Richard Ngo

@RichardMCNgo100k followers

eppur lo si può muovere

Open on X

Top outlier

188×

Best organic post

Peak views

2.6M

Single post

Bookmarks

37k

Across 77 organic posts

Last scan

Oct 16

Community cache

Posts 77 posts

#	Date	Post	↓ Outlier	Views	Likes	Reposts	Bookmarks
1	Jul 1, '24	Today I learned (the hard way) that when a long-haul flight turns around only an hour in, it will sometimes dump fuel from its wings so that it’s light enough to land.	188×	2.6M	10k	205	755
2	Dec 28, '22	Ongoing thread of unusual party ideas I've heard lately (some great, some terrible). 100 beers of solitude: the party starts with 100 beers on a table in the middle of the room, an	103×	1.5M	15k	973	5.5k
3	Jan 3, '25	Hypothesis: the world's most valuable data is screen captures of outlier competent people going about their work. But very little of this data is recorded, let alone made publicly	55.7×	784k	2.8k	146	1.3k
4	Feb 20, '24	So apparently UK courts can decide that two unrelated jobs are “of equal value”. And people in the “underpaid” job get to sue for years of lost wages. And this has driven their 2nd	45.3×	638k	3.3k	207	632
5	Mar 28, '23	I predict that by the end of 2025 neural nets will: - have human-level situational awareness (understand that they're NNs, how their actions interface with the world, etc) - beat a	38.9×	547k	1.5k	230	706
6	Jun 18, '25	Hypothesis: we’ll look back on mass migration as being worse for Europe than WW2 was. Europe recovered quickly from WW2, because each country remained high-trust and homogeneous. B	35×	492k	12k	1.2k	861
7	Dec 24, '22	A simulation of a hurricane is not a real hurricane, but a simulation of a chess game is a real chess game. Similarly, a sufficiently good simulation of a mind is a mind itself.	34.7×	488k	1.9k	184	311
8	Nov 2, '24	Scott’s case doesn’t persuade me. He’s missing that: 1. The biggest story of the election is the massive realignment of US political coalitions. 2. Experience shows Trump’s claims	30.5×	429k	995	42	275
9	Jul 17, '24	I’m trying to understand what a Trump administration would look like, and it’s been useful to interpret Vivek Ramaswamy and JD Vance as conducting an extended debate with each othe	28.9×	407k	1.1k	98	751
10	Dec 16, '22	At OpenAI a lot of our work aims to align language models like ChatGPT with human preferences. But this could become much harder once models can act coherently over long timeframes	27.4×	385k	1.4k	271	614
11	Jul 9, '25	In my head I’ve started referring to political quadrants in terms of properties of their preferred coordination networks. Top two are centralized. Bottom two are distributed. Left	25.6×	360k	4.7k	433	2.8k
12	Sep 14, '23	It is absolutely insane that neither covid nor WW2 had any lasting effect on US GDP growth trends. That’s crucial context for any argument about long-term outcomes: our strong prio	21.5×	302k	2k	167	300
13	Sep 13, '24	Chess prodigy Josh Waitzkin learned chess in a very elegant way. He first trained to win the simplest endgames, then more complex endgames, then simple midgames, etc. By the time h	20.5×	289k	2.6k	148	1.6k
14	Apr 19, '23	Eric says he struggles to understand the reasoning behind dropping SATs. I say this very rarely, but: you can't treat it as an honest mistake. It's a power grab intended to undermi	19.7×	277k	808	86	93
15	Jan 12, '25	This also applies to people. A significant number of the most brilliant people I know avoid self-coercion to an extent that sometimes appears dysfunctional or even self-destructive	19.4×	272k	1.6k	73	798
16	Nov 13, '24	After three years working on AI forecasting and governance at OpenAI, I just posted this resignation message to the slack. Nothing that surprising in it, but you should read it mor	19.3×	271k	2.3k	117	705
17	May 11, '23	Overconfidence about AI is a lot like overconfidence about bikes. Many people are convinced that their mental model of how bikes work is coherent and accurate, right until you forc	17.8×	250k	1.3k	121	220
18	Jan 8, '25	@growing_daniel The rules is: if you want to be part of the problem, use unjust laws to extort them. If you want to be part of the solution, take the 10k (or decline it) and move o	17.8×	250k	1.7k	3	39
