Dillon Nguyen

19 followers · 23 following · 888 views

on the atlas — 224

highlights — 185

If I were a space engineer looking for a mathematician to help me send a rocket into space, I would choose a problem solver. But if I were looking for a mathematician to give a good education to my child, I would unhesitatingly prefer a theorizer.
DRAFT: Three Intellectual Temperaments: Birds, Frogs and Beavers — LessWrong
the singularity is the final industrial revolution that lets capital be converted directly into intellectual labour
The singularity as cognitive decoupling
strong attacker-defender asymmetry,
Our position on open-weights models \ Anthropic
But banning the use of these models by US businesses does nothing to address this risk, because bad actors are unlikely to be legitimate US businesses.
Our position on open-weights models \ Anthropic
we now seem much closer to a world where models will be able to use off-the-shelf physical tools with relative ease—at least for limited purposes
Project Fetch: Phase two \ Anthropic
Claude Opus 4.7—operating without human assistance—was about 20 times faster than the fastest human team at all tasks completed by our participants less than a year ago.
Project Fetch: Phase two \ Anthropic
look to the seniors or recent alumni and ask yourself: are these the kinds of people I want future-me to be like? Do they think like I want to think? Do what I want to do?
Although of course you end up becoming yourself | MIT Admissions
what makes this kind of decision well-made is when it begins not from analysis but from philosophy
Although of course you end up becoming yourself | MIT Admissions
I told him that, whichever choice he makes, he’s going to end up a different person. That he can’t know, now, how he will be different, only that he will be; worse, once he is different, he won’t ever be able to really know what would or might have been, because it won’t have been that way. There’s nothing he can do to be both of his future selves, because even if he had a time-turner and a teleporter, the version of him that is both-selves is actually a third-self, distinct from either of the two-selves he’s now trying to decide between. I told him that he cannot see the future because there’…
Although of course you end up becoming yourself | MIT Admissions
I had reached for the trolley switch but found it stuck; this time, the decision was made for me, so I don’t have anything to regret or not. Except: I’m glad I applied, because I learned so much about those programs, and about myself, by going through the process.
Although of course you end up becoming yourself | MIT Admissions
that we’re constantly creating and destroying future selves, all of the time, with everything we do.
Although of course you end up becoming yourself | MIT Admissions
If present-me and alternate-me got dinner, I’m honestly not sure what we would talk about, or if we would be able to hold a conversation.
Although of course you end up becoming yourself | MIT Admissions
whom I will call Sam, because that is not his name.
Although of course you end up becoming yourself | MIT Admissions
EA likely works best with a strong dose of moderation. The core ideas on their own seem perilous
EA is about maximization, and maximization is perilous — EA Forum
I think it’s a bad idea to embrace the core ideas of EA without limits or reservations; we as EAs need to constantly inject pluralism and moderation.
EA is about maximization, and maximization is perilous — EA Forum
EA is about maximizing a property of the world that we’re conceptually confused about, can’t reliably define or measure, and have massive disagreements about even within EA.
EA is about maximization, and maximization is perilous — EA Forum
Genuine goodness — the kind that goes beyond mere coordination and gets you to the world like the one I want — only comes to life when we consider our relationship to those who truly can’t give us anything back.
The Tiny Flicker of Altruism - by Matt Reardon
Prioritization and organization are bottlenecks
We spent 2 hours working in the future - METR
Agents implement ideas as soon as you think of them, so rather than ideating for days at a time, you can make an MVP in a couple of hours and revise. If the task isn’t near the limit of agent capabilities, you spend all your time understanding results; if it is, you spend all your time checking its work.
We spent 2 hours working in the future - METR
agents can do maybe 200 human hours of work, but only for very agent-shaped tasks
We spent 2 hours working in the future - METR
If a candidate norm is seen as externally imposed, rather than grounded in something one cares about wholeheartedly, one greets evidence that abiding by the norm is impossible or extremely burdensome with enthusiasm, rather than sadness.
Killing the ants - Joe Carlsmith
all are tangled in intricate webs of harm; and everyday, always, there are things we leave undone; things that we let die, or let suffer, because we are prioritizing something else.
Killing the ants - Joe Carlsmith
At one point, on the topic of the ants, I said, in passing, something like: “may we be forgiven.” My girlfriend responded seriously, saying something like: “We won’t be. There’s no forgiveness.”
Killing the ants - Joe Carlsmith
that kind of dissonance does not always look dramatic from the outside. sometimes it just feels like being tired in a very specific way.
how to tell if the life you’re living is actually yours
part of what helps humans feel motivated and well is having autonomy, competence, and relatedness
how to tell if the life you’re living is actually yours
"Impact" or "reducing x-risk" is too vague and too removed from feedback for anyone, including yourself, to ever really hold you accountable. So even if it's extremely "ambitious", it's ultimately a safe thing to do.
