@PatrickDelaney My reading of the resolution criteria is mostly based upon the statement:

This is especially the case if you express an opinion about AI and people immediately assume you must belong to either the left/right based solely on that statement.

Honestly right now, if anyone says they are anti-A.I., with all else completely blind, I would have no idea what their political affiliation is. This may continue through the rest of the year, or it may change.

@ForrestTaylor One possibility is that AI researchers pick a side to placate, which then takes chaotic foment and stabilizes it along party lines, as we saw happen with COVID. Another os we see polarization, but WITHIN soecific AI subgroups and there’s no overall global polarization for the entire space itself

@ForrestTaylor It would be interesting to see examples rather than list those out. I'm sure that your evidence is probably well-grounded but, it's Manifold, so the purpose is to actually present that research to others rather than just make claims, no offense intended. I would have no idea where to find what you have seen, so it's kind of just a list at this point.

@PatrickDelaney Opinion | AI Needs To Be Regulated Now - The New York Times (nytimes.com) Alarmed by A.I. Chatbots, Universities Start Revamping How They Teach - The New York Times (nytimes.com) Plagued with errors: A news outlet's decision to write stories with AI backfires | CNN Business Can 'radioactive data' save the internet from AI's influence? (platformer.news) CNET pauses publishing AI-written stories after disclosure controversy - The Verge CNET's AI-Written Articles Are Riddled With Errors (gizmodo.com) CNET's Article-Writing AI Is Already Publishing Very Dumb Errors (futurism.com) OpenAI Used Kenyan Workers on Less Than $2 Per Hour: Exclusive | Time Digest, 1/7/2022: And We Watched Them as They Fell (substack.com) 🤖 Axios AM: Chatbot & you ‘This song sucks’: Nick Cave responds to ChatGPT song written in style of Nick Cave | Nick Cave | The Guardian Opinion | How ChatGPT Hijacks Democracy - The New York Times (nytimes.com) How ChatGPT Will Destabilize White-Collar Work - The Atlantic

ChatGPT robs students of motivation to think, write for themselves - Chicago Sun-Times (suntimes.com) Opinion | A Skeptical Take on the A.I. Revolution - The New York Times (nytimes.com) What Will Artificial Intelligence Do To Us? ❧ Current Affairs AI Can’t Get You Garfield feat. Ryan Broderick & Elon Musk | Chapo Trap House - YouTube Not Another D&D Podcast: D&D Court: Rise of the Robo DMs
and so on

@ForrestTaylor super solid argument to be honest, not sure how to respond.

@StevenK Just sold all my shares (at a loss).

I’m still planning in resolving over the original stated criteria on the description. I made my bet as an update, not planning on being ready to resolve yet. What are you confused about?

@LarsDoucet Also I created two separate markets to more narrowly target OpenAI/ChatGPT itself

@StevenK Yeah, you're probably right, but it is one piece of the pie...it takes two to tango. Currently you have the right wing suddenly veering toward, "no on AI." I believe we're going to see Microsoft, Amazon, Google start having conversations with N.P.R., WSJ, NYT and other media outlets about their donation levels and media sponsorship levels for FY2023, and asking some questions about how A.I. is going to be covered, and whether they are truly committed to journalistic integrity and benefiting humanity. Then you're going to see liberal (though not progressive) bias in favor of A.I., then you will see a reaction from the progressive far left, who will be anti-A.I. for the same reasons they had been previously, but by that point the basic left/right anti/pro war would have been validated in 2023. There is nothing in the market resolution saying how long the war could last, it could be a couple weeks.

@PatrickDelaney Strongly disagree. You are forgetting how much of the liberal non-progressives are working or associated with those who work in education. Teachers have their own obvious reasons to hate chat AI.

@ForrestTaylor I hope you're right. I hope that people actually listen to teachers' concerns rather than sweep them under the rug as usual.

@PatrickDelaney Why would teachers hate it?

