AI ALIGNMENT FORUM
AF

Ramana Kumar
Ω50871350
Message
Dialogue
Subscribe

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by
Newest
No wikitag contributions to display.
Is there any rigorous work on using anthropic uncertainty to prevent situational awareness / deception?
Answer by Ramana KumarSep 26, 202410

Vaguely related perhaps is the work on Decoupled Approval: https://arxiv.org/abs/2011.08827

Reply
Consent across power differentials
Ramana Kumar1y10

Thanks for this! I think the categories of morality is a useful framework. I am very wary of the judgement that care-morality is appropriate for less capable subjects - basically because of paternalism.

Reply
Consent across power differentials
Ramana Kumar1y20

Just to confirm that this is a great example and wasn't deliberately left out.

Reply
Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover
Ramana Kumar1y84Review for 2022 Review

I found this post to be a clear and reasonable-sounding articulation of one of the main arguments for there being catastrophic risk from AI development. It helped me with my own thinking to an extent. I think it has a lot of shareability value.

Reply
Systems that cannot be unsafe cannot be safe
Ramana Kumar2y30

I agree with this post. However, I think it's common amongst ML enthusiasts to eschew specification and defer to statistics on everything. (Or datapoints trying to capture an "I know it when I see it" "specification".)

Reply
Why do we care about agency for alignment?
Answer by Ramana KumarApr 23, 202340

This is one of the answers: https://www.alignmentforum.org/posts/FWvzwCDRgcjb9sigb/why-agent-foundations-an-overly-abstract-explanation

Reply
Teleosemantics!
Ramana Kumar2y10

The trick is that for some of the optimisations, a mind is not necessary. There is a sense perhaps in which the whole history of the universe (or life on earth, or evolution, or whatever is appropriate) will become implicated for some questions, though.

Reply
AI and Evolution
Ramana Kumar2y32

I think https://www.alignmentforum.org/posts/TATWqHvxKEpL34yKz/intelligence-or-evolution is somewhat related in case you haven't seen it.

Reply
$500 Bounty/Contest: Explain Infra-Bayes In The Language Of Game Theory
Ramana Kumar2y90

I'll add $500 to the pot.

Reply
Load More
22Dialogue on What It Means For Something to Have A Function/Purpose
1y
0
28Consent across power differentials
1y
6
19Refining the Sharp Left Turn threat model, part 2: applying alignment techniques
3y
5
36Threat Model Literature Review
3y
3
45Clarifying AI X-risk
3y
16
30Autonomy as taking responsibility for reference maintenance
3y
2
37Refining the Sharp Left Turn threat model, part 1: claims and mechanisms
3y
3
58Will Capabilities Generalise More?
3y
28
13ELK contest submission: route understanding through the human ontology
3y
2
22P₂B: Plan to P₂B Better
4y
13
Load More