In the end of 2028, will the public generally accept that talking with AI Girlfriends is cheating?
Plus
17
Ṁ3692029
33%
chance
1D
1W
1M
ALL
If at the year of our lord Jesus Christ of 2028 there is a general consensus in the internet (therefore skewing towards Millenials and GenZers) that having an AI girlfriends is cheating in relationship, I'll resolve this market to YES.
I see what is the current majority opinion on Twitter and the relevant subreddits. I'll ask for something that my guts tell that ~70%+ of the relevant public thinks it's cheating. This means a focus in the anglophone market.
If hard data from prolific pollsters come, I'll too consider that.
This market resolves to my understanding on Dec 31st 2028.
I won't bet.
This question is managed and resolved by Manifold.
Get
1,000
and3.00
Sort by:
Related questions
Related questions
In 2028, will at least 350,000 Americans (1/1000) be talking at least weekly to an AI they consider a romantic companion?
74% chance
Will more people interact with AI via voice than via text at the end of 2027?
40% chance
In 2028, will at least 350,000 (1/1000) Americans be talking at least monthly to an AI therapist or coach?
73% chance
In 2028, will at least 3.5m Americans be talking at least weekly to an AI they consider a romantic companion?
23% chance
When AI girlfriends will become socially acceptable to at least one age group?
Will we believe that AI character bots make it harder for kids to form human friendships by 2028?
61% chance
Will more than 10% of US young men have tried an "AI girlfriend" by end of 2024.
10% chance
Will AI "love bots/virtual girlfriends" be good enough to reduce the % of "unwarranted male attention" by 50% by 2027?
29% chance
Will AI girlfriends overtake real girlfriends before 2035?
13% chance
In 2028, will at least 3.5m (1/100) Americans be talking at least monthly to an AI therapist or coach?
63% chance