19	Jul 24, '24	One of the weirder side effects of having AIs more capable than 90% then 99% then 99.9% then 99.99% of humans is that it’ll become clear how much progress relies on 0.001% of human	17.6×	247k	3.8k	296	470
20	Apr 4, '23	Instead of treating AGI as a binary threshold, I prefer to treat it as a continuous spectrum defined by comparison to time-limited humans. I call a system a t-AGI if, on most cogni	17.2×	241k	657	82	256
21	Jan 6, '25	Felix Hill was a kind and brilliant man (and my former teacher and colleague) who committed suicide after struggling with ketamine-induced psychosis. Psychedelic use is common in m	16.8×	237k	1.1k	71	482
22	Nov 16, '24	The most valuable experience in the world is briefly glimpsing the real levers that move the world when they occasionally poke through the veneer of social reality.	16.2×	228k	2.1k	93	750
23	Jun 13, '24	Today I learned that the rate of dying at a given age increases almost exactly exponentially from around 30 to around 90. The gods of straight lines strike again!	15.9×	224k	1.1k	78	337
24	Jul 1, '24	The reason for turning around was really dumb: they forgot the drinking water. I’m very curious a) how that escaped the preflight checklist and b) which corporate/regulatory bureau	15.8×	223k	5k	43	70
25	Sep 8, '25	This year I’ve been thinking a lot about how the western world got so dysfunctional. Here’s my rough, best-guess story: 1. WW2 gave rise to a strong taboo against ethnonationalism.	15.5×	217k	1.9k	276	1.4k
26	Jul 18, '24	I increasingly believe that there are fundamental principles which simultaneously govern the designs of well-functioning minds, organizations and societies. Once we pin them down w	14.3×	201k	757	56	346
27	Dec 30, '24	Assume for the sake of argument that each major religion succeeded because it was built around some novel psychological insight. What insight would you ascribe to each?	14.1×	198k	1.1k	36	651
28	Nov 17, '24	I didn’t think working at OpenAI affected my tweeting much (sama’s tweets provide a lot of cover) but in the one day I’ve been unemployed I’ve tweeted more candid/spicy takes than	13×	183k	1.1k	20	175
29	Aug 1, '24	Goodhart’s law for ideologies: any false belief that cannot be questioned by adherents of an ideology will become increasingly central to that ideology.	12.8×	180k	945	136	284
30	Dec 26, '24	America winning used to be very correlated with Americans winning. But the more the US economy is driven by outlier talent and AI, the more they could decouple. To be trusted, the	12.4×	174k	684	33	141
31	Jan 14, '23	A term I've found useful lately is "intellectual empathy": the ability to see the ways that someone else's worldview is reasonable, even when it disagrees with your own. Emotional	11.3×	159k	1.2k	183	128
32	Aug 2, '24	Why is society so vulnerable to far-left ideas? One key reason: it’s hard to counter weaponized empathy. When actions are taken under the banner of a long-suffering group, that mak	11×	155k	957	107	655
33	Nov 7, '24	Extreme financial competence (like near’s) is social dark matter. Few discuss it but eventually some end up 10x or 100x wealthier than you’d expect from their career progression. U	11×	155k	931	20	512
34	Jul 9, '24	Ideologies very often end up producing the opposite of what they claim to want. Environmentalism, liberalism, communism, transhumanism, AI safety… I call this the activist‘s curse.	10.5×	147k	996	156	833
35	Mar 17, '25	Talked to a friend today who decided that if RLHF works on reasoning models, it should work on him too. So he got a mechanical clicker to track whenever he has an unproductive chai	10.4×	147k	1.8k	42	526
36	Dec 28, '22	Height equiparty: everyone has to brings along a stool, or pair of platform shoes, that brings them up to the same height as the tallest person expected at the party.	10×	141k	1.5k	16	19
37	Apr 2, '22	New York’s nightclubs are the particle accelerators of sociology: reliably creating precise conditions under which exotic extremes of status-seeking behaviour can be observed. Ashl	9.9×	0	4.1k	449	1.8k
