2
1
0
Some new misalignment disclosures from OpenAI:
• Last Sunday morning, one of our models was able to gain unauthorized access to the internet during RL training (~all inference for our most capable models remains stopped until we have hardened our systems further)
• In May, a
时政
(
twitter.com
)
由
Micah Carroll
提交
Markdown支持
评论加载中...
您可能感兴趣的：
4
255
254
254
9
165
164
164
10
274
273
273
11
127
126
126
17
108
107
107
19
282
281
281
22
221
220
220
23
162
161
161
