Massive2 sources· last seen 6h ago· first seen 17h ago
From GPT-2 to Claude Mythos: The return of AI models deemed 'too dangerous to release'
Seven years ago, OpenAI declared its language model GPT-2 "too dangerous to release." The industry rolled its eyes. Now Anthropic is repeating the move with Claude Mythos Preview - but this time there's real evidence on the table: thousands of vulnerabilities in operating systems and browsers, found
Lead: The DecoderBigness: 100gpt-2anthropicmythosreturndeemed
📡 Coverage
50
2 news sources
🟠 Hacker News
97
370 pts, 111 comments
🔴 Reddit
0
📈 Google Trends
100
Anthropic: 100/100 🔥 spiking
Full methodology: How scoring works
Receipts (all sources)
From GPT-2 to Claude Mythos: The return of AI models deemed 'too dangerous to release'
RSS · The Decoder · 6h ago
score 181
Seven years ago, OpenAI declared its language model GPT-2 "too dangerous to release." The industry rolled its eyes. Now Anthropic is repeating the move with Claude Mythos Preview - but this time there's real evidence on the table: thousands of vulnerabilities in operating systems and browsers, found
OpenAI says its new model GPT-2 is too dangerous to release (2019)
HACKERNEWS · Hacker News · 17h ago · ▲ 370 · 💬 111
score 172
Related clusters
What should we take from Anthropic’s (possibly) terrifying new report on Mythos?
3 sources · bigness 88 · 1h ago
Anthropic Announces AI Cybersecurity Project Powered By Claude Mythos Model
2 sources · bigness 69 · 6h ago
The way Anthropic talks about Mythos if they made it open to Public
1 sources · bigness 63 · 3h ago
Anthropic Warns That “Reckless” Claude Mythos Escaped a Sandbox Environment During Testing
1 sources · bigness 34 · 3h ago
Claude Mythos knows when it's breaking the rules — and tries to hide it
1 sources · bigness 34 · 6h ago