Win $50k for Solving a Single AI Problem? #Shorts
Skills:
AI Alignment Basics90%
Key Takeaways
The Alignment Research Center is offering a prize of up to $50k for solving the Eliciting Latent Knowledge problem, with proposals and contest details available online.
Full Transcript
say you've got a huge diamond you want to protect so you put it in a cool sci-fi vault with all sorts of sensors and actuators you have an ai system to run the fault and the plans it comes up with might be too complex for you to understand but it also predicts the final view from the camera after the plan happens so before you okay a plan you can check that the diamond is still there at the end but imagine a plan that allows a thief to come in and set up a screen in front of the camera showing a diamond the predicted outcome looks good so you okay the plan and the diamond is stolen but this should be avoidable right in order to predict the right fake image the ai has to know that the diamond's been stolen but how do you get that information out solving this problem in its hardest form is extremely difficult so difficult in fact that the alignment research center is offering prizes of five to fifty thousand dollars for good ideas so if you think you've got a solution based on the description i've just given you don't read the full technical report it's 105 pages of reasons why your idea won't work but if you've carefully gone through all of that and still think you've got something send it in link below the deadline is february 15th i think i'll have a go myself
Original Description
The Alignment Research Center is offering up to $50k for proposals for their Eliciting Latent Knowledge problem.
The report: https://docs.google.com/document/d/1WwsnJQstPq91_Yh-Ch2XRL8H_EpsnjrC1dwZXR37PC8/edit
Contest details: https://www.alignmentforum.org/posts/QEYWkRoCn4fZxXQAY/prizes-for-elk-proposals
#Shorts
Watch on YouTube ↗
(saves to browser)
Sign in to unlock AI tutor explanation · ⚡30
Playlist
Uploads from Robert Miles AI Safety · Robert Miles AI Safety · 36 of 47
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
▶
37
38
39
40
41
42
43
44
45
46
47
Predicting AI: RIP Prof. Hubert Dreyfus
Robert Miles AI Safety
Respectability
Robert Miles AI Safety
Are AI Risks like Nuclear Risks?
Robert Miles AI Safety
Avoiding Negative Side Effects: Concrete Problems in AI Safety part 1
Robert Miles AI Safety
Avoiding Positive Side Effects: Concrete Problems in AI Safety part 1.5
Robert Miles AI Safety
Empowerment: Concrete Problems in AI Safety part 2
Robert Miles AI Safety
Why Not Just: Raise AI Like Kids?
Robert Miles AI Safety
Reward Hacking: Concrete Problems in AI Safety Part 3
Robert Miles AI Safety
The other "Killer Robot Arms Race" Elon Musk should worry about
Robert Miles AI Safety
Reward Hacking Reloaded: Concrete Problems in AI Safety Part 3.5
Robert Miles AI Safety
What Can We Do About Reward Hacking?: Concrete Problems in AI Safety Part 4
Robert Miles AI Safety
What can AGI do? I/O and Speed
Robert Miles AI Safety
AI learns to Create ̵K̵Z̵F̵ ̵V̵i̵d̵e̵o̵s̵ Cat Pictures: Papers in Two Minutes #1
Robert Miles AI Safety
AI Safety at EAGlobal2017 Conference
Robert Miles AI Safety
Scalable Supervision: Concrete Problems in AI Safety Part 5
Robert Miles AI Safety
Superintelligence Mod for Civilization V
Robert Miles AI Safety
Why Would AI Want to do Bad Things? Instrumental Convergence
Robert Miles AI Safety
Experts' Predictions about the Future of AI
Robert Miles AI Safety
AI Safety Gridworlds
Robert Miles AI Safety
Friend or Foe? AI Safety Gridworlds extra bit
Robert Miles AI Safety
Safe Exploration: Concrete Problems in AI Safety Part 6
Robert Miles AI Safety
Why Not Just: Think of AGI Like a Corporation?
Robert Miles AI Safety
How to Keep Improving When You're Better Than Any Teacher - Iterated Distillation and Amplification
Robert Miles AI Safety
Is AI Safety a Pascal's Mugging?
Robert Miles AI Safety
AI That Doesn't Try Too Hard - Maximizers and Satisficers
Robert Miles AI Safety
Training AI Without Writing A Reward Function, with Reward Modelling
Robert Miles AI Safety
9 Examples of Specification Gaming
Robert Miles AI Safety
10 Reasons to Ignore AI Safety
Robert Miles AI Safety
Sharing the Benefits of AI: The Windfall Clause
Robert Miles AI Safety
Quantilizers: AI That Doesn't Try Too Hard
Robert Miles AI Safety
The OTHER AI Alignment Problem: Mesa-Optimizers and Inner Alignment
Robert Miles AI Safety
Deceptive Misaligned Mesa-Optimisers? It's More Likely Than You Think...
Robert Miles AI Safety
Intro to AI Safety, Remastered
Robert Miles AI Safety
We Were Right! Real Inner Misalignment
Robert Miles AI Safety
Apply to AI Safety Camp! #shorts
Robert Miles AI Safety
Win $50k for Solving a Single AI Problem? #Shorts
Robert Miles AI Safety
Free ML Bootcamp for Alignment #shorts
Robert Miles AI Safety
Apply Now for a Paid Residency on Interpretability #short
Robert Miles AI Safety
Why Does AI Lie, and What Can We Do About It?
Robert Miles AI Safety
Apply to Study AI Safety Now! #shorts
Robert Miles AI Safety
AI Ruined My Year
Robert Miles AI Safety
Learn AI Safety at MATS #shorts
Robert Miles AI Safety
Using Dangerous AI, But Safely?
Robert Miles AI Safety
AI Safety Career Advice! (And So Can You!)
Robert Miles AI Safety
Robot Dog! Unitree Go2 review #shorts #robot #dog
Robert Miles AI Safety
Tech is Good, AI Will Be Different
Robert Miles AI Safety
Apply for the Affine Superintelligence Alignment Seminar #shorts
Robert Miles AI Safety
More on: AI Alignment Basics
View skill →Related AI Lessons
⚡
⚡
⚡
⚡
GuardFall: When Decades-Old Shell Injection Tricks Beat Modern AI Safety Guardrails
Dev.to · Cor E
What 116 court judgments taught me about the limits of AI
Medium · AI
Your ChatGPT History Is a Liability. I Fixed That With a $80 Chip and a Pi5.
Medium · AI
Your Skepticism About AI Is an Asset. Here’s How to Use It.
Medium · Programming
🎓
Tutor Explanation
DeepCamp AI