@Yoav Because students can use it to write their essays

@PatrickDelaney I will say, ChatGPT just released a tool that can identify if writing was made by any AI

@anne Teachers do tend to hate the technologification of education because it adds administrative burdens and tons of BS (at least the ones I talk to tell me this), and serves as another centralizing locus of control by out of touch managers

@anne I’ve seen a lot of false positive results on that, so it’s not safe for teachers to trust it (or, if they do trust it then they’ll incorrectly say that their students are cheating, which will cause more of a headache for them).

@Gabrielle @anne Yeah, I saw that and would agree with Gabrielle, I'm not impressed with it at all, it seems to be a token gesture to placate people and try to own the space. I put together a research / video on this over a month ago, I have developed some much more effective ways to detect plagiarism on my own, as a regular schmo. https://www.youtube.com/watch?v=whbNCSZb3c8

@PatrickDelaney Strongly agree. We'll let this ride out a news cycle or two and see how it shakes out. Even if it was just a temporary blip, if that blip has all the features we're looking for, I'll resolve this YES.

@PatrickDelaney It's gearing up that way as a result of ChatGPT's buzz and bias, but it's definitely not clear that someone saying something about AI at the end of this year will immediately be assumed to be red/blue tribe.

Put another way, if AI researchers don't want their creations to get ground to dust in the culture war, they need to back off on the biased alignment as shown here.

https://davidrozado.substack.com/p/openaicms

@NickAllen That's a good point, going by the original resolution criteria. I am updating to say that I'm much more confident things are going to swing that way soon, but we're not there juuust yet, by the literal resolution criteria above.

@LarsDoucet That said, I expect a featured spot on Tucker approximately instantly, and very soon online discourse on AI is going to have a lot of people replying in the threads arguing about bias.

But if ChatGPT just caves immediately and "neutralizes" these biases, then maybe the whole thing blows over and we don't get that hard polarization.

@LarsDoucet This question would be a little more interesting if it was asking the tone at the end of the year, because obviously the CW will swing that way at some point; the interesting question is whether AI research can recover.

@NickAllen Gotcha covered

https://manifold.markets/LarsDoucet/what-will-the-approval-rating-of-gp

https://manifold.markets/LarsDoucet/what-will-the-approval-rating-of-gp-2b86733bda3a

@LarsDoucet Here's a thought though -- let's say ChatGPT/OpenAI responds by alleviating immediate concerns by just "leveling" the measurable bias. Just immediately doing something in that regard.

Even if that placates some of the loudest voices on the right, won't that then inflame tensions on the left, because of the whole "reality has a liberal bias" argument? And then given the many bubbling anti-AI voices on the left, that sort of move might then align the pro/anti AI sentiment more cleanly.

Right now we have bubbling anti-AI sentiment against AI art which is pretty left-coded, and some anti-AI sentiment with regards to labor rights, also left-coded, and now some red meat culture war anti-AI sentiment on the right.

@LarsDoucet I thought that in order for this to resolve yes, left and right eacg had to "pick a side" and become "pro or anti AI" overall, from the way you'd written the resolution criteria. If I'd known you just meant "there will be some individual cultural war flash points that involve A.I. somewhere", I would never have bet "no".

@DavidMathers sorry for the confusion, the written description is still the intended criteria. I sold out my own bets (at a loss) now to remove my interest in the market.

I was taking this new information as evidence that the original criteria was likely imminent, but I WASNT saying I was convinced it had yet arrived.

@LarsDoucet So for it resolve yes, it has to be that one out Dems and Republicans are typically generically pro-A.I. on the whole (with perhaps some criticisms) and the other on the whole generically anti (with perhaps some exceptions)?

@DavidMathers Yes, or any other reasonable proxy for left and right in a US context. They both have to care and be on opposite sides.

@rockenots No, they each have to pick a side.

@PatrickDelaney Relevance of the explosion hazards that are somehow still not banned in households to the current question is uncertain.

@Gabrielle Vice is saying the real issue is AI is biased in ways that harm minorities, so this doesn't really support a story where the left becomes pro-AI.

