Scoring Show HN Submissions for AI Design Patterns
Analyzing the Rise of AI-Generated Design on Hacker News
Show HN submissions have tripled recently, with many displaying a generic, sterile look that suggests AI generation. To quantify this subjective observation, a developer created a system to score 500 Show HN landing pages against 15 identified AI design patterns.
AI Design Patterns Identified
The patterns fall into four main categories based on analysis from designer feedback:
Fonts
- Inter used for everything, especially centered hero headlines
- LLM font combos like Space Grotesk, Instrument Serif
- Geist Serif italic for one accent word in otherwise-Inter hero
Colors
- "VibeCode Purple"
- Perma dark mode with medium-grey body text and all-caps section labels
- Barely passing body-text contrast in dark themes
- Gradient everything
- Large colored glows and colored box-shadows
Layout Quirks
- Centered hero set in a generic sans
- Badge right above the hero H1
- Colored borders on cards, on the top or left edge
- Identical feature cards, each with an icon on top
- Numbered "1, 2, 3" step sequences
- Stat banner rows
- Sidebar or nav with emoji icons
- All-caps headings and section labels
CSS Patterns
- shadcn/ui
- Glassmorphism
Scoring Methodology
The scoring system uses deterministic checks rather than subjective AI judgment:
- A headless browser (Playwright) loads each site
- A small in-page script analyzes the DOM and reads computed styles
- Each pattern is a deterministic CSS or DOM check
- Manual QA showed 5-10% false positive rate
Results from 500 Show HN Submissions
Sites were grouped into three tiers based on how many of the 15 patterns they triggered:
- Heavy slop (5+ patterns): 105 sites (21%)
- Mild (2–4 patterns): 230 sites (46%)
- Clean (0–1 patterns): 165 sites (33%)
The increase in Show HN submissions correlates with Claude Code's release, leading HN moderators to restrict submissions for new accounts with /showlim in March 2026.
While not necessarily bad, the prevalence of these patterns suggests many projects are using AI-generated defaults rather than crafted designs. The author notes this parallels the pre-AI era when everything looked like Bootstrap, and suggests people may return to more distinctive designs to stand out.
📖 Read the full source: HN AI Agents
👀 See Also
Claude Max 20x Plan: Limit Increases Not Applied Despite Announcements — User Confirms with Math
A paying Claude Max 20x ($200/month) user reports that the 2x session and 1.5x weekly limit increases announced by Anthropic have not been applied to their account. They provide mathematical proof and share a complete lack of support response.
Claude Agents on Bedrock Get Autonomous Micropayments via x402 Protocol
AWS AgentCore Payments lets Claude agents on Bedrock hold wallets and make USDC micropayments mid-task via the x402 HTTP standard, enabling autonomous paid API calls and subtask delegation without human approval.
Exploring n8n as an Alternative to OpenClaw Skills for Automation
The OpenClaw community on Reddit debates the pros and cons of using n8n over OpenClaw Skills for automation tasks. Key discussion points include ease of use, flexibility, and real-world application examples.
Qwen3.5-122B on Blackwell SM120: fp8 KV Cache Corruption Issue and Performance Findings
Testing Qwen3.5-122B on 8x RTX PRO 6000 Blackwell hardware revealed that fp8_e4m3 KV cache silently produces corrupt output without errors, requiring bf16 KV cache instead. MTP optimization provided a 2.75x single-request speedup while DeltaNet constraints blocked other optimizations.