January 31, 2026
Blame the vibes?
Automatic Programming
Proud of “Automatic Programming” — but who owns the bugs
TLDR: A creator declares AI‑assisted “Automatic Programming” is human‑led and the output is theirs, sparking a brawl over credit and blame. Commenters clash on accountability, whether “vibe coding” is a dumb label, and if coding is a spectrum—asking if speed from AI can coexist with real skill growth and transparency.
A YouTuber planted a flag: AI‑assisted coding is “Automatic Programming,” and the code is mine. “Vibe coding,” they say, is when you toss a vague idea at a chatbot and walk away. Cue the fireworks. Commenters stormed in asking: if you’re taking the credit, are you taking the blame when the app explodes?
One side came in hot. “Vibe coding is an idiotic term,” snapped one critic, arguing labels just dunk on people and ignore how teams actually work. Another called the whole thing a “false dichotomy,” saying real projects use a mix of hands‑on steering and hands‑off generation—sometimes in the same afternoon. The spiciest thread? A nod to “strong anti‑disclosure,” with some reading the post as a rallying cry to stop announcing when AI helped—while others warned that secrecy plus bugs equals reputation disaster.
Beyond the branding brawl, a thoughtful question landed: can we go faster with AI and still grow our skills? Fans loved the “collective mind” idea—training data as a shared gift—while skeptics joked, “Cool, but when production goes down, do we blame the vibes?” The meme energy was high: mood‑ring metaphors, “who owns the bug” punchlines, and pride vs. panic gifs in spirit. The debate isn’t just about code—it’s about credit, accountability, and what it means to call software ‘yours’.
Key Points
- •The author defines “automatic programming” as AI-assisted software development guided by human vision and continuous steering.
- •They differentiate “vibe coding” as minimal-involvement prompting that accepts initial AI outputs without deeper guidance.
- •Automatic programming aims for high-quality results tightly aligned with the producer’s design across multiple abstraction levels.
- •The author claims programmers can consider AI-assisted code as their own, given models train on human-generated data.
- •Redis is cited to illustrate that software value often derives from ideas and vision more than technical novelty.