HarkMahlberg

HarkMahlberg , 2 months ago

I mean... they didn't specify it had to be random (or even uniform)? But yeah, it's a good showcase of how GPT acquired the same biases as people, from people..

HarkMahlberg , 3 months ago

Which is still weird.

Alexander Sawchuk, then an assistant professor of electrical engineering at the University of Southern California ... along with a graduate student and the SIPI lab manager, was hurriedly searching the lab for a good image to scan for a colleague's conference paper. ... Just then, somebody happened to walk in with a recent issue of Playboy. The engineers tore away the top third of the centerfold so they could wrap it around the drum of their Muirhead wirephoto scanner...

https://en.wikipedia.org/wiki/Lenna

Everything about the story sounds like it was a rush job, a decision made on a whim, after exhausting their existing catalog of test images. And who bring a Playboy mag to their university's computer lab, and advertises their possession? They don't even say who it was, probably to protect them from any embarrassing professional consequences. To me, that's probably the strongest reason to retire it: it's unprofessional.

HarkMahlberg , 3 months ago

Well said.

HarkMahlberg , 3 months ago

I would say that the "positive vibes only" trait is part of it, but the far bigger problem was the character limit. Even when it was double from 140 to 280, that still doesn't not leave room for nuanced opinions. And then, the least nuanced opinions also become the most easily spreadable. Both traits really reward our worst instincts.

HarkMahlberg , 3 months ago

Easiest block of my life.

HarkMahlberg , 3 months ago

Wouldn't this enable, for example, Trump claiming he didn't make the "bloodbath" comment, calling it a deepfake, and telling Youtube to remove all the new coverage of it? I mean, more generally, what stops someone from abusing this system?

HarkMahlberg , 3 months ago (edited 3 months ago)

The author does have a way with words lol. I love this paragraph in particular, emphasis mine:

As we speak, the battle that platforms are fighting is against generative spam, a cartoonish and obvious threat of outright nonsense, meaningless chum that can and should (and likely will) be stopped. In the process, they're failing to see that this isn't a war against spam, but a war against crap, and the overall normalization and intellectual numbing that comes when content is created to please algorithms and provide a minimum viable product for consumers. Google's "useless" results problem isn't one borne of content that has no meaning, but of content that only sort of helps, that is the "right" result but doesn't actually provide any real thought behind it, like the endless "how to fix error code X" results full of well-meaning and plausibly helpful content that doesn't really help at all.

And he describes exactly what I have to deal with on the regular, "content that only sort of helps" that "steals your attention from the content you actually want." Even moving from Google to DDG has only mitigated this problem, it hasn't fully gone away.

But yeah, one of his conclusions seems to be the Death of the Hyperlink? Which, I mean, not even LLM's can kill that. I doubt <a href is going away any time soon.

HarkMahlberg , 3 months ago

The author pretty freely admits he shares some blame, having PII on the same phone he uses Lemmy, using Lemmy while not paying attention/being half asleep. I'm sure he does know better and agrees with your statement. And yet, when mistakes happen and people prove to be fallible, Lemmy proves it is not capable of handling the problem.

I also can't believe the Lemmy developers would be so indignant about being presented with such an oversight. GDPR or no GDPR, federated to other servers or not, the idea of PII being hard/impossible to delete from a social media platform is an embarrassment to the developers.

HarkMahlberg , 3 months ago (edited 3 months ago)

I'm kind of impressed by the amount of research they did to figure out why this guy's bill was so high, then immediately offered a resolution, and then immediately offered another avenue if the resolution wasn't good enough. Shout out to the customer service rep.

HarkMahlberg , 4 months ago

Ich bin ein kbinner?

HarkMahlberg , 4 months ago

There are far more important facets to truthfulness and semantics than yes/no questions. If this is the only way you evaluate LLM's, you will quickly fall for confirmation bias.

HarkMahlberg , 4 months ago

No

HarkMahlberg , 4 months ago

The car or the account?

HarkMahlberg , 4 months ago

Por que no las dos?

