Ned Nguyen
@NedNguyen
Followers
99
Following
1K
Statuses
651
Full time social media user. All opinions are my own.
Berkeley, California
Joined September 2012
The best time to learn about AI is 10 years ago. The second best time is today.
0
0
0
@motisorkin @legaltweetz This kind of victimhood mentality is part of the reason the left is losing. "I am the poor innocent person who is stabbed by the other side" is a very wrong way to view people with different opinions from you.
1
0
0
@motisorkin @Noahpinion I think most people care very little about identity of a person doing a job in general these days, or at least this is true in the US. This is the key reason of why US wokeism is stupid: it implicitly assumes majority are racists whereas people really just don't care.
0
0
2
People are missing the inherent tradeoff between short-term efficiency and long-term viability of major breakthroughs. Can't really have the later with sacrificing some of the former.
*Every single* cure for a disease ultimately flowed from basic exploratory research. Stopping basic research is like stopping the mountain rains and expecting rivers of cures to still flow. Examples: 1) studying saliva of Gila monster -> GLP1's 2) studying funghi -> first statins 3) mRNA biology -> gene therapy for spinal atrophy 4) studying bacterial genetics -> CRISPR gene therapies 5) studies of nuclear magnetic resonance -> MRI scans this list can go on and on. Not only in biology but all aspects of technology.... e.g. 6) curvature of spacetime -> GPS 7) quantum mechanics -> semiconductors 8) electromagnetism -> fiber optics -> internet ...
0
0
0
@motisorkin @clairlemon Great point, I completely forgot about Europe post WWII. I asked AI and it gave the following answers about what made Marshall plan different, wdyt?
0
0
0
This gonna be mostly a human's endeavor for a while because the best work here require one to blend into the business' daily grind to observe what they do, understand problems that the business can't even articulate.
The classic software startup writes code to solve users' problems. If AI makes writing code more of a commodity, understanding users' problems will become the most important component of starting a startup. But it already is.
0
0
0
@motisorkin @clairlemon I can't think of any country in the history in the world that has succeeded changing course of another country succesfully positively. There is even a famous quote from Richard Rumelt like "it's easier to go to Mars than to change Afghanishtan to a democracy".
1
0
0
@motisorkin @JTLonsdale "Something that didn't need to exist" => one way to ensure humans still have jobs in the AI era
0
0
0
I hope the new US administration removes all nonsense regulations similar to this
GERMANY: EU regulators think it is too dangerous to simply break a bottle across the bow of a new ship, instead they insist on the use of this contraption. This is why the EU is failing. This is Economy Minister Habeck flailing.
0
0
0
@_TomHoward @JeffDean @trajan317 There is the balance between selecting for the best talents of current time and seeding the support culture for more talents in the future. I would argue only doing the former leaves everyone to be stuck in the local maxima.
3
0
1
AI barely scratches the 5% of human cognitive abilities and it already feels so excited. Imagine what would happen when it crosses the 10%-20% thresholds.
I looked at AIME problems and one thing strikes me. All problems are about computing a number. This is a tiny part of math. I was trained a a mathematician in France. And I almost never had to solve a problem of that kind. All the math work was about proving mathematical properties of mathematical objects. For instance, prove that a given group is isomorphic to another given group. This is to say that getting good at computing numbers specified by some mathematical setting is not the same as getting good at math in general. It is definitely part of math, but only a tiny part of math. There is no wonder AI focuses on number finding math problems. It is because checking the result is simple. Tackling the full spectrum of math requires a much more complex result checking machinery (formal proof checker) It is also interesting to note that AI math benchmarks only care about the final number. If that number was accidentally found via a flawed mathematical proof, then it is still considered a success.
1
0
1
In the age of AI fact checking, those who are open minded and willing to be wrong will advance so much faster than people who can only accept being right.
It is kind of shocking how little AI assistance has influenced the discourse on this site, given there is a button next to every post that lets you ask a solid AI to fact check the details. It does a pretty good job. This has made no difference at all in the spread of nonsense.
0
0
0