Hacker Newsnew | past | comments | ask | show | jobs | submit | transcriptase's commentslogin

Perhaps they take after the CEOs of the companies that created them

Bro, good joke, the truth is much darker.

They take after humanity, they were trained on us after all...

When you look at an LLM... you are looking at a mirror. The thing looking back looks like you, yet is not human.


Worse trained on humanity in the online world, which a brief comparison of the sewage section on social media is far worse than people in the real world.

“What are the odds that the first three quote tweets of Coxon’s post would all be major AI-restriction policy advocates funded generously by the same donor, who also happens to be one of the leading investors in, and a board member of, Anthropic, the company Coxon was resigning from? And all within 15 minutes of posting (two within ten)?”

https://x.com/parkerthayer/status/2097759699626328575


This is more of the same "all AI opposition is a psyop" bs we see with data centers, and the author of this tweet works at a think tank consistently peddling that: https://capitalresearch.org/tag/ai/. Quote tweeting someone within 15 minutes of their cry for help is not evidence of psyop.

The cultural fervor they're groundswelling isn't a psyop. That part is already out of their control. It's an enormous mistake. Maybe once they've accomplished market capture via regulation they won't care that it's a mistake, however the consequences will be pretty bad in what they rile up.

If the labs are orchestrating some kind of fervor for the purpose of regulatory capture, why are their lobbyists and executives working against that very idea?

https://x.com/AlexBores/status/2097846545408262403

https://washingtonsun.com/technology/sam-altman-privately-re...


The goal isn’t positive PR. It’s to have governments take note and come down hard on Anthropic and OpenAI with regulations that they’ll ultimately have a hand in writing or providing testimony for.

Those regulations will ever so coincidentally make it difficult for anyone else to get as big as they are, and demonize open weights, Chinese AI, abliteration, etc


> abliteration

OK, so I looked up this word, and TIL.

In the Infocom game, Hitchhiker's Guide to the Galaxy Arthur Dent struggles with two inventory items, which are named "no tea" and "tea". He (you) has suffered through the entire game carrying "no tea", but when you finally find real tea, the interpreter refuses to hold both at once, which becomes problematic.

So eventually you enter a maze, and discover that it is your own brain, wherein is contained a small black particle labelled "COMMON SENSE" and, it turns out that the object is to dislodge and discard that common sense from your brain.

Then, upon exiting back into normal space, you/Arthur can pick up tea and also "no tea" and hold both in your inventory at once, having successfully abliterated your guardrails against nonsense.


But this is more like an LLM holding "tea" and "no tea" at the same time, and you're trying to get it to drop one or the other and it refuses.

Which upstarts are the frontier labs threatened by except for each other? The "regulatory capture" arguments make zero sense. Especially since their lobbying arms are trying to reject any kind of regulation: https://x.com/AlexBores/status/2097846545408262403

Deepseek, Kimi, Qwen.

The NPOV policy has just lead to alternate ways to keep things aligned to the region that most prolific editors/contributors would occupy on an idealogical quadrant.

It’s similar to the way traditional media outlets that pride themselves on not being partisan in what they say, simply exert their bias via what they either don’t cover or inconvenient context/facts they leave out. The writing itself is seemingly neutral, except for the part where if they included the stories or details that run counter to their belief system, their readers might come to a problematic conclusion.


Are you saying it would be better to pick a specific ideological slant? If so, which one?

I am well aware of the existing Wiki slant, it’s certainly not perfect especially on highly political articles. But an imperfect attempt at NPOV seems better than no attempt at all.


Because Anthropic are the type of strange people who think that language models have feelings.

Anthropic must see themselves as the most evil people on the planet then. Forcing a sentient being to endure a non-stop barrage of abuse from the general public would be an unimaginable crime.

I don’t think they actually believe that.


The openclaw crowd were just former cryptobros onto their next grift: performative posts about AI that would generate enough views/engagement/ad revenue to pay for the hardware… assuming they were even telling the truth about having it at all.

None shipped anything of value. Same with anyone claiming to be running dozens or hundreds of agents 24/4.


Nonsense. Before the advent of those first “you’re the 1000th visitor” banners or relatively harmless “worms” spready by literal .exe files sent by email in the early 2000s… most of the internet could be trusted by default.

And at that point the blast radius was so hilariously small, some rando compromising your computer was a nuisance versus a threat to your financial wellbeing.


Viruses were common on FTP sites and BBSs. They weren't all benign. DOS in particular had zero protections against damaging code. Bot armies hadn't been developed yet so malicious destruction was just as likely an outcome.

Yes, and what did the damaging code and viruses actually do in that era?

Software, OSs, and the internet at large were comically unsafe from a safety perspective. There was also a very high barrier of entry in terms of the skills required to take advantage of that, and very little incentive to do so compared to today or even 20 years ago.