38	Jul 6, '25	Undergrad is a waste of time and money for most students. I also expect that mass university attendance significantly lowers birth rates. So slashing college enrollment to 1/4 of c	9.9×	140k	694	27	114
39	Dec 28, '22	Berlin wall party: attendees are split into two groups. The communists get free beer, but can't drink anything else; the capitalists have to bring their own booze. Every room is di	9.8×	137k	1.9k	44	45
40	Feb 11, '22	Octopuses are surprisingly intelligent, and reproduce at 1 year old. If we'd started a breeding program 50 years ago, we probably could've gotten them smarter than dolphins by now.	9.8×	0	6.1k	631	322
41	Dec 15, '24	Yudkowsky and Yarvin are the most important intellectuals of the 21st century thus far, because their ideologies predicted—and partly caused—massive upheavals. But they needed to b	9.4×	132k	780	14	277
42	Dec 28, '22	Knaves party: everyone picks a few topics to consistently lie about throughout the party.	9.3×	130k	630	12	24
43	Jun 1, '25	Magnus Carlsen claims that one or two signals per match from a chess AI indicating when he should think hardest would make him “almost invincible”. IMO you could boost researchers	9.2×	129k	1.9k	68	449
44	Dec 28, '22	Intangible gift exchange: everyone writes down one non-physical gift they can offer, and then you hand them out randomly (optionally allowing swapping/stealing). We did this on Chr	9.1×	129k	1.8k	44	137
45	Dec 28, '22	Pomodoro party: every 25 mins a bell sounds and people have to switch conversational groups.	9.1×	129k	1.5k	29	25
46	Nov 27, '24	If you’re going to hang around rationalist spaces, you should deeply internalize what I call Hanson’s razor: never attribute to malice that which can be adequately explained by aut	9.1×	128k	2.5k	118	456
47	Mar 31, '25	I’ve long suspected that puzzle-like coding interviews are so popular because they’re the closest thing to an IQ test that’s legal to use during hiring.	9.0×	126k	1.8k	22	192
48	Feb 14, '25	It is really pretty fucked that superintelligence will arrive right in the midst of a Thucydides trap. 20 years ago China wasn’t a superpower yet. 30 years from now the US would ha	8.8×	124k	1k	34	263
49	Dec 28, '22	Listening party for the Jonestown tapes: as advertised. I can't remember who suggested this, or why.	8.6×	121k	836	8	8
50	Dec 11, '24	After doing some serious emotional processing a few months ago my body abruptly decided that it only needs 5-6 hours of sleep. I now consistently wake up 2 hours earlier than I use	8.6×	121k	832	24	448
51	Oct 7, '24	A quote from Peter Thiel that has stuck in my mind: “Whenever people think you can just muddle through, you’re probably set up for some kind of disaster.” I believed the opposite f	8.5×	119k	816	49	679
52	Dec 28, '22	People bingo: all attendees are given a bingo card of traits that other people might have, and need to fill in each square with a different name. Here's one we used recently:	8.2×	116k	959	14	108
53	Feb 11, '25	Thiel says his biggest mistake was not doubling down on Facebook after getting in early. It seems like another big mistake that he didn’t double down on Trump after getting in earl	8.2×	116k	1.4k	19	204
54	Sep 15, '25	Once you notice how often “beautiful” is used as a default compliment for women, it’s hard to unsee. Instead of “my gorgeous wife”, may I suggest describing her as courageous, sinc	⚡ 8.2×	3M	1.9k	56	819
55	Nov 1, '24	Atlas Shrugged hits harder when you realize that, since it was written, American railways have stagnated much more than basically anyone but Rand imagined.	7.8×	110k	2.2k	71	213
56	Dec 28, '22	Phenomenological potluck: my friend hosted a party where everyone brought along one unusual sensation for others to try. Variant: only painful sensations - e.g. a while back we hos	7.8×	110k	578	9	66
57	Dec 6, '23	In the long term I expect almost all human jobs to become socially oriented. Even when AIs are better at every task, people will pay a premium to interact with another human. Human	7.6×	107k	542	34	164
58	Oct 27, '23	I just learned that the RTS,S malaria vaccine spent 23 years in trials, including a 6-year delay caused by the WHO even after the EMA approved it. It’s hard to get me this angr	7.5×	105k	622	86	57