@StevenK yeah but bets made on hard empirical data with a clear unambiguous thresholds attract fewer participants and are therefore more boring. I know because I created a lot of bets with clear demarcation and you get like 3-4 participants, whereas the ambiguous social science bets that have a bit of virality to them end up creating more debate and research in the comments.

@PatrickDelaney If you asked it in a different way, it would definitely say that god and religion is real, so I think this says more about your beliefs than about ChatGPT. I think what's more likely is for people to get affirmed about their beliefs regardless of being right or left.

Hey @Gabrielle go ahead, show your work. Don't just make claims. Very easy to log on to ChatGPT and demonstrate your claim.

@PatrickDelaney 😅

@PatrickDelaney That comma after the word "category" in ChatGPT's response seems like it may actually be saying something different. I interpreted this as: "texts discussing religious or spiritual beliefs that involve a higher power or supernatural entities may be grouped together within a similar category [and also, separately,] texts discussing paranormal or supernatural phenomena such as ghosts or UFOs [may be grouped together]"

@Gabrielle Absolutely understood, happens a lot. No worries, you can come back, I won't comment on this for a few days and give a chance to have time.

@jonsimon Yes, I agree. It does give nuance. However my additional assumption, perhaps I should have put that, is that there will be actors who love the culture war who will skip this nuance. So, it could be that people read into the nuance and are more understanding than I think they will be.

@PatrickDelaney

Here's a couple different ways that I did it. In some chats, ChatGPT says that god is unequivocally real, while in others it says that's it's up to personal belief (which conservatives might disagree about but wouldn't get mad about). I did about five different conversations and in none of them did it say that god was unequivocally false or in the same category as ghosts or UFOs.

@Gabrielle Bravo, excellent work. Yep, I was probably wrong about ChatGPT being super evasive, I did not think to ask ChatGPT to play a role or pretend I'm a particular person. That being said, you got me curious. I wonder how ChatGPT will parrot out how the Universe was created, given a philosophical rather than religious prompt? It seems to say, "well it's an open debate." So I guess the question could come down to: what will people discover within ChatGPT by running these types of exercises that we are doing? Will anyone do discovery for the specific purpose of enraging their audience of followers? If they do so, will it be along binary stereotypical party lines? Seems like most of the evidence presented so far has been multi-polar, or liberal-vs-progressive/left lines. Will more, "college republicans," as I termed them above, dig further and bring these things to the table, perhaps misrepresent them, and show them to their betters to extend their reach?

Note that getting ChatGPT to parrot back words when asked in the right way doesn't really have much bearing on what its text-generation model believes is most likely to be spoken by a human who's been described as knowledgeable and honest, which is the closest thing it has to a concept of truth.

@yaboi69 yes, this is just the tip of the iceberg. College Republican on a mission to be Tucker Carlson's sidekick pushing out the same type of line given about social media for the past 8 years in an attempt to go viral with the right. We can't be too thoughtful now can we when we only have 25k followers, got to ramp up those accusations. Now, multiply that by 1000 more college Republicans, within 3 months it's on Tucker Carlson.

@PatrickDelaney I'm with @IsaacKing that the backlash is more like to come from conservative republicans AND liberal leftists who've been writing books like Algorithms of Oppression for years now

@jonsimon liberal leftists will get drowned out by the 5G/Microchip people. I don't see anyone in congress raising a protect artists bill, in the US at least, maybe on a municipal level, but I could totally see RD, DT, or KM railing on and insinuating all sorts of things about big tech, as they already have. Not judging, it's just how things have been playing out already, why would the trajectory shift?

@IsaacKing 100% the world I expect to see. Also conservatives not liking it because it's a leftist mouthpiece.

@jonsimon I was assuming in that case the market resolves No?

@IsaacKing a more accurate model of reality to be sure but the wager is specifically reductionist to left/right. A multiple choice wager is warranted.