HarkMahlberg , 4 months ago

This is the same criticism that was made of cryptocurrency's claim to fame regarding decentralization, consensus, and resilience to authoritarian takeover.

"If you take all these different parts of your identity, all the games you play, all the things you buy, all the groups you join, and stick them into one system, that's a central system. It doesn't matter how many servers that system spans, you've pooled all that data in one place."

And ultimately we can make the same criticism of the Fediverse itself. It's nice that there are different platforms, different instances, different communities... but it's still just one entity at the end of the day. This is especially apparent with the spam wave we just saw. Misskey, Mastodon, Lemmy, even kbin was not invulnerable. You don't need to attack them individually, you can attack them all at once, and then they will naturally spread your attack to other instances for you.

HarkMahlberg , 4 months ago

Venture Capital

HarkMahlberg , 4 months ago

None? I don't debate that Blue Sky is corporate-owned while Bitcoin and the Fediverse aren't. Rather, I'm saying the thing they all have in common is that they like to think of themselves as "decentralized" federations of independent systems and users, but in reality they are all "centralized" systems with shared weaknesses. This is the "ideological contradiction" I thought you were referring to.

HarkMahlberg , 4 months ago

Super bowl ads don't mean anything. The crypto world also ran ads before/at the super bowl, look how they're doing now.

Air Canada had to shut down their AI chatbot because it was making up refund policies and costing them money.

New Hampshire is pissed that someone used an AI Biden voice to spread voter disinformation.

If it's not already there, AI is quickly approaching the FO phase of FAFO.

HarkMahlberg , 4 months ago

Poor Duolingo. Once upon a time I used it to learn Japanese, but by the time I could start reading kanji and noticed that duolingo was still constructing sentences entirely out of hiragana, I knew I had outgrown it and moved on to Anki.

Using AI to learn a new language has to be incredibly frustrating - you can either tell where's messing up, or you can't tell at all and then you learn incorrect information..

HarkMahlberg , 4 months ago

Almost makes me nostalgic for the way clothing used to work in Cyberpunk 2077.

HarkMahlberg , 4 months ago

"We have Fandango at home."

Fandango At Home at home:

HarkMahlberg , 4 months ago

Yeah that's the guy. Hilarious to see he thinks his garbage biased opinion is worth any amount of money.

HarkMahlberg , 4 months ago

The value-add is the comedy of a man pretending an Intel Q6600 is better than a Ryzen 3600X.

HarkMahlberg , 4 months ago

I love my Grados for that exact reason

HarkMahlberg , 4 months ago

The irony of this being crossposted from ML.

HarkMahlberg , 4 months ago

Oh I'm not giving you grief, I just think it's funny.

the article is on a CO site

🤌

HarkMahlberg , 4 months ago

For what it's worth, I guess they saw this coming.

HarkMahlberg , 4 months ago

How the tables have turned.

Sony won the case against Universal that allowed people to record TV shows with their VCR. I wonder how they'd feel if I pointed OBS at their streams.

HarkMahlberg , 4 months ago

Probably not without some tinkering, but DRM can always be defeated.

HarkMahlberg , 4 months ago

What a catchy name.

HarkMahlberg , 4 months ago

Let's not confuse ourselves here. The opposite of one evil is not necessarily a good. Police reviewing their own footage, investigating themselves: bad. Unreliable AI beholden to corporate interests and shareholders: also bad.

HarkMahlberg , 4 months ago

It’s fine to not understand what “AI” is and how it works

That's highly presumptive isn't it? I didn't make any statement about what AI is, or the mechanics behind it. I only made a statement regarding the owners and operators of AI. We're talking about the politics of using AI to aid in police accountability, and for those intents and purposes, AI need not be more than a black box. We could call it a sentient jar of kidney beans for all it matters.

So for the sake of argument - the one I made, not the one I didn't make - what did I misunderstand?