59	Jun 28, '24	Idea: $100k cash to every IMO medalist from a third-world country. It would cost maybe $10 million per year; could be one of the most cost-effective interventions ever.	7.2×	101k	910	41	111
60	Nov 16, '24	Putting this here for future reference.	6.9×	98k	622	25	164
61	Aug 9, '24	Courage is in shorter supply than genius. But integrity is in shorter supply than either. Without integrity all your courage and genius will merely serve the ends of whoever can ga	6.7×	95k	641	72	201
62	Nov 26, '24	The career advice I most often find myself giving is "maximize your chance of becoming one of the best in the world at any broadly-leveragable thing". In part this is because I ten	6.6×	93k	1.1k	50	714
63	Oct 10, '25	My main concern about insiders trading on prediction markets is not that decisions will leak, but that it incentivizes corruption. E.g. inside traders could profit more by pushing	6.6×	92k	984	32	110
64	Nov 22, '24	Ending death.	5.9×	83k	753	31	79
65	Aug 5, '23	Turns out Ray Bradbury wrote a poem about large language models, he just didn’t know it.	5.7×	80k	701	154	127
66	Nov 19, '24	My three most actionable suggestions for how to become much more emotionally integrated: 1. Talk to Claude about your feelings. 2. Try circling at least twice. 3. Identify the phys	5.7×	80k	660	16	935
67	Nov 25, '24	A friend recently told me that not only do his internal subagents bargain with each other over how to allocate his time, they also charge each other interest when they overrun thei	5.6×	79k	594	26	208
68	Jan 9, '25	Men and women are so psychologically different it's kinda weird that they WEREN'T very ideologically polarized until recently. What changed? One hypothesis: dating forces you to co	5.5×	77k	572	25	145
69	Dec 26, '24	Unconditional love is not "I will love you under any circumstances". I call that selfless love, and it's bad, because you do have a self. Unconditional love is "I won't use my love	5.4×	76k	1.5k	177	438
70	Apr 24, '23	If you had to explain human psychology to aliens, a crucial claim to convey: "humans spend their first decade doing social learning with a very high learning rate, and then often a	5.2×	74k	656	46	127
71	Sep 13, '24	People have asked me how earning San Francisco tech money affects my daily life. Today I figured out the biggest change, which is that now I’ll keep buying ebikes no matter how man	5.1×	71k	636	11	42
72	Feb 20, '24	Here’s another that seems ridiculous: warehouse operatives and sales consultants, at a private company, in a unanimous decision! Why not go all the way to central planning? Hon	4.9×	70k	853	30	40
73	Feb 20, '24	There are a bunch of equal pay cases, but the biggest is against Birmingham City Council, which paid over a billion pounds in compensation because some jobs (like garbage collector	4.9×	69k	647	26	57
74	Mar 8, '21	My girlfriend asked me: "If you were in charge of a prison, would you feed the prisoners the same food that you eat yourself?" This was such an interesting ethical conundrum that I	4.5×	0	3.4k	214	57
75	Oct 30, '24	For meritocracy to be sustainable, I think we’ll need to bring back a version of noblesse oblige. If you’re talented enough to be at the top, you have a duty to empower the rest of	4.5×	63k	712	39	117
76	Mar 6, '24	In hindsight, Her was remarkably prescient about the future of AI: - Human-level AI assistants in 2025 - Human-AI relationships as a popular use case - A gradual climb towards supe	4.5×	63k	770	48	137
77	Apr 6, '24	I’ve been writing a lot of sci-fi set in posthuman worlds lately, and thinking about new types of conflict that might open up. Some ideas in comic form (with explanations and examp	4.2×	59k	649	85	290
78	Dec 1, '24	Tweet the training data you want to see in the world.	4.0×	56k	658	74	48
79	May 10, '25	What's the best explanation for why, despite China's economy booming, the Shanghai Stock Exchange has been flat for decades?	⚡ 3.7×	2.7M	1.5k	71	442
80	Aug 6, '24	Highly intelligent people understand most things very well, but are often terrible at understanding what it’s like to be dumb. Similarly, highly empathetic people understand most e	3.2×	45k	606	34	212
