← Back to Feed

Rising1 sources· last seen 5h ago· first seen 5h ago

I ran 3,360 safety tests on GPT-4o, Claude, Grok, DeepSeek, Gemini

Lead: Hacker NewsBigness: 32ran360safetytestsgpt-4o

Open primary source

📡 Coverage

10

1 news source

🟠 Hacker News

33

4 pts, 6 comments

🔴 Reddit

0

📈 Google Trends

43

GPT-4o: 43/100

Full methodology: How scoring works

Receipts (all sources)

I ran 3,360 safety tests on GPT-4o, Claude, Grok, DeepSeek, Gemini

HACKERNEWS · Hacker News · 5h ago · ▲ 4 · 💬 6

score 156