Unreliable

On June 22, 2023, Judge P. Kevin Castel of the Southern District of New York released a lengthy order sanctioning two attorneys for submitting a brief drafted by ChatGPT. Judge Castel reprimanded the attorneys, explaining that while “there is nothing inherently improper about using a reliable artificial intelligence tool for assistance,” the attorneys “abandoned their responsibilities” by submitting a brief littered with fake judicial opinions, quotes, and citations.

Judge Castel’s opinion offers a detailed analysis of one such opinion, Varghese v. China Southern Airlines Co., Ltd., 925 F.3d 1339 (11th Cir. 2019), which the sanctioned lawyers produced to the Court. The Varghese decision is presented as being issued by three Eleventh Circuit judges. While according to Judge Castel’s opinion the decision “shows stylistic and reasoning flaws that do not generally appear in decisions issued by the United States Court of Appeals,” and contains a legal analysis that is otherwise “gibberish,” it does in fact reference some real cases. Additionally, when confronted with the question of whether the case is real, the AI platform itself doubles down, explaining that the case “does indeed exist and can be found on legal research databases such as Westlaw and LexisNexis.”

https://www.natlawreview.com/article/artificially-unintelligent-attorneys-sanctioned-misuse-chatgpt

Regardless of how ChatGPT made this error, be it "hallucination" or otherwise, I would submit this as exhibit A that AI, at least currently, is not reliable enough to do legal analysis.

Beholden to corporate interests

Most of the large, large language models are owned and run by huge corporations: OpenAI's ChatGPT, Google's Bard, Microsoft's Copilot, etc. It is already almost impossible to hold these organizations accountable for their misdeeds, so how can we trust their creations to police the police?

The naive "at-best" scenario is that AI trained to identify unjustified police shootings sometimes fails to identify them properly. Some go unreported. Or perhaps it reports a "justified" police shooting (I am not here to debate that definition but let's say they occur) as unjustified, which gums up other investigation efforts.

The more conspiratorial "at-worst" scenario is that a company with a pro-cop/thin-blue-line sympathizing culture could easily sweep damning reports made by their AI under the rug, which facilitates aggressive police behavior under the guise of "monitoring" it.

As reported by ProPublica, Patterson PD has a contract with a Chicago-based software company called Truleo to examine audio from bodycam videos to identify problematic behavior by officers. The company charges around $50,000 per year for flagging several types of behaviors, such as when officers use force, interrupt civilians, use profanities, or turn off their cameras while on active duty. The company claims that its data shows such behaviors often lead to violent escalation.

How does Truleo determine what is "risky" behavior, what is an "interruption" to a civilian? What is a profanity? Does Truleo consider "crap" to be a profanity? More importantly, what if you disagree with Truleo's definitions? What recourse do you have against a company that has zero duty to protect you? If you file a lawsuit alleging officer misconduct, can Truleo's AI's conclusions be admissible as evidence, and can it be used against you?

(1/2)

HarkMahlberg , 4 months ago

And shareholders

He couldn’t have imagined the drama of this week, with four directors on OpenAI’s nonprofit board unexpectedly firing him as CEO and removing the company’s president as chairman of the board. But the bylaws Altman and his cofounders initially established and a restructuring in 2019 that opened the door to billions of dollars in investment from Microsoft gave a handful of people with no financial stake in the company the power to upend the project on a whim.

https://www.wired.com/story/openai-bizarre-structure-4-people-the-power-to-fire-sam-altman/

Oh! Turns out I was wrong... "a handful of people with no financial stake in the company" doesn't sound like shareholders, and yet they could change the direction of the company at will. And just so we're clear, whether it's four faceless ghouls or Sam Altman, 1 or 4, the fact that the company is beholden to so few people, who themselves are not democratically elected, nor necessarily law experts, nor necessarily have any history being police officers... their AI is what decides whether or not to hold a police officer accountable for his misdeeds? Hard. Pass.

Oh, and lest we forget Microsoft is invested in OpenAI, and OpenAI has a quasi-profit-driven structure. Those 4 board directors aren't even my biggest concern with that arrangement.

