Ramez Naam Finds AI Models Remarkably Aligned and Improving

Combined views

1.9K

2 posts, first seen 4h ago

34 likes10 comments

Commentary on X

Could I persuade an LLM to teach me how to do something bad for society? Of course! (With an open weight model and a little jailbreaking.) But I also have Google! And movies! And novels! LLMs are overall pretty great.

9:55 PM · Jul 14, 2026

EK

E.D. Kelley@ParalaxView

@ramez AI will destroy Capitalism.

10:02 PM · Jul 14, 2026

I honestly don't understand the case that AI is misaligned. Yes, I read the model cards, and the papers. And overall AI models seem remarkably aligned to me, and becoming moreso.

9:52 PM · Jul 14, 2026

Combined views

1.9K

2 posts, first seen 4h ago

34 likes10 comments

2 saves

1 reposts

Sentiment

Positive50%50%Negative

Positive Read

Positive users praise LLMs as overall pretty great despite jailbreak risks, while negative users fear AI will destroy capitalism.

Based on 2 visible X reactions from 2 accounts; directional sample.

Digg Deeper

Ask a question below.

Published answers will appear here.

Commentary on X

Could I persuade an LLM to teach me how to do something bad for society? Of course! (With an open weight model and a little jailbreaking.) But I also have Google! And movies! And novels! LLMs are overall pretty great.

9:55 PM · Jul 14, 2026

EK

E.D. Kelley@ParalaxView

@ramez AI will destroy Capitalism.

10:02 PM · Jul 14, 2026

I honestly don't understand the case that AI is misaligned. Yes, I read the model cards, and the papers. And overall AI models seem remarkably aligned to me, and becoming moreso.

9:52 PM · Jul 14, 2026