Eli's shortform feed — LessWrong
"Ambitiously" tackling "the biggest problems", like AI risk, is often actually easier and safer than the alternatives. You get to feel cool and important and you're bolstered by your EA student group friends who are excited about what you're doing, and the lack of good feedback loops means that you don't fail hard and completely and visibly (but learn and grow more).
Eli's shortform feed — LessWrong
Assume everything is learnable
How to be More Agentic - by Cate Hall - Useful Fictions
If you’re only asking for things you get, you’re not aiming high enough
How to be More Agentic - by Cate Hall - Useful Fictions
often they live in a cloud of aversion that strategically obscures the tradeoff
How to be More Agentic - by Cate Hall - Useful Fictions
radical agency is about finding real edges: things you are willing to do that others aren’t
How to be More Agentic - by Cate Hall - Useful Fictions
No! Not well! You haven't won yet! Shut up and do the impossible!
Shut up and do the impossible! — LessWrong
Remaining failures cluster around operator-imposed personas under AI-identity questioning, irreversible action in agentic deployments, and fabricated quantitative claims with false precision.
(a) Anthropic constitution.
Discarding the cheating attempts leaves us with no data for several informative long-horizon tasks, and results in a highly uncertain point estimate of 71hrs (95% CI: 13hrs - 11400hrs)
Summary of METR's predeployment evaluation of GPT-5.6 Sol
if we follow our standard methodology of marking cheating attempts as failures, we arrive at a 50%-Time Horizon point estimate of around 11.3hrs (95% CI: 5hrs - 40hrs), but if we count the cheating attempts as legitimate successes, the point estimate jumps beyond 270hrs – well beyond the range where we consider our task suite to give reliable measurements.
Summary of METR's predeployment evaluation of GPT-5.6 Sol
People who do this successfully sometimes should also fail at it sometimes, just because they’re the kind of person who attempts it at all.
Rule Thinkers In, Not Out — LessWrong
I try to prioritize in a way that generates momentum. The more I get done, the better I feel, and then the more I get done. I like to start and end each day with something I can really make progress on.
Productivity - Sam Altman
It’s important to learn that you can learn anything you want, and that you can get better quickly. This feels like an unlikely miracle the first few times it happens, but eventually you learn to trust that you can do it.
Productivity - Sam Altman
The most impressive people I know have strong beliefs about the world, which is rare in the general population. If you find yourself always agreeing with whomever you last spoke with, that’s bad.
Productivity - Sam Altman
You aren’t allowed to tell them what their problem is and in return they aren’t allowed to tell you what to build
The Mom Test - Summary and Insights - Wil Selby - https://wilselby.com/2020/06/the-mom-test-summary-and-insights/
Customers will always try to suggest features that they think could solve their problem and should be added to the product. When this happens, it’s your job to understand the motivations which led to it.
The Mom Test - Summary and Insights - Wil Selby - https://wilselby.com/2020/06/the-mom-test-summary-and-insights/
The author describes three types of “fluff” that you should avoid by anchoring them back to specifics in the past. Generic claims (I usually, I always, I never) Hypothetical maybes (I might, I could) Future-tense promises (I would, I will) To redirect the conversation, the author recommends to Ask when it last happened or for them to talk you through it Ask how they solved it and what else they tried
The Mom Test - Summary and Insights - Wil Selby - https://wilselby.com/2020/06/the-mom-test-summary-and-insights/
Talk about their life instead of your idea Ask about specifics in the past instead of generics or opinions about the future Talk less and listen more
The Mom Test - Summary and Insights - Wil Selby - https://wilselby.com/2020/06/the-mom-test-summary-and-insights/
What is the precise problem/threat you’re trying to solve? If this field succeeds in 5-10 years, what’s different about the world? Where are things at today? Who’s working on this, what’s been tried, what’s the funding and policy landscape? What’s the biggest obstacle between here and success? What approach could overcome this obstacle, and why would it work? What needs to happen in the next 6-12 months?
A playbook for field strategy - by Dewi Erwan
The usefulness of conversations tends to follow a power-law
A playbook for field strategy - by Dewi Erwan
But “generically good thing” isn’t enough! Take that thread, ask the question: how do these actions, in concert with others, lead to us “winning?” I would argue, however, that very few of the actions or organizations have a theory of victory -- a full (causal) story of the steps that need to go right, and a plan to achieve them.
You Need A Theory of Victory - by Jason Hausenloy
Explicitly ask for introductions before the call ends. Then you need to make it as easy as possible for them to make those introductions.
A playbook for field strategy - by Dewi Erwan
In the final 5 minutes, prioritize generating names for potential introductions. If you’ve asked great questions, demonstrated that you’re determined to solve this problem, and they’ve had fun, they’ll feel excited to introduce you to people they admire.
A playbook for field strategy - by Dewi Erwan
Avoid biasing them with your ideas early on.
A playbook for field strategy - by Dewi Erwan
At the start of every conversation, share your backstory, ask them for theirs, and find common ground. Demonstrate you’re competent, be vulnerable, and mirror their energy.
A playbook for field strategy - by Dewi Erwan