General Politics Three: But what is left/right?

[uppi]

Are there any situations where the right answer to "I want this bit of accurate information" is "Ask a generative AI large language model"?

There are none, but that does not stop people.

This is very likely an accurate statement. I'm not saying there aren't any, but voting districts are not generally commensurate with ZIP codes.

It could conceivably be misleading information, but to avoid that, one would positively have to program the chatbots to correct the human user: "You said zip code, but you're asking about voting; did you mean voting district? Here are the precincts within the 19121 zip code."

Google does that if one misspells things, so it could conceivably be done. But the programmers would have to take on themselves to anticipate such errors.

LLMs are not programmed in the traditional sense. The underlying software is, of course, but the output depends on what the model has learned from the training data. It gets the start of a conversation and then tries to predict how it will end. So it is very tricky to get it to answer questions factually correct. You can of course fine tune it for election-related questions, but only those you can think of.

 

[Gori the Grey]

LLMs are not programmed in the traditional sense.

Right, but on top of their base operation, the developers can impose constraints. Although nudes are among the commonest forms in Western art, DALL-E won't paint you one. Developers maintain a second-order control over the output of the programs, once they discover that they do or do not provide a certain kind of output. That kind of thing would have to happen here. Oh, our programs can give correct but misleading answers to questions about voting. We (as a society) don't want that. So let's intentionally direct how the program responds to certain kinds of queries.

Most of what I'm saying is that the article misrepresents the nature of the problem. Zip code 19121 hasn't been singled out because of it's racial composition. Chat just doesn't know what a person is really asking for when they put in a particular query. Or more precisely, what society wants as the right kind of answer to particular kinds of queries (namely, answers that don't discourage people from voting).

tl;dr, the answer to Samson's question is "no"

 

[Lexicus]

I suspect the number of people depending on Chat GPT to find out where to vote is close to zero

 

[Kyriakos]

I suspect the number of people depending on Chat GPT to find out where to vote is close to zero

Why rely on suspicions, when you could ask Chat GPT?

 

[Gorbles]

I suspect the number of people depending on Chat GPT to find out where to vote is close to zero

I have no evidence on voting, but man people are using it at work for way too many things. Mainly technical, which is where I guess the answers might be more accurate (as it just crawls known snippets online and relays them, or an abstract), but again, folks are using it almost as an alternative to Google here. It's depressing (because, like . . . just search for it), but the convenience wins nearly every time.

 

[Kyriakos]

While officially ChatGPT doesn't keep logs of your discussions, it is pretty obvious those are used to train it for free.

 

[Birdjaguar]

I don't think it's accurate, I think it's censorship.

Who has been censored? Certainly not any of the hate speakers on conservative talk radio over the past 30 years. their goal ha been to demonize anyone who thinks differently and keep the country divided into us versus them groups.

 

[Gori the Grey]

I suspect the number of people depending on Chat GPT to find out where to vote is close to zero

Mostly I'm upset by how the article is framing the matter:

Chatbots' inaccurate, misleading responses about U.S. elections threaten to keep voters from polls

That title, plus the specific zip code on which the article focuses, is trying to generate an anxiety: OMG Black voters are going to be disenfranchised by Chat GPT. Everything in the article is technically correct as far as it goes, but the anxiety that the article is trying to generate (for clicks, of course) is unwarranted. For a number of reasons. First, as you say, most people won't use Chat GPT to find their polling place. Second, any who do won't put in the query in the dumb way the article does. Third, any who do that and get the stated results, won't just stop there: "Oh, herp derp, I guess people in my zip code don't get to vote." Fourth, Chat would return the exact same kind of result about a majority white district if asked this same stupid question.

The sense of crisis that the article is trying to generate is unwarranted, even while every assertion in it is technically correct. That kind of thing in our modern media (or anywhere, I guess) pisses me off and I have a compulsion to call it out.

 

[Core Imposter]

We have situations where lawyers use it and it spits out citations of cases that do not exist.
One model was taken offline when it refused to generate images of white people, leading to black popes, black Vikings and black American patriots.
We had the situation yesterday where an author found that his book critizing Big Tech had run into this:

Gemini spat out summaries of four negative reviews, ostensibly from the Washington Free Beacon, The New York Times, The New York Times Book Review and Wired, including quotes from each:

• "Book review by Matthew Continetti in The Washington Free Beacon: This review criticizes the book for relying on ‘anecdotal evidence’ and ‘cherry-picking examples’ to support its claims.
• Book review by Emily Bazelon in The New York Times Book Review: This review notes that the book 'lacks a deep understanding of how these companies work' and relies on 'unproven accusations.'
• Article by Emily Dreyfuss in Wired: This article explores the ‘dearth of evidence’ for the book's claims about Big Tech bias and highlights examples of cherry-picking.
• Article by Ben Smith in The New York Times: This article discusses the book's lack of rigorous research and its reliance on 'unsubstantiated claims.'"

But here’s the catch. None of those reviews were real. Not one. And neither were any of the quotes.

I believe it is fair to say that it's going to be a problem.

who is birdjaguar at civfanatics?

