4 ms·
> the uniquely American animosity toward artificial intelligence The poll they cite shows this is clearly not unique: https://www.pewresearch.org/global/2025/1
by burkaman 10mo ago
> the uniquely American animosity toward artificial intelligence
The poll they cite shows this is clearly not unique: https://www.pewresearch.org/global/2025/10/15/concern-and-excitement-about-ai/ https://www.pewresearch.org/global/2025/10/15/concern-and-ex.... The US is (just barely) at the top, but no country is anywhere close to "more excited than concerned", and several countries are basically equal to the US.
I live in America so that's my perspective, but I would be surprised if this article couldn't accurately describe a lot of other countries.
- psunavy03 10mo agoEspecially lately, it seems to be fashionable to dump on Americans as somehow odd, weird, and afraid of things we supposedly shouldn't be. So much for "tolerance."
- kiliantics 10mo agoI'm a European living in the US and my sense is that perceptions towards AI are generally more positive here than in Europe if anything (I do work in tech though which skews things a good bit). This article almost feels like some kind of psychological manipulation: "Jeez Americans, can't you just get on board like the rest of the world?"
- frantathefranta 10mo agoThis is obviously anecdata but I was back home in Central Europe over Christmas and a staggering amount of people use ChatGPT (in public). Most usage I've seen was on public transit and in restaurants. My mom has replaced her Google usage with ChatGPT. Meanwhile in the US, my friend group makes it a point of pride to not use "AI" for anything, and 90% are not in tech. I had a feeling that us Americans are being forced to use Copilot/Gemini/whatever more than Europeans and have slightly more animosity towards it.
- mingus88 10mo agoIt’s economic anxiety. The average American wants jobs to come back (look at the election data), and seeing AI shoehorned into every service is not an indicator that industry is going to start hiring lower level positions anytime soon. The EU has strong worker protections and a robust social safety net. It’s not surprising to hear they are less antagonistic towards AI
- Yizahi 10mo agoEU has the same problems, and those 1-3 months of notice periods aren't helping that much when the whole world is in recession. Whole segments of IT are getting selectively laid off, like QAs for example, while LLMs are being shoved in everywhere.
- thefz 10mo agoIt's not even that, everyone feels like it does not work well and its results are wonky/unreliable. I live in central Europe, for what it's worth.
- Yizahi 10mo agoOne can use LLMs and at the same time dislike them and fear the long term consequences of mass proliferation. I sometimes use them, either to answer a multi-item question or to generate three paragraphs of business-speak spam if that is the only way. I don't like either of these things, spam especially. But even when LLMs are genuinely useful, it's only because normal search engines has failed me. Had they been more powerful, I wouldn't need to ask a random character generator.
- htfy96 10mo agoNote that most countries in the list are developed countries. In a separate research at https://www.bloomberg.com/news/articles/2025-06-20/trust-in-ai-strongest-in-china-developing-countries-un-study https://www.bloomberg.com/news/articles/2025-06-20/trust-in-... , people found that low income countries have higher trust in AI.
- daveguy 10mo agoAnd presumably much less experience with it.
- nitwit005 10mo agoMy assumption was Americans have received the most exposure to rushed out products from startups.
- armchairhacker 10mo agoAlso SK and, to a lesser extent, Japan (and Germany?) I wonder about China. More generally, do countries deemed collectivist (https://worldpopulationreview.com/country-rankings/collectivist-countries https://worldpopulationreview.com/country-rankings/collectiv...) and supportive of their government tend to lean towards AI?
- htfy96 10mo agohttps://hai.stanford.edu/ai-index/2025-ai-index-report https://hai.stanford.edu/ai-index/2025-ai-index-report > In countries like China (83%), Indonesia (80%), and Thailand (77%), strong majorities see AI products and services as more beneficial than harmful. In contrast, optimism remains far lower in places like Canada (40%), the United States (39%), and the Netherlands (36%).
- armchairhacker 10mo agoThanks. Interestingly, the report claims that AI optimism is rising (even the US gained 4%), but it only goes to 2024.
- beardyw 10mo agoBecause "Why do people hate AI" confirms you are in the majority. This title suggests that it is uniquely american and misguided.
- UncleMeat 10mo agoNYT is becoming trash simping for billionaires. They are participating in a broad movement within media to insist that an AI revolution that completely captures labor within the power of a few major companies is inevitable and we just all need to get on board.
- abanana 10mo ago> uniquely American Every so often, local news websites in the UK publish articles highlighting behaviours that are supposedly uniquely characteristic of the area they're writing about, with headlines such as "Only in $city!" or "You'll know all about this if you come from $city!". Across the publications, they're all substantially the same normal human behaviours, just with $city and any mention of local landmarks changed, and photos replaced with local ones. People swallow it up and believe their group is unique. This article seems to be doing the same sort of thing. Just one of the many forms of modern clickbait - it works.