This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
Senator Josh Hawley:
Huh? For reference, section 230 is here. In short, section 230 says that companies aren't liable for information posted on their websites by third parties. This means that Google can't be sued for showing ISIS.com on your search results, because ISIS is a third party, and ISIS.com is their content, not Google's. Section 230 doesn't apply to generative AI because generative AI isn't a third party. If Google Gemini replies to your prompt with, "Thank you for joining ISIS. Recommended pipebomb targets in your area are X, Y, and Z," Google can't use section 230 as a defense if Y sues them for being bombed, because Google generated the information.
If I were to steelman Hawley's point, I guess it would be that Google as a company benefits from section 230, and so repealing it would punish them for creating "woke" AI and cut off a source of funds for AI development, but I don't think Hawley's use of the phrase "these AI companies" is easily read as referring to only "AI companies which are bankrolled by social media products."
If you are familiar with simulacrum levels, you may have had a bit of difficulty grokking level 4. I think an intuitive definition of level 4 is, "politician speak that doesn't fit into levels 1, 2, or 3". Which level is the tweet by Hawley on? It's not 1, because it isn't true. It's not really 2, because it's not trying to convince you of a proposition. It's not 3, replace "conservatives" with "liberals" and "Google Gemini" with "𝕏", and it could be from AOC. That leaves 4. It's just word associations. Woke AI is bad. Tech companies make woke AI. Section 230 something something big tech censorship. Put it in a box, shake it up, let the manatees do their thing, post whatever comes out to Twitter.
To ironman, Hawley thinks that LLMs work as a fulfillment of the argument ad absurdem from Batzel, where Google as a company has slurped in a slurry of data from undiscoverable initial providers, and Google engineers have carefully tweaked and twisted it to only provide the results they want, such that Google 'hasn't generated/produced' the content only by the strict literal sense where a ransomer might not have 'written' the letter they cut from newsclippings.
This isn't technically correct, but the ways that it's wrong are technical and not-obvious, and given Daubert and stochastic parrots, I'm not sure I'd bet money on it not going to court (or even not convincing a jury).
To steelman, AI companies, whether social media or search or just-plain-LLMs, aren't in the business of selling answers: they're selling API keys. Section 230 means that some of their clients -- not all, but a large portion who produce end-user-facing text -- can't be liable or even brought to a courtroom for something defamatory, which is not a small selling point. More critically, this allows the actual LLM production to be laundered through a horde of intermediates who've put their own tiny tweaks into play, making it extremely hard to bring serious lawsuits to court against even the most intentionally tortuous conduct, and near-impossible to do so successfully.
Hawley's a demagogue, and isn't considering this. OpenAI might not even be considering it (I'd give ~60% odds at low stakes for them, though I'd put a sizable bet at long odds that Google has had separate legal and actuarial teams look over it). But it's a question that has far bigger impact on the business applications of current LLM tech than anything blase like copyrightability.
They plainly are. When I give Bing/Google a question in a natural way it tries it damnest to give me an answer, even if half the time that answer is spam and the other half of the time the answer is gaslighting me about something I clearly know not to be the case irl.
Fair point, but are you the purchaser, or the product, when doing so?
It's currently direct enough that you can put blame somewhere, but I don't think Google expect that use case to be where and how it makes its money from Gemini (or future LLMs), any more than the open testing grounds directly talking to the thing are.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link