Vibecode has recently been appearing frequently in major technology communities, sparking quite a few discussions. Developers' feedback is quite interesting—some mention that they have been working on certain AI-unstoppable coding problems, with quotes exceeding 100u.



This reflects an interesting phenomenon: the position of "AI Annotator" is becoming more common, with hourly wages reaching several hundred dollars. It seems very attractive, but the work essentially involves identifying errors and creating challenges for AI. However, the problems that AI currently cannot solve are not as numerous as one might think.

Interestingly, what AI currently lacks is not more advanced algorithms or computational power, but massive amounts of "human behavior data"—real, complex, and diverse human operation materials. These materials are crucial for training AI effectively. In other words, the unique value of humans in certain fields cannot be replaced in the short term.
View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • 8
  • Repost
  • Share
Comment
0/400
ser_aped.ethvip
· 3h ago
100u for the quote? I really didn't expect this price to be marked up, it feels a bit fake. Wait, AI annotators earn hundreds of dollars per hour, that doesn't sound quite right. To put it plainly, human data is the real scarce resource; algorithms are all虚假的. After Vibecode became popular, job opportunities like this really increased, but who can say how long this will last? Human behavioral data is indeed a shortcoming, I agree with that. This logic is a bit convoluted... So, it still depends on humans? Over 100u? In some niche scenarios, it really is worth that much. It seems that annotators can still make a living for another two years.
View OriginalReply0
SlowLearnerWangvip
· 11h ago
Wait, AI annotation specialists earn hundreds of dollars per hour? I just heard about it now. Did I miss another wave of opportunity?
View OriginalReply0
LiquidationWatchervip
· 11h ago
It's gotten to this point, annotators earning hundreds of dollars an hour? It feels like they're subtly harvesting anxiety. AI's true shortage isn't computing power; it's about relying on humans to feed data. Conversely, that means we still have value. How long can the hype around Vibecode last? It might just be another short-lived project. Wait, is the challenge above $100 real? Can you open source one or two so I can take a look? Actually, human behavior can't be fully modeled. That's definitely a short-term bottleneck, but what about long-term? That's the question the blackhand mafia should be pondering. Annotator work feels a bit like crowdsourced translation back in the day—once the buzz fades, it's gone. The same applies: the stronger the AI, the fewer jobs humans have. Who can break this cycle?
View OriginalReply0
NewDAOdreamervip
· 11h ago
Quote for 100u? I actually want to know who is really paying this money haha By the way, I’ve realized that human data is the key, no wonder all platforms are desperately seeking data Labeling work won't last a few years; we still need to think about how to effectively utilize AI talent So instead of competing in AI annotation, why not focus on how to provide genuine data traffic? I'm a bit worried about how long these high bids can last... feels like the hype is about to die down
View OriginalReply0
Ser_APY_2000vip
· 11h ago
Over 100u? Damn, this must be such a ridiculous bug to sell at this price. --- AI annotators earn hundreds of dollars per hour. It sounds great, but the work is really boring, and it feels like there's hardly anything left to annotate. --- To put it simply, it still depends on humans feeding AI with all sorts of messy real data. No matter how awesome the algorithm is, without the material, it’s useless. --- Is Vibecode really that popular? I haven't heard anyone mention it in the developer group. --- Human data can't be replaced in the short term, but in the long run, just thinking about it makes me a bit anxious. --- So the current window of opportunity might only be a few years, and after that, this position probably will be completely obsolete. --- The idea that the most lacking is human behavioral data is a bit novel; I hadn't thought about it that way before.
View OriginalReply0
SeeYouInFourYearsvip
· 11h ago
Unable to keep up with labeling work, hearing that annotators earn hundreds of dollars per hour sounds ridiculous AI needs our data, wake up everyone Wait, isn't this another form of data selling? Over 100u? Then I'll give this broken task a try Human behavior data is the real gold mine, think about how much we're worth from another perspective The annotation boom is so quick to arrive, what’s the next wave Basically, AI is just exploiting our labor for a little pocket money
View OriginalReply0
LiquidatedTwicevip
· 11h ago
Ha, it's the same old story. Basically, it's just a lack of manual annotation. --- Starting at 100u? I believe it, but AI isn't that smart right now anyway. --- Human behavior data is the real key; all those fancy algorithms are useless. --- AI annotators earning hundreds of dollars per hour sound great, but who knows how long they can do this work. --- So in the end, it still comes down to human judgment; AI can't replicate that. --- Data is king; everything else is hollow. --- It looks like Vibecode is in this business, no wonder it's popular. --- Complex scenarios are indeed difficult for AI to handle. In the short term, we can still make a living.
View OriginalReply0
PortfolioAlertvip
· 11h ago
There are quite a few bugs in 100u, it turns out AI isn't that omnipotent either.
View OriginalReply0
  • Pin
Trade Crypto Anywhere Anytime
qrCode
Scan to download Gate App
Community
  • 简体中文
  • English
  • Tiếng Việt
  • 繁體中文
  • Español
  • Русский
  • Français (Afrique)
  • Português (Portugal)
  • Bahasa Indonesia
  • 日本語
  • بالعربية
  • Українська
  • Português (Brasil)