You need fairly high IQ to become an AI researcher

Jason Voorhees

Jason Voorhees

𝕸𝖊𝖗𝖈𝖊𝖓𝖆𝖗𝖞 𝕮𝖔𝖗𝖕 • 𝟐𝟎𝟐𝟒🥇
Joined
May 15, 2020
Posts
88,173
Reputation
264,367
If you want to make it as a real AI researcher not just some prompt monkey on Upwork you need fairly high IQ. The field is very very competitive you’re not just copying code from GitHub or doing vibe coding on claude, you’re inventing new architectures, new algorithms, scaling laws, and beating SOTA models on benchmarks. All this require actual mathematical intuition and pattern recognition most people simply don’t have. There are many things in the CS world where you can just brute force your way through sheer hard work but this is not one of them


Average or even aboveaverage IQ will get you filtered out in interviews. The math, the research papers, the debugging at scale it’s all high g territory. You have to be a math genius and unlike most jobs that are repetitive and mundane, you do have to have high cognitive power in your day to day tasks.



If you’re sub 130 and still dreaming of frontier AI research. Just forget it, you’re coping, saying it how it is.
 
Last edited:
  • +1
  • Hmm...
  • JFL
Reactions: hypernormie, ewri, Wuzzdio and 27 others
All that work just for ngas to goon with the ai
 
  • JFL
  • +1
Reactions: Wuzzdio, lnceIs, Jgns and 9 others
yeah
 
  • +1
Reactions: ecstazy and Jason Voorhees
If you want to make it as a real AI researcher not just some prompt monkey on Upwork you need fairly high IQ. The field is very very competitive you’re not just copying code from GitHub or doing vibe coding on claude, you’re inventing new architectures, new algorithms, scaling laws, and beating SOTA models on benchmarks. All this require actual mathematical intuition and pattern recognition most people simply don’t have. There are many things in the CS world where you can just brute force your way through sheer hard work but this is not one of them


Average or even aboveaverage IQ will get you filtered out in interviews. The math, the research papers, the debugging at scale it’s all high g territory. You have to be a math genius to be good at these things

https://looksmax.org/threads/my-fri...0k-and-built-a-nlp-than-mogs-chatgpt.1875664/

If you’re sub 130 and still dreaming of frontier AI research, you’re coping just saying it how it is.
I mog you Jason
 
  • +1
Reactions: Jason Voorhees
@imontheloose @Glorious King @Petsmart @Wuzzdio
 
  • +1
Reactions: Wuzzdio and Glorious King
  • JFL
  • +1
Reactions: Wuzzdio, lnceIs, Thief and 3 others
‘AI RESEARCHER’ HAHAHAHAHAHAHAHAHAHHAHAHAHAHA
 
  • +1
Reactions: Jason Voorhees
@mohito @Saint Casanova
 
  • +1
Reactions: lnceIs, ecstazy and mohito
@ecstazy @Kara
 
  • +1
Reactions: ecstazy and Kara
If you want to make it as a real AI researcher not just some prompt monkey on Upwork you need fairly high IQ. The field is very very competitive you’re not just copying code from GitHub or doing vibe coding on claude, you’re inventing new architectures, new algorithms, scaling laws, and beating SOTA models on benchmarks. All this require actual mathematical intuition and pattern recognition most people simply don’t have. There are many things in the CS world where you can just brute force your way through sheer hard work but this is not one of them


Average or even aboveaverage IQ will get you filtered out in interviews. The math, the research papers, the debugging at scale it’s all high g territory. You have to be a math genius and unlike most jobs that are repetitive and mundane, you do have to have high cognitive power in your day to day tasks.



If you’re sub 130 and still dreaming of frontier AI research. Just forget it, you’re coping, saying it how it is.
What’s the next best thing you can do within the AI space without getting involved in this complex research ?
 
  • +1
Reactions: ecstazy and Jason Voorhees
The new frontier of tech and science, yet ppl treat it like a routine office job. Literally the fourth industrial revolution and sum still think it’s a joke
the joker GIF
 
  • +1
Reactions: lnceIs, ecstazy, Saint Casanova and 1 other person
What’s the next best thing you can do within the AI space without getting involved in this complex research ?
Applied AI. The researchers I mentioned build and you as an Applied AI engineer implement them into buisness in way that is a useful to clients or the buisness
 
  • +1
Reactions: Jgns, ecstazy, Kara and 2 others
Applied AI. The researchers I mentioned build and you as an Applied AI engineer implement them into buisness in way that is a useful to clients or the buisness
What IQ range is required for that type of role ?
 
  • +1
  • Ugh..
Reactions: Jason Voorhees and topology
Can confirm this as true.. The reason I'm doing so well is due to my confirmed IQ of 160+.
 
  • +1
Reactions: Jason Voorhees
What IQ range is required for that type of role ?
It is very similar in responsibilities and nature to a backend engineer but with just an AI twist. Dont need to be some math genius. Just understand system design and data structures well which most people can learn and get good at with practice. Say like 100-110 IQ.
 
  • +1
Reactions: Kara and Saint Casanova
Yeah AI research is not for most people.

This is work you'd do in top of the line academia or elite industry.

Although that doesn't mean someone decently smart can't work AI related jobs at all.

A lot of ML devs exist, and they basically set up architecture rather than making breakthroughs 24/7.
 
  • +1
Reactions: Jason Voorhees
@Acquiescence
 
  • +1
Reactions: Acquiescence
Nobody cares
 
  • +1
Reactions: Jason Voorhees
Yeah AI research is not for most people.

This is work you'd do in top of the line academia or elite industry.

Although that doesn't mean someone decently smart can't work AI related jobs at all.

