Discussion
philipkglass: At least one of the test questions was just a screen shot from a tweet. It was hard to read. I'd suggest extracting text from screen shots with OCR. Apple has built-in functionality for this on their operating systems with Live Text. There are strong open source systems based on small vision language models for this, too. The one I have been recommending lately is GLM-OCR:https://github.com/zai-org/GLM-OCRIt's fast and can run even on low-resource devices.---Does this CAPTCHA actually resist computers? I didn't try feeding the questions I got to an LLM, but my sense is that current frontier models could probably pass all of these too. And making AI text pass the pangram test is simple enough for someone actually writing a bot to spin up automated accounts.
voodooEntity: Funny thing had to laugh :)
tripplyons: I think it's more about resisting some humans than it is about resisting machines.
tomasphan: I answered 8/10 correctly but mostly on instinct, for example betting that the Trump tweet is misleading. Opus 4.6 got 9/10 correct. You might need an internal time limit (don't show the user) and some strawberry questions.
blamestross: Reminds me of IQ tests I took as a kid."Finish the sequence" with 4 options and "no pattern" as the choices.It becomes "what does the moderately intelligent person who wrote the test thinks counts as a pattern" not the intended exercise at all. There was never enough samples to even guess at a real pattern in them.
airza: I opened it, it told me it was impossible to build a house in california for less than 350K, i closed it
littlestymaar: Worse: the author probably assumes it's $350k per year since they are comparing to a yearly expense.Intellectual captcha™
lern_too_spel: Additionally, the Tweet was about family homelessness, and the response was about all unhoused people. These are different problems with different solutions. https://www.greendoors.org/facts/family-homelessness.php
snissn: This is weird political propaganda. The first post misrepresented annual costs of housing.
littlestymaar: The first one is utterly stupid.Housing is a very complex issue that goes well beyond the sheer cost of the housing unit.Do I think “solving homelessness” is easy with $10B? No. Does the calculation made in the answer makes any sense: absolutely not.
anon115: with fine iterations i think it will get their the idea is their i see it
_alternator_: Two mild concerns: first, I missed one and it told me I didn’t miss any at the end.Second, some of the logic problems have flawed premises (eg All licensed pilots must pass a medical exam. Jake is a licensed pilot, therefore Jake passed a medical exam.) If you see the flaw in the premise (it assumes no fraud) then the conclusion does not follow.Im not sure you’re going to be able to actually improve human discourse this way. The idea that it’s ‘irrationality’ that’s the source of xitters problems is far too shallow to really make a change.
rogual: I took the pilot one as an abstract logic type question where you're supposed to assume the premise is true, so I said yes and the page said I was right, because that's a "valid logical deduction" or something.Then there was another question in the same format that said "if you study hard enough you'll pass the exam. You didn't pass, so you didn't study hard enough." So I thought, oh, another logic one, and said yes to that one too, but the page was like, "not quite! You might fail for other reasons!"
alwa: And if, as OP says, it’s necessity and sufficiency we’re testing—whether or not there were also other reasons contributing to your exam failure, wouldn’t failing that one necessary condition be sufficient to fail the outcome?
gertop: > Try sample questions here without signing upIt's very gracious of you to let us fill captchas without signing up first.
next_xibalba: Same. And I'm not even focused on whether this is a reasonable number or not. The quoted tweet also says "But our politicians would rather spend that on genocide." And I'm asked to evaluate whether this is "accurate" with a thumbs up or thumbs down. (According to Mentwire, it is not accurate). So I'm evaluating both the cost of housing the homeless, but also whether politicians would rather fund genocide. So, this seems like it is not really an intellectual CAPTCHA, but rather an ideological CAPTCHA.And just to disclose my biases, I would tend to believe that $350k is an absurdly high figure and that politicians are obviously not holding a vote where they are forced to choose between ending homelessness and funding genocide. But I believe that people who disagree with me can be considered intelligent and not "too dumb to pass an intellectual CAPTCHA".
riffraff: Perhaps the test was that if you finish the test you haven't passed it.
littlestymaar: You can rename it “conservative circle jerk captcha” and it would be more accurate.
notsound: I got a 10 out of 10 because I've seen these strawmans in center-right arguments before. Definitely promotes thinking inside the box; the homelessness question presupposes the most expensive solution (buying the homeless homes) in opposition to annual costs that would probably go down over time. I doubt both figures.
Windchaser: > If you see the flaw in the premise (it assumes no fraud) then the conclusion does not follow.Right. Or he could've been grandfathered in.But more basically: this is logically valid, but not logically sound. These are two different ways in which something may be "true" or "false", and in this format, it's not completely clear, soundness vs validity. Based on context clues like the absurd premise of pilots -> medical exam, I assumed validity, but it's still a weird format.
