Sam Altman is a huge risk to META. He has similar morals to Zuck and a much better technical team. If OpenAI turns on the slop generator, they could hit Facebook and Instagram hard. Wang is probably smart enough to help navigate that risk.
For every Quinton Tarantino there are dozens of people who could make amazing films but never had the opportunity for a wide variety of reasons that now mostly go away.
And for each of those dozens, there are now millions of untalented people who can generate terrible films with a few clicks. How does a Tarantino, or even one of your new talents, ever get discovered?
We've seen unprecedented democratization of media over the last 30 years, primarily due to the internet. I'm personally not seeing a corresponding positive impact on cultural creativity. I see a lot of tweets and a lot of Etsy stores, and the audience for long novels (for example) is aging out.
You're not looking hard enough: indie video games, educational YouTube content being on par if not better than TV documentaries, countries generally not making great series, now being able to be seen by people anywhere, all sorts of great musicians and writers.
Like with everything one can see the positives or focus on the negatives. The Internet today has great, good, medium and bad content, just as before, what's different is the effort and excitement we used to have back then.
The entire reason OpenAI has a high valuation is the expectation