A lot of ML devs exist, and they basically set up architecture rather than making breakthroughs 24/7.
True but that ain't where those insane headlines breaking pay packages exist innit? I see AI research as similar to quant research. The funnel is so small so heavily gate kept and so competitive it doesn't make sense for someone who is not extra ordinary to aim for it.
 
  • +1
Reactions: Saint Casanova and Kara
If you want to make it as a real AI researcher not just some prompt monkey on Upwork you need fairly high IQ. The field is very very competitive you’re not just copying code from GitHub or doing vibe coding on claude, you’re inventing new architectures, new algorithms, scaling laws, and beating SOTA models on benchmarks. All this require actual mathematical intuition and pattern recognition most people simply don’t have. There are many things in the CS world where you can just brute force your way through sheer hard work but this is not one of them


Average or even aboveaverage IQ will get you filtered out in interviews. The math, the research papers, the debugging at scale it’s all high g territory. You have to be a math genius and unlike most jobs that are repetitive and mundane, you do have to have high cognitive power in your day to day tasks.



If you’re sub 130 and still dreaming of frontier AI research. Just forget it, you’re coping, saying it how it is.
yeah? I thought this was water

high iq job requires high iq?

@Bryce
 
  • +1
Reactions: Bryce and Jason Voorhees
True but that ain't where those insane headlines breaking pay packages exist innit? I see AI research as similar to quant research. The funnel is so small so heavily gate kept and so competitive it doesn't make sense for someone who is not extra ordinary to aim for it.
I guess to get extraordinary pay, you have to be extraordinary yourself. Something like that.

I used to want to get into quant as well, but damn. The competition is insane.
 
  • +1
Reactions: Jason Voorhees
@Sayori
 
  • +1
Reactions: Sayori
I agree you have to be pretty well educated to work in AI
 
  • +1
Reactions: Jason Voorhees
If you want to make it as a real AI researcher not just some prompt monkey on Upwork you need fairly high IQ. The field is very very competitive you’re not just copying code from GitHub or doing vibe coding on claude, you’re inventing new architectures, new algorithms, scaling laws, and beating SOTA models on benchmarks. All this require actual mathematical intuition and pattern recognition most people simply don’t have. There are many things in the CS world where you can just brute force your way through sheer hard work but this is not one of them


Average or even aboveaverage IQ will get you filtered out in interviews. The math, the research papers, the debugging at scale it’s all high g territory. You have to be a math genius and unlike most jobs that are repetitive and mundane, you do have to have high cognitive power in your day to day tasks.



If you’re sub 130 and still dreaming of frontier AI research. Just forget it, you’re coping, saying it how it is.
bad day to be low iq
 
  • +1
Reactions: ecstazy and Jason Voorhees
Not true

I work in AI…







Pulling the fiber optics in data centers :KEKLaugh:
 
  • +1
  • JFL
Reactions: Jgns, Sayori, Jason Voorhees and 1 other person
  • Love it
  • Hmm...
Reactions: lnceIs and Sayori
@Jgns
 
  • +1
Reactions: Jgns
id would say quant is also another field

where you need a 120+ iq to be successful

physics too
 
  • +1
Reactions: Jason Voorhees
id would say quant is also another field

where you need a 120+ iq to be successful

physics too
Quant, AI and high end bio medical research. All need very high IQ
 
  • +1
Reactions: kisslessvirgin
@SomaliSub5 @davidlaidisme67
 
  • +1
Reactions: SomaliSub5 and kisslessvirgin
Quant, AI and high end bio medical research. All need very high IQ
yea

most careers that require intense knowledge and skill

but also require creativity and creation of something new

are high iq
 
  • +1
Reactions: Jason Voorhees
@hypernormie @aids
 
  • +1
Reactions: hypernormie and aids
@ewri @Aox Ofwar @lnceIs
 
  • +1
Reactions: lnceIs and Aox Ofwar
I mean, if you are seriously studying properly papers, forming questions as you read them, draft hypotheses, you won't just be completely hopeless even if you have 100iq, you can just learn how things work by doing
 
  • +1
Reactions: Jason Voorhees
If there exists a single person who grinded properly like 6h a day for 3 years who didn't make it in their desired industry, i would think they are a needle in a haystack
 
If there exists a single person who grinded properly like 6h a day for 3 years who didn't make it in their desired industry, i would think they are a needle in a haystack
Maybe AI field in general but AI research in particular no.
 
  • +1
Reactions: aids and lnceIs
Bros venting he got rejected from phd app or something
 
I have a hard time feeling like you write these posts yourself because they read so fucking weird but yeah obviously.

Any high level mathematical task that is directed at creating the framework rather than calculating based on a previously established framework will require superior fluid reasoning.

AI as in crunching big data and LLM's maybe isn't the most groundbreaking stuff for those who have expertise but if they are actually trying to create AGI somewhere that will necessarily attract some real whiz kid braniacs because it is a highly complex and novel frontier with no pre established framework to rely on. That's the exact kind of task that would require top percentile fluid reasoning.
 
AGI is a theoretical concept, not a guaranteed possibility. This pushes the task into the realm of philosophy or abstract logic and math and extremely abstract logic and math is the most intellectually demanding thing we are aware of.
 

Similar threads

M
Replies
13
Views
68
jaaba
jaaba
JOTAROSON
Replies
5
Views
29
giodd
giodd
giodd
Replies
2
Views
36
lnceIs
lnceIs
Iblame2023
Replies
0
Views
15
Iblame2023
Iblame2023
sqaqyi
Replies
1
Views
21
droidmaxxing
droidmaxxing

Users who are viewing this thread

Back
Top