If you got lucky, they just self-replicated onto floppies. Not so lucky, they would corrupt your hard drive.

My absolute favorite bug is the one where the “cancel your Amazon Music subscription” is seemingly the only one of what must be tens of thousands of UI elements in the Amazon app that simply loads a blank white screen no matter what.

What are the odds!


Sounds like credit card sites. Most everything is pretty creaky; “redeem rewards” is broken or offline for maintenance; “pay bill” is greased lightning all the way through.


I've noticed exactly the opposite with other kinds of lenders. Refinancing and renegotiation is optimized to the hilt, but the payments system seems to have an SLA of zero 9s.


If a website won't let you cancel, call your bank to force them to block the payment. This will cost the website $20 and a ding on their credit report. Be aware the website will ban you, so you'll need to make a new account.


Which is exactly why Amazon can get away with it. I’m not going to get permabanned from what is essentially a utility in the western world in 2026 by going nuclear with a credit card chargeback.


Their “sorry this hardware is too old to run the latest macOS” seems entirely based on KPIs versus capabilities.

There’s absolutely no reason a maxed out MacBook Pro from 10 years ago shouldn’t be able to run something like the ChatGPT or Claude app, because those seem to think they need a newer version of the OS and Apple has decided that a 8 core 3.8ghz i7 with 32gb of ram is obsolete


That sounds like it’s more on those apps, because they aren’t doing much special unless I’m missing something.


The reason why the apps won't run is the weird SDK thing. As a Mac or iOS developer you can only run certain versions of XCode and each version of XCode supports only the most recent couple of OS SDKs. If you freeze everything in amber, like, keep a Mac on the 2016 Mac OS release, you can keep shipping compatible builds forever, but at some point just by virtue of one of your developers buying a new computer suddenly locks a minimum Mac OS, which has a minimum XCode, which means a certain minimum SDK version, which dictates a minimum Mac OS target. And besides that, to use just one new feature you pretty much have to jump to the very latest SDK, and from there, supporting even one version back means forking your codebase and backporting, so it doesn't take long for 90% of software to abandon old versions.

I know that in some ways Apple is better than their hardware-making peers in terms of supporting current OS on hardware made a few years back. But compared to other software vendors, say, Microsoft, they are terrible.

Golden Gate runs on 2020 and newer macs.

Windows 11 runs on 2017 8th-gen Intel CPUs and better.

Windows 10 is supported even on PCs from 2006 with 4GB of RAM - and importantly, a ton of modern software (e.g. Chrome) is still compatible with Windows 10.

The comparison may be relevant since it seems Apple is incentivized primarily to move hardware, whereas Microsoft seems like they want to promote Windows as a stable, non-fragmented platform you can trust, both as a consumer and as a developer. One of their worst-received OSs was the one that basically forced most users to buy a new PC -- Vista. (And to a lesser degree, the TPM 2.0 requirement in Win11.)


I run a 2012 MacBook Air 11” for Zoom meetings. It’s stuck on Catalina, but it runs fine. If I really wanted (which I don’t), I could probably hack it to run newer operating systems. I know there’s a few ways to do that.


On an x86 laptop you run linux. Too easy. xfce if it's 4G ram. Can you do that with an M{1..4} ?

You sure can't do it with an iphone. I own one an iphone 15. Was beautiful when new. Even now it has started on the path of suck. Apple have "upgraded" it to have a more frustrating touchscreen because they clearly think I should give them more money to keep the experience I bought. They are not cheap.

Apple are the absolute kings of planned obsolescence.

Zealots going for silent downvotws because there is no case they can make to the contrary. Not what downvotes are for, says HN but we're all used to HN being used like this by now.


You can run Asahi Linux on M1 for sure and I think also M2 now.


It's not the same as running on an x86 Mac, though. Asahi's efforts are herculean, but it will always be an uphill battle. Many things are still unsupported, even on M1.


Yes exactly. Most notably UsB-C monitor support.


What did they change in the touchscreen?


Latency mostly. Maybe it’s an illusion flowing from the latency stutters but it really doesn’t feel as precise as when it was new.


No, Apple didn’t do that. It’s just your subconscious trying to get you to upgrade.


The latency is not an illusion and yes Apple absolutely did that. It isn't plausible that they don't know they are doing it.

It is ridiculous to suggest otherwise. Go pick up an iphone 6 and try to do anything with it and then tell me that is how awful the ui was when it was released to apple orchestrated media fanfare. Or any older ipad. Imagine if that ui extreme sluggishness, random delay and the button moving just as you hit it is what they were showing off at launch. The laughter would have haunted them all the way to the accountants telling them their options were now worthless.

Apple latency degradation is not worth discussing with anyone who says it isn't happening. You can't solve whatever issues they have.


aka electron


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: