of plate tectonics along until being forced back down underneath another plate and recycled back into goop.

Material is recycled this way, what is significant is this might be a sample of OG earth goop and it might suggest some of the OG earth goop doesn’t mix well with modern earth goop and it might influence modern indigestion in earth as in Hawaii or Yellowstone type stuff.

Remember, we are talking about goop in a lava lamp here, ostensibly it is always mixing so somehow finding a bit of goop that has remained isolated in composition for a long time is going to be fascinating to big lava lamp nerds.

Environmental Conservation -> because only a psycho would want LESS kinds of cats in the world

Aksala-Dnalneerg is the resource rich border region in the far tundras of the north between Alaska and Greenland, it is largely undeveloped because no real estate developer has had the proper vision for it yet (a great flat low expanse with easy access to water).

Can’t we just fly Trump to a random ice covered rock and just hire some people to pretend it is Greenland and he can have a little senior assisted living situation inside Trump Castle in Newer New York City, Greenland safely isolated from the rest of us?

This is a good thing honestly, fuck Spotify it ruined music as much as any single company/service could.

How do you get all the candy into the baseball bat tho?

Is this what hearing Vogon Poetry is like?

Republicans obviously take majority responsibility for this turd of a deal and the corrupt shifting of TikTok ownership to Trump’s buddies. But it can’t be overstated what an own-goal supporting this whole dumb thing was for Democrats, who not only helped Trump’s friends steal partial ownership of TikTok, they saber-rattled over a ban during an election season where they desperately needed young people to vote.

As I’ve spent years arguing, if these folks were all so concerned about U.S. consumer privacy, they should have passed a functional modern internet privacy law applying to all U.S. companies and their executives.

The World’s Leading Pioneer Of Disrupting The Genocide Industry With Cloud Computing And AI

–> That Is The Microsoft Edge

See what YOUR small business can accomplish with Microsoft Azure now!

Try asking Cortana about Smart AI Assisted Pogromming Technology that will help you build the society of your dreams!

ActivityPub is extensible, though. As part of our E2EE program, Mallory, Tom and I adapted the Messaging Layer Security (MLS) standard as an extension of ActivityPub to make the MLS over ActivityPub specification. The protocol fits the great MLS E2EE system onto the ActivityPub API and federation protocol.

But a protocol specification is not enough; it must be implemented. That’s why we’re so happy to announce that the Sovereign Tech Fund has commissioned work with the Social Web Foundation to coordinate two new interoperable implementations of MLS over ActivityPub. This investment by the Sovereign Tech Fund will help move the Fediverse towards more privacy for social web users, no matter what server they use.

This locks in the coming collapse by destroying the possibility of evolving our responses to the climate crisis intelligently and according to real world data. I would say buckle up, but I am not sure how much a seat belt is going to do for any of us at this point but convey a dangerous and false sense that things will work out ok in the end.

What a shithole country I live in.

“The current LLM tech landscape positions [neurodivergent people] to dominate,” according to the application. “Pattern recognition. Non-linear thinking. Hyperfocus. The cognitive traits that make the neurodivergent different are precisely what make them exceptional in an AI-driven world.”

What a load of bullshit, LLMs will be used in a million ways to sideline neurodivergent people in society whether it be BS AI “help” for a neurodivergent student replacing a human teacher or job applications using AI to illegally screen and filter out neurodivergent people, this is a bad decade for neurodivergent people and it is likely only to get worse as societies collapse into bigotry from the endless stresses and catastrophies of runaway climate change.

“Is this a sub for ants?!”

Sure, but personal blogs, esoteric smaller websites and social media are where all the actual valuable information and human interaction happens and despite the awful reputation of them it is in fact traditional news media and associated websites/sources that have never been less trustable or useless despite the large role they still play.

If companies fail to integrate the actual valuable parts to the internet in their scraping, the product they create will fail to be valuable past a certain point shrugs. If you cut out the periphery of the internet paradoxically what you accomplish is to cut out the essential core out of the internet.

In the realm of LLMs sabotage is multilayered, multidimensional and not something that can easily be identified quickly in a dataset. There will be no easy place to draw some line of “data is contaminated after this point and only established AIs are now trustable” as every dataset is going to require continual updating to stay relevant.

I am not suggesting we need to sabotage all future endeavors for creating valid datasets for LLMs either, far from it, I am saying sabotage the ones that are stealing and using things you have made and written without your consent.

