AI confidently renamed my working variables and somehow broke everything while optimizing the code.
The AI Hype vs Reality
17 Comments
@[HakoneMatate] Google AI is particularly useful for finding short code snippets with one click. e.g when you want to create a CSS animation when you have not done an animation before or not for some time and forget how exactly to do it...for example. But occasionally it gives you a total bum steer, e.g. deprecated use of the link tag for favicons.
Please log in to add a comment.
Spot on 😄 My favorite genre of AI failure is the confident wrong answer, it'll invent a function that doesn't exist, swear it's in the docs, and apologize only after you paste the error. The real skill now isn't prompting, it's knowing enough to catch when it's bluffing. Which is exactly why juniors who lean on it without understanding the output are the ones at risk, not seniors.
Please log in to add a comment.
https://dev.to/neoncyan0x/the-code-that-broke-the-silence-5-impactful-truths-from-the-architecture-sovereign-2ko more like a big tech company fail for me
Please log in to add a comment.
chatGPT and google search AI also suggests deprecated use of html tags such as
<link rel="icon" type="image/x-icon" href="/favicon.ico">
When the industry has moved on to the following apparently.
<link rel="icon" type="image/x-icon" href="/favicon.ico">
Artificial intelligence my butt! More like a mindless parrot.
Please log in to add a comment.
@[Aether-Intel]
The 'singularity' is simply not possible based on current silicon based technology.
The human brain is a mathematically chaotic system. And human intelligence and wisdom etc. is an emergent property of that chaotic system.
It intelligence in any particular domain is not built into the architecture of our brains.
Silicon based technology is 100% deterministic. No emergent properties are possible in a deterministic system.
Simulating neurons and neural networks with software and data running on powerful; Nvidia chips does not change the fact that you are trying to build intelligence into the architecture of a deterministic system.
In my view we have reached a technological plateau with AI and its current iterations (chatGPT, google AI etc)
Sure they can be highly useful tools in particular narrow domains (e.g. protein folding).
But also notoriously unreliable ones if your try to do more with them than their fundamental silicon architecture can cope with (e.g. self driving cars on other than dedicated roads or tightly controlled highways etc.)
If humanity ever achieves C3P0 like AGI then it won't be based on our current silicon based technology. Or at least not based on silicon transistor networks.
Please log in to add a comment.
Please log in to comment on this post.
More Posts
- © 2026 Coder Legion
- Feedback / Bug
- Privacy
- About Us
- Contacts
- Premium Subscription
- Terms of Service
- Refund
- Early Builders
More From bilalmlkdev
Related Jobs
- Full Stack Java/Go Developer (Bilingual English/Spanish)Dev Technology · Full time · Arlington, VA
- Operational Technology Server Specialist - Windows & HyperVAstra North Infoteck Inc. · Full time · Canada
- Language Data Annotator ( Spanish)Innova software Services Inc · Full time · Canada
Commenters (This Week)
Contribute meaningful comments to climb the leaderboard and earn badges!