BirdJaguar is a prominent member of the CivFanatics community, known for their contributions and engagement related to the Sid Meier’s Civilization series. Here are some notable aspects associated with BirdJaguar:

1. Chiyu’s Extra Mod: BirdJaguar created and maintains the Chiyu’s Extra Mod, which combines various material from official Firaxis mods (including Civ4 - Colonization) into one comprehensive package. The mod enhances gameplay by integrating additional content and features1.
2. Active Participation: BirdJaguar actively participates in discussions, shares insights, and collaborates with other CivFanatics members. Their contributions span various topics related to Civilization games, including strategy, modding, and historical context.
3. Avatar Inspiration: BirdJaguar’s username draws inspiration from history. In 629, Bird Jaguar III was enthroned as the king of Yaxchilan, an ancient Maya city. Yaxchilan is known for its impressive architecture, including the Santiago Temple and intricate glyphic relief panels. BirdJaguar III’s reign left a mark on the region, and their legacy continues to inspire players and historians alike23.

In summary, BirdJaguar is an active and respected member of the CivFanatics community, contributing to the enjoyment and depth of the Civilization games.

who is core impostor at civfanatics?

Core Impostor is an active member of the CivFanatics community, known for their contributions and engagement related to the Sid Meier’s Civilization series. Here are some notable aspects associated with Core Impostor:

1. Chiyu’s Extra Mod: Core Impostor has been involved in the creation and maintenance of the Chiyu’s Extra Mod. This mod combines various materials from official Firaxis mods (including Civ4 - Colonization) into one comprehensive package. The mod enhances gameplay by integrating additional content and features1.
2. Thoughtful Discussions: Core Impostor actively participates in discussions on the CivFanatics forums. Their contributions span various topics related to Civilization games, including strategy, game mechanics, and historical context. They engage with other community members, share insights, and contribute to the overall knowledge base.
3. In-Depth Analysis: Core Impostor often provides thoughtful analysis and perspectives on game-related matters. Whether discussing game balance, historical accuracy, or modding techniques, their posts contribute to the depth of understanding within the community

I certainly enjoyed working with Birdie on Chiyu's Mod, whatever that is.

 

[Gori the Grey]

Did Chat add the smilies at the end of the Bird treatment? Nice touch.

 

[Core Imposter]

Yes, the smilies for the Bird and on the CI response it added a footnote to a thread I created that had no responses.

 

[Core Imposter]

Gonna be a lot of fun when law enforcement starts to use the AI to solve crimes.

 

[Ziggy Stardust]

In 2016 the GOP introduced Alternative Facts when describing the size of the crowd on Trump's inaugeration.
Alternative Facts are the exact opposite of Facts.

In 2016 the GOP also has introduced Alternative Intelligence, and called it MAGA.
Alternative Intelligence is the exact opposite of Intelligence.

 

[Samson]

I believe it is fair to say that it's going to be a problem.

Why? It seems to me that it is so obviously bad that people will soon get over the thing that well written text must be right. It much the same way that photoshopped images may have confused some people for a short time, but as soon as it became clear what was possible everyone got used to their existence and it was never really a problem.

 

[Lexicus]

Gonna be a lot of fun when law enforcement starts to use the AI to solve crimes.

Yeah, this actually needs to be made illegal ASAP

 

[bernie14]

Who has been censored? Certainly not any of the hate speakers on conservative talk radio over the past 30 years. their goal ha been to demonize anyone who thinks differently and keep the country divided into us versus them groups.

We could "both sides this", to high heaven given that "existential threat orange man" is, as @Lexicus likes to say, literally a demon!

 

[Ziggy Stardust]

We could "both sides this", to high heaven given that "existential threat orange man" is, as @Lexicus likes to say, literally a demon!

You forgot to answer who is censored.

 

[bernie14]

You forgot to answer who is censored.

I wasn't thinking about it in that way, I was thinking about those who want to ban speech.

 

[Core Imposter]

I wasn't thinking about it in that way, I was thinking about those who want to ban speech.

Therein lies the rub. There exists no gatekeepers anymore. Anything you read, hear or see can be true, or not.

Is lying a crime? Not in most contexts.

What is truth? As a Christian I have truths that might give James Carville a stroke. Truth is elastic. Facts are propositions. Logic is customizable. Prophecy is of no private interpretation but... I have a speculation about the number of the beast- 666. Perhaps soon, in order to eliminate confusion, we will all be issued a number, a prefix. a Christian might be a 7, thus identifying her truth, might be a 72, indicating a Christian using the pronouns she, her, perhaps or perhaps not. We get to choose the truth, our truth, my truth. But oh, the confusion. We need a quick way to understand people. Perhaps a command level male atheist would have the prefix 666? Just speculation.

We don't want to offend anybody, why not tattoo a code on everyone's forehead containing pertinent information? People have tattoo and numbers already. Oh wait, all we need to do is wear the meta headsets and all that info would be virtually displayed.

 

[Lexicus]

We could "both sides this", to high heaven given that "existential threat orange man" is, as @Lexicus likes to say, literally a demon!

Both sides? And all you've got is my posts?? Rush had his eponymous radio show before I was born dude

I wasn't thinking about it in that way, I was thinking about those who want to ban speech.

You mean like how the government has banned teachers in your state from talking about LGBTQ anything? Give us all a break; like any right-winger, your BS about "censorship" is just a rhetorical move to sidestep defending the substance of speech to which libs object. Like exactly what you did in this thread; Birdjaguar criticized the substance of Rush's speech and rather than even try to defend it you simply brought up censorship as a red herring. And you just continue the same pattern of deflection by mentioning my name.