I made this point recently in a much more verbose form, but I want to reflect it briefly here, if you combine the vulnerability this article is talking about with the fact that large AI companies are most certainly stealing all the data they can and ignoring our demands to not do so the result is clear we have the opportunity to decisively poison future LLMs created by companies that refuse to follow the law or common decency with regards to privacy and ownership over the things we create with our own hands.

Whether we are talking about social media, personal websites… whatever if what you are creating is connected to the internet AI companies will steal it, so take advantage of that and add a little poison in as a thank you for stealing your labor :)

Not arguing against trying to stop this as much as possible but I also recommend assuming your website will be scraped by bots and taking advantage of that to poison all the AI models you can. Feed in nonsense to the robots in places on your website that aren’t public facing to humans on your website, have 5% of your content be blatant nonsense that asserts obviously untrue statements confidently but in a way that doesn’t disguise the clear intent of purposeful absurdity to human viewers.

See it as an opportunity not a vulnerability. Text is cheap, it doesn’t even really take up storage space on your website so why not?

Be the change you want to see, from everything I have read it takes a shockingly small amount of “poisoned” information to undermine AI models, especially if multiple different non-consenting inputs to an AI model are participating in this strategy the impacts will grow exponentially as problem bits of data mix and mingle and become impossible to fully extract from bulk datasets scraped from the internet.

Thank you!

Experience in the past few years makes it seem that the viability of tank-based warfare has dramatically declined.

I do disagree here though, I think this is a serious miscalculation that arose from as a narrative primarily from two things. The first was Ukraine having to innovate with what they were actually given (not enough traditional AT) and had access to in order to stop Russian assaults (quadcopters) and the second is Russian armor has fatal flaws that haven’t been meaningfully been addressed despite decades of feedback and indicators of those fatal flaws.

Drones have radically changed land warfare, but in the end I think they will make armored vehicles more crucial as part of combined arms land operations.

Take the Bradley for example, it simply outclasses almost all Russian armor, Russia can’t compete even against much older cold war western military equipment like this. On armor thickness alone most Russian armor fails to meet battlefield realities, even smaller artillery calibers shred their armor to pieces. This forces Russia to focus on drone tactics and also to HEAVILY propagandize the idea that traditional armored vehicles are obsolete lest they look weak and stuck in the past on a dead end of obsolete armor design like they are.

Drones have transformed the role of armor not made it obsolete, Russia is just trying to desperately bullshit the rest of the world this isn’t the case with a firehose of propaganda about it.

Look at the most recent iteration of the Abrams, it incorporates a capacity for hull mounted PERCH systems for launching loitering muition/surveillance drones from within the vehicle, integrating the use of drones tightly in with the use of main battle tanks and infantry fighting vehicles, further the CROWS system on Abrams tanks highly emphasizes the capability to observe and target fast moving targets with advanced optics and apply kinetic force to them. The Bullfrog turret program meant for Bradleys and other armored vehicles fulfills a similar role. This is the way forward rather than considering tanks obsolete unless you build a massive unwieldy metal cage on top of them and pretend artillery and other direct fire weapons don’t exist as decisive counters.

Drone cages/cope cages are likely here to stay, I am talking about the Russian turtle “tanks” that are basically barely moving deathtraps for the crews.

As a modular system, PERCH is designed to be simply bolted onto an armored vehicle; in the case of the Abrams, it is fixed in place using existing attachment points. In the MARS event, PERCH was operated via a tablet interface, although GDLS says that future iterations will be fully integrated with existing vehicle computer systems.

By utilizing the Switchblade, PERCH provides the vehicle with not only extended-range surveillance but also over-the-horizon lethality. In certain circumstances, this can even be extended to beyond-line-of-sight (BLOS), in which the loitering munition is used in an autonomous, preprogrammed mode to fly a route and/or hit a fixed target.

https://www.twz.com/land/m1-abrams-tank-armed-with-switchblade-drones-tested-by-army

The Bullfrog is equipped with a .50 caliber (12.7mm) weapon and a cyclic rate of fire of 600 rounds per minute. It is designed to defeat Group 1 through Group 3 UAVs and features both autonomous and semi-autonomous engagement modes. At just 165 pounds without ammunition and accurate to less than 1 MOA, the system is optimized for mobile operations and fixed-point defense.

Company specifications state the Bullfrog can engage aerial targets at ranges of up to 1,500 meters. In addition to battlefield deployment, the system can be used to protect critical infrastructure such as power substations.

https://defence-blog.com/bradley-abrams-get-drone-defense-upgrade/

The muscle memory is hard to fix

Also this is software, we should celebrate and embrace the fact that the same tool can be customized to look and be organized differently to maximally ease users into learning it. This is one of the super powers of software!