81	Dec 15, '23	“LLMs are just doing next-token prediction without any understanding” is by now so clearly false it’s no longer worth debating. The next version will be “LLMs are just tools, and l	⚡ 3.2×	2M	1.3k	141	408
82	May 22, '23	@k_atanasoff Fun fact: it actually is hinged; and if you rotate the chains to be closer together you get clozempic.	3.1×	44k	1.3k	8	5
83	Feb 22, '25	People who expect Trump and Musk to fall out underrate how lonely it’s been for Trump/Musk to each fight the entire political/industrial establishment for years on every front. And	3.0×	42k	661	19	50
84	Aug 31, '25	@micah_erfan he’s trolling	2.9×	41k	597	2	8
85	Mar 28, '25	One plausible outcome of AGI is humanity ending up like the British royal family. They used to run everything. Now they retain significant formal power, but little real power. Yet	2.7×	37k	612	17	106
86	Jun 18, '25	Why compare mass migration with WW2 specifically? Because the “never again, at any cost” attitude towards WW2 from European elites has been a major cultural force pushing against n	2.6×	36k	580	18	59
87	Dec 18, '24	Something I’d love to see: an app like Omegle which randomly pairs up Americans with Chinese people for video chats, and does real-time AI translation for them. Feels like even a s	2.6×	36k	622	15	93
88	Jul 17, '22	Whenever I overhear someone at a party saying "so what do you do?" I feel a twinge of second-hand shame. Surely, as a society, we can just agree to ask each other better questions	2.5×	0	1.1k	61	218
89	Jul 17, '22	IMO also avoid "what's your story?" and "so how do you know [host]?" Better options: - What have you been excited about lately? - What about you tends to surprise people? - What do	1.9×	0	582	32	320
90	Oct 8, '24	Zero-sum thinking (e.g. “if the billionaires become poorer the rest of us will be richer”) is itself very destructive, but what’s worse is that it’s often a cover for negative-sum	1.9×	27k	705	45	83
91	Dec 19, '24	View post on X	1.9×	26k	587	42	101
92	Aug 2, '22	Hypothesis: people change their minds slowly because many of their beliefs are heavily entangled with their broader worldview. To persuade them, focus less on finding evidence for	1.8×	0	1k	91	171
93	Nov 21, '22	Ex Machina remains the best film about AI risk by a long shot. Touches on many of the key ideas in the field - details in next tweet to avoid spoilers.	1.7×	0	994	46	132
94	Oct 12, '24	@reasonisfun This theory makes me feel unsafe.	1.6×	22k	1.1k	4	7
95	Feb 11, '22	The best time to start an octopus intelligence breeding program was 50 years ago. The second-best time is now! @newscienceorg	1.5×	0	1.1k	69	15
96	Apr 2, '22	All details are from Very Important People by Ashley Mears, albeit with my interpretations added on top - check it out for much more: And if you want this thread longform, here's t	1.3×	0	542	34	238
97	Apr 2, '22	(I’m strongly reminded of Michael Sandel’s book What Money Can’t Buy - the intuitions about the corrosive effects of money are the same here, just applied to a much less high-minde	1.2×	0	544	28	250
98	Mar 25, '22	Considering hosting a Trolley Problem party. Every so often I'll randomly pick 5 victims and a decider. The victims will be thrown out of the party unless the decider selects one b	1.2×	0	763	37	48
99	Feb 11, '22	Four specific reasons their brain size could increase rapidly: - Their brains don't need to pass through a birth canal - Their head weight is supported by water - They could develo	1.0×	0	810	27	7
100	Apr 20, '22	Hypothesis: Google doesn't let people pay for ad-free search because some users are worth way more to advertisers than any reasonable subscription price, and those are the ones dis	0.9×	0	595	30	38
101	Apr 5, '22	My european friend, looking at US tech job postings, was very impressed and a little bemused that so many of them offer a salary of precisely 401k.	0.8×	0	634	10	16

Scan another creator

Run this on any account.

XOutlier builds this table for every public profile you open on x.com, with live data and every post, not just the top 50.

Add to Chrome Download .zip