(2/2)

HarkMahlberg , 4 months ago

Man, you said everything I wanted to in less than half the words. Shoulda just linked to your comment lol

HarkMahlberg , 4 months ago

All those racks of hard drives are taking up the space they need for racks of Nvidia GPU's.

HarkMahlberg , 5 months ago

May also indicate that users were shopping around for a blocker that worked against Youtube. Maybe some of those users actually just settled with AdGuard coming from ABP, or uBlock, or whoever.

HarkMahlberg , 5 months ago

Seen plenty of people talking about the crazy ads they see on Youtube. Right wing propaganda, blatant grifting, scams... Folding Ideas has done not one but two videos talking about the ads he saw and picking them apart. Surely the people complaining about these ads know adblockers exist right? Why don't they use them? I'm sure there are several reasons but, it's been a known quantity for decades that you have the power to control how many and what kind of ads you see.

HarkMahlberg , 5 months ago

I moved from Vivaldi to Firefox during the crackdown, signed out all of my Google accounts, and immediately noticed the problems went away. Sorry Vivaldi...

HarkMahlberg , 5 months ago

Always happy to see Simon Stalenhag's work lol

HarkMahlberg , 5 months ago

my android phone, which I’ve paid off completely

I think this is about where I realized your anxiety has more to do with your financial situation than your technology situation. Your worries are about the way you spend money, how much you spend, what you spend it on, and how corps try to part you from your money. Like another commenter said, all the free and open technology in the world isn't going to magically balance your checkbook... though of course, it will help!

Yeah installing Ubuntu is great, learning to code is great, these are valuable endeavors if for no other reason than just to learn and try new things, but you don't need to learn programming to "convert your chromebook to Ubuntu."

I have no idea what to do about Amazon or Amazon Prime. ... things that, in a small town with a particular disability keeping me from driving, I can only get on Amazon.

If using Amazon is unavoidable, then it is what it is. There's no shame in using them to get what you need. If you're concerned about, say, your habit of impulse buying (not an accusation, just an example), you could try setting up a secured credit card with a spending limit so you can only use it for exactly what you need.

death consciousness of mindlessly scrolling through Facebook

Block Facebook in your router settings (or get a Raspberry Pi, install Pihole, and set up a block rule there). If you need Facebook to communicate with friends and family, could you rely solely on Messenger? That way you don't need to see anything on Facebook other than your DM's.

If your mental health is dire enough that all that's not enough, you probably need a therapist. You can even get it through some tele-health programs (YMMV). Hope this helps!

HarkMahlberg , 5 months ago

Sounds like a product they're gonna kill off soon.

HarkMahlberg , 5 months ago

But Bard will also analyze the private content of messages “to understand the context of your conversations, your tone, and your interests.” It will analyze the sentiment of your messages, “to tailor its responses to your mood and vibe.” And it will “analyze your message history with different contacts to understand your relationship dynamics… to personalize responses based on who you're talking to.”

Reading between the lines: they want to use your private messages to train their AI.

HarkMahlberg , 5 months ago

I wish I could find the article, but when Musk first started breaking shit and locking everything down, local meteorological accounts realized people could start missing important public information like tsunami and earthquake warnings, and they had no other way to reach the public than through Twitter.

Twitter being accessible only via direct links to tweets is still not an acceptable solution, because how would I know what the URL is for the latest Icelandic volcano warning (for example)?

HarkMahlberg , 5 months ago

I wish network effects weren't a self fulfilling prophecy.

HarkMahlberg , 5 months ago (edited 5 months ago)

Yikes.

Your comment is entirely fact-free, it's almost impressive. There's no point refuting anything because it's clear nothing will change your mind, and I don't have to anyway because it's all been dismantled piece by piece already.

https://youtube.com/watch?v=YQ_xWvX1n9g

https://vivaldi.com/blog/why-vivaldi-will-never-create-thinkcoin

I know it's hard to admit you've gambled big on a losing horse, but the solution is not to gamble bigger, it's to quit gambling.

Edited links for clarity.

HarkMahlberg , 5 months ago

Of course he does, he quotes Marcus Aurelius in his profile. May as well just block him and move on.