AI ALIGNMENT FORUM
AF

Nominated Posts for the 2019 Review

Posts need at least 2 nominations to continue into the Review Phase.
Nominate posts that you have personally found useful and important.
Sort by: fewest nominations

2019 Review Discussion

Alignment Research Field Guide
58
Abram Demski
6y

This field guide was written by the MIRI team with MIRIx groups in mind, though the advice may be relevant to others working on AI alignment research.

⠀

Preamble I: Decision Theory

Hello! You may notice that you are reading a document.

This fact comes with certain implications. For instance, why are you reading this? Will you finish it? What decisions will you come to as a result? What will you do next?

Notice that, whatever you end up doing, it’s likely that there are dozens or even hundreds of other people, quite similar to you and in quite similar positions, who will follow reasoning which strongly resembles yours, and make choices which correspondingly match.

Given that, it’s our recommendation that you make your next few...

(Continue Reading – 4972 more words)
0Marius Adrian Nicoară1mo
It's not quite clear to me what a MIRIx chapter means.  The only thing I can connect it to is TEDx, which are events that are planned and coordinated independently, on a community-by-community basis, under a free license from TED. Is a MIRIx chapter similar to a TEDx event?
Abram Demski1mo20

The name was by analogy to TEDx, yes. MIRI was running official MIRI workshops and we (Scott Garrabrant, me, and a few others) wanted to run similar events independently. We initially called them "mini miri workshops" or something like that, and MIRI got in touch to ask us not to call them that since it insinuates that MIRI is running them. They suggested "MIRIx" instead. 

Reply
Load More
20How common is it for one entity to have a 3+ year technological lead on its nearest competitor?
Q
Daniel Kokotajlo
6y
Q
1
1 • 0
27AGI will drastically increase economies of scale
Wei Dai
6y
15
1 • 0
25Misconceptions about continuous takeoff
Matthew Barnett
6y
12
1 • 0
52The Commitment Races problem
Daniel Kokotajlo
6y
36
1 • 0
60Selection vs Control
Abram Demski
6y
14
2 • 2
26Classifying specification problems as variants of Goodhart's Law
Victoria Krakovna
6y
5
2 • 1
28But exactly how complex and fragile?
KatjaGrace
6y
30
2 • 1
29Reframing Impact
Alex Turner
5y
4
2 • 1
38Reframing Superintelligence: Comprehensive AI Services as General Intelligence
Rohin Shah
6y
29
2 • 2
66Chris Olah’s views on AGI safety
Evan Hubinger
6y
30
2 • 2
63Debate on Instrumental Convergence between LeCun, Russell, Bengio, Zador, and More
Ben Pace
6y
17
2 • 2
53Gradient hacking
Evan Hubinger
6y
33
2 • 2
58Alignment Research Field Guide
Abram Demski
6y
8
2 • 2
39Soft takeoff can still lead to decisive strategic advantage
Daniel Kokotajlo
6y
34
2 • 4
33The strategy-stealing assumption
Paul Christiano
6y
45
2 • 3
57Evolution of Modularity
johnswentworth
6y
6
2 • 1
106What failure looks like
Paul Christiano
6y
28
2 • 2
27Strategic implications of AIs' ability to coordinate at low cost, for example by merging
Wei Dai
6y
22
2 • 1
31Six AI Risk/Strategy Ideas
Wei Dai
6y
14
2 • 1
50AI Safety "Success Stories"
Wei Dai
6y
11
2 • 1
51Thoughts on Human Models
Ramana Kumar, Scott Garrabrant
6y
9
2 • 1
55Utility ≠ Reward
Vladimir Mikulik
6y
16
2 • 2
39The Credit Assignment Problem
Abram Demski
6y
32
2 • 1
55Seeking Power is Often Convergently Instrumental in MDPs
Alex Turner, Logan Riggs Smith
6y
34
2 • 2
47Why Subagents?
johnswentworth
6y
12
2 • 1
53Understanding “Deep Double Descent”
Evan Hubinger
6y
35
3 • 4
58Risks from Learned Optimization: Introduction
Evan Hubinger, Chris van Merwijk, Vladimir Mikulik, Joar Skalse, Scott Garrabrant
6y
33
3 • 3
91The Parable of Predict-O-Matic
Abram Demski
6y
16
5 • 4