Picture

Jordan Taylor @JordanTensor

Working on new methods for understanding machine learning systems and entangled quantum systems. sites.google.com/view/jordanten… Brisbane Joined December 2009
  • Tweets
    437
  • Followers
    368
  • Following
    1K
  • Likes
    29K
Jordan Taylor retweeted
AmandaAskell Profile Picture

Amanda Askell @AmandaAskell

a week ago

I'm learning the true Hanlon's razor is: never attribute to malice or incompetence that which is best explained by someone being a bit overstretched but intending to get around to it as soon as they possibly can.

Jordan Taylor retweeted
NeelNanda5 Profile Picture

Neel Nanda @NeelNanda5

2 months ago

It was great to be part of this statement. I wholeheartedly agree. It is a wild lucky coincidence that models often express dangerous intentions aloud, and it would be foolish to waste this opportunity. It is crucial to keep chain of thought monitorable as long as possible

balesni Profile Picture

Mikita Balesni 🇺🇦 @balesni

2 months ago

It was great to be part of this statement. I wholeheartedly agree. It is a wild lucky coincidence that models often express dangerous intentions aloud, and it would be foolish to waste this opportunity. It is crucial to keep chain of thought monitorable as long as possible

Jordan Taylor retweeted
balesni Profile Picture

Mikita Balesni 🇺🇦 @balesni

2 months ago

A simple AGI safety technique: AI’s thoughts are in plain English, just read them We know it works, with OK (not perfect) transparency! The risk is fragility: RL training, new architectures, etc threaten transparency Experts from many orgs agree we should try to preserve it:…

Jordan Taylor retweeted
AISecurityInst Profile Picture

AI Security Institute @AISecurityInst

2 months ago

Can we leverage an understanding of what’s happening inside AI models to stop them from causing harm? At AISI, our dedicated White Box Control Team has been working on just this🧵

Jordan Taylor retweeted
JBloomAus Profile Picture

Joseph Bloom @JBloomAus

2 months ago

🧵 1/13 My new team at UK AISI - the White Box Control Team - has released progress updates! We've been investigating whether AI systems could deliberately underperform on evaluations without us noticing. Key findings below 👇

AISecurityInst Profile Picture

AI Security Institute @AISecurityInst

2 months ago

🧵 1/13 My new team at UK AISI - the White Box Control Team - has released progress updates! We've been investigating whether AI systems could deliberately underperform on evaluations without us noticing. Key findings below 👇

Jordan Taylor retweeted
OwainEvans_UK Profile Picture

Owain Evans @OwainEvans_UK

8 months ago

New paper: We train LLMs on a particular behavior, e.g. always choosing risky options in economic decisions. They can *describe* their new behavior, despite no explicit mentions in the training data. So LLMs have a form of intuitive self-awareness 🧵

Jordan Taylor retweeted
McaleerStephen Profile Picture

Stephen McAleer @McaleerStephen

8 months ago

Ok sounds like nobody knows. Blocked off some time on my calendar Monday.

McaleerStephen Profile Picture

Stephen McAleer @McaleerStephen

8 months ago

Ok sounds like nobody knows. Blocked off some time on my calendar Monday.

Jordan Taylor retweeted
yonashav Profile Picture

Yo Shavit @yonashav

9 months ago

Observation 5: Technical alignment of AGI is the ballgame. With it, AI agents will pursue our goals and look out for our interests even as more and more of the economy begins to operate outside direct human oversight. Without it, it is plausible that we fail to notice as the…

Jordan Taylor retweeted
anthrupad Profile Picture

w̸͕͂͂a̷͔̗͐t̴̙͗e̵̬̔̕r̴̰̓̊m̵͙͖̓̽a̵̢̗̓͒r̸̲̽ķ̷͔́͝ @anthrupad

9 months ago

ais will increasingly attempt schemes, bits, skits, shenanigans, mischiefs, plots, plans, larps, tomfoolery, hooliganry

Trends for United States

Karen
94,8 B posts
GiveRep V1
1.341 posts
Good Saturday
19,3 B posts
#SaturdayVibes
1.948 posts
Rachel
79,1 B posts
FKJD FANDAY24 SG
178 B posts
#saturdaymorning
1.601 posts
Ken Dryden
5.344 posts
AJ Lee
264 B posts
Mahomes
69,6 B posts
Herbert
59 B posts
#OverstockIsBack
Marvel Tokon
3.329 posts
Chargers
122 B posts
Bader
17,4 B posts
Big Brother
25,4 B posts
Elizabeth Warren
46,6 B posts
Kelce
42,3 B posts
Rose Garden
30,3 B posts
Eswatini
23,7 B posts

You might like

pre-rat, or as we used to say, aspiring rat. strongly in favor of niceness, community, and civilization your friendly beloved shapeshifter
morphillogical 🔍
@morphillogical
431 Followers
Quantum physicist @ift_uam_csic . Will apply Cauchy-Schwarz for food.
Alvaro M. Alhambra
@AlhambraAlvaro
2K Followers
Automated/strongly-augmented AI safety research. Past: AI safety independent research and field-building - ML4Good, AGISF; ML academia (PhD, postdoc).
Bogdan Ionut Cirstea
@BogdanIonutCir2
2K Followers
Freelance software & GPT-4 application consultant. Recent Boston University grad. Effective altruist. Likes != endorsement; epistemic status: twitter
Ezra Newman
@EzraJNewman
844 Followers
Quantum Information PhD student at Scuola Normale Superiore in Pisa
Francesco Anna Mele
@FrancescoMeleAn
846 Followers
PostDoc Researcher in #QuantumComputing and #QuantumOptics @ University of Pavia | PhD in Physics
Simone Roncallo
@SimoneRoncallo
49 Followers
You save the world from unaligned AI, I'll intercept paperwork and bureaucrats for you.
Pip Foweraker 🔍
@PipFoweraker
382 Followers
post-post rat | 0.1x engineer
Sam
@samsmisaligned
83 Followers
I crash into stuff
crash
@crash23001
157 Followers
For Alignment News see: https://t.co/uK1pLqXM5u
Casebash
@casebash
612 Followers
AI, national security. All our best work is private. Fellow at RAND. Georgetown IR.
Everett Smith
@DefenseBased
212 Followers
Alignment at UKAISI and PhD student at NYU
Jacob Pfau
@jacob_pfau
2K Followers
prioritization researcher, master of logic, baby parent
Erin Braid
@erinbraid
168 Followers
all that i Do; from my little corner of Existence; is for the Love of all Beings
shakaz
@shakaz_
492 Followers
CambrianImplosion ⏸...
@CambrianImplos1
11 Followers