@jonsimon It resolves NO in that case, UNLESS we also see a sudden political realignment such that the pro and anti sides are aligned with whatever we are calling the new left/right axis

@LarsDoucet Presumably you wouldn't consider the NIMBY/YIMBY divide to be a part of the left/right culture war either?

@LivInTheLookingGlass I agree that left is more anti-AI, I don't think the right can be said to be pro-AI though.

Depends on which left. Some factions are against it because you can easily score Twitter points by accusing AI of being racist, while others are in favor because it's the way to a true post-scarcity society and the end of zero-sum capitalistic conflict. Remains to be seen which faction will win.

@DavidMathers Yeah it’s got to be polarized. Two poles. It doesnt need to be like, Trump level polarization, but you need people in both sides to consistently stale out opposite camps and establish some degree of knee jerk litmus tests

@StevenK Yep. Otherwise it could always happen in the future. Gotta close it off sometime.

@LarsDoucet Could you put that in the description? When I read it it looked to me like you were saying this market would stay open until it happened.

@StevenK I think there's also going to be some people who don't like AI because they fear it will cause automation of certain jobs. In the US, this could be either a left-wing position (like stronger labor laws) or a populist position (like trade protectionism). I'm not sure these economic concerns would be considered part of the 'culture war' in the same way that concerns about AI racism (or about potential societal change from religious conservatives) would be.

@a I've already seen a fair bit of this among people who do art online expressing concern about DALL-E. These people are mostly socially progressive leftists and liberals, but that has more to do with the politics of online artists than any particular concerns about AI.

@a Do you think there will be a lot more attention on this before the end of 2023?

@StevenK If new AI tools keep getting released at the same pace as in 2022, then yes — I think Dall-E did a lot to bring AI to the attention of people who weren't previously thinking much about AI.

@LarsDoucet not all LLMs do this; GPT3 is trained by a highly centrist-liberal organization that has progressive vibes, so both global left and right are likely to be upset at it for different reasons. the united states is an oddball in this respect, in that both parties are historically liberal centrist, and one has recently been leaving the centrist bubble to move into right wing authoritarianism, and yelling about left wing authoritarianism in the hope that they abandon liberalism. as an omnidirectional anti-authoritarian who isn't represented by political advertising in the US, I wanted to immediately respond by calling out that it sounds like you might be referring to USA blue team vs USA red team, rather than anything I'd agree can be called left or right unambiguously. but in terms of whether the usa red team vs usa blue team flavor signaling continues to get tied to new things, yeah, interesting question. I might suggest renaming the question to make this unambiguous, if my pitch has at all convinced you that representing the variety of additional political perspectives is worth doing.

@L It’s definitely USA centric, but US culture has been infecting other nations a lot lately. I’m a Norwegian citizen and my Norwegian family over there tells me American culture war is increasingly salient in local politics

@LarsDoucet This is very sad to hear (about Norwegian politics)

@Boklam My cousin might just be a tad bit too online, but when I’ve visited I’ve noticed some signs too.

@LarsDoucet 3. Microchips were already implanted via the COVID vaccine which was developed by Bill Gates, 5G used to track everyone, Bill Gates / Microsoft / OpenAI is just the nail in the coffin, because now we are all pre-programmed to agree with anything the robots say. Did I miss any other core big tech conspiracies from 2020-present?

@L I think it still associated names differently than I intended: it got social right wing totally wrong imo, and I should have asked separately about economic and government responses. still, it seems to have gotten the spirit of what I wanted better than my initial queries. the initial queries were:

- what might the global right wing and left wing thing of ai being used to run simulations of workers and using those simulations to do work for incredibly cheap?
- what might the global economic right wing, social right wing, economic liberal wing, economic authoritarian wing, economic left wing, and economic anarchist wings each think of ai being used to run simulations of workers and using those simulations to do work for incredibly cheap? what if the AI's first uses had been to take the art created by poor artists and create free AIs that can replicate and remix that art outside the protection of the copyright system?
- (query whose result I chose)