This resolves to YES if Peter Wildeford reports that AI spam has had a discernable impact, however he defines that by March 23, 2024. It resolves to NO if he reports it hasn't. If we can't get a response from him by April 1, 2024, I will resolve this based on whether I think he'd have said yes if he did tell us.
1,000
3.00@PeterWildeford how does this resolve? I realize I said I would resolve it today without a reply but I figure I should give a notification first and wait a few days.
If I don't get a response, I will resolve this NO, but I wouldn't find it unreasonable if he said it resolves YES.
@ZviMowshowitz I'm nervous now that we may end up getting into disputes over the definitions of "discernable impact" and "AI spam".
I think I have noticed an uptick in spam getting through my Gmail spam filter and I've noticed an uptick in "wrong message text scams/spams".
In neither case has the uptick been enough to be really annoying, which is what I think I originally meant by "discernable", but "discernable" literally means "perceptible" or "noticable", which is definitely true here.
Also in none of these spam cases do I think there is clear evidence that AI was used. Like they're not insanely personalized or anything cool. I think it's possible the wrong number scams use LLMs or something but I'm not sure? They don't seem different than before, just increased in frequency. I'm not sure I'd go as far as to say my use of email/text/etc. is "worse" than before.
I've gotten some LinkedIn spam from kalendar.ai which is AI, but not very good IMO.
So I think I'm leaning now towards a "yes" resolution now, but maybe going more off the literal language of the market rather than the original spirit I had intended. The original spirit I had in mind was that LLM spam would make email/text like truly annoying to use and that anti-spam measures would be rendered more-or-less useless, but this over-the-top spirit doesn't match a plain reading of the question and may feel like moving the goal posts or that I'm being influenced by wanting my earlier bets to pay out in my favor.
Not sure what you think? I clearly should've defined terms or the original spirit more.
I have bought this down to 5%, consistent with my initial position, so you can bet against me. I won't bet on it further because it is subjective and dependent on my view.
To be clear, my original position was "discernable impact on MY LIFE" not "discernable impact generally". My life is already set up to be pretty robust to spam and so far I haven't noticed any notable difference on Twitter, email, etc. So could be very different from what @MarkIngraham is saying.
@ZviMowshowitz not sure if you want to clarify that?
@PeterWildeford
Would something like "additional safety features that make your internet experience (or a service you use within it) noticeably worse" count?
E.g. would introducing Captchas into a non-Captcha world make an analogous market resolve positive?
@TobiasHaeberli Yes, if I had to complete CAPTCHAs or other "proof of human" security systems in an obtrusive and more than one-off way where I did not have to do that before, I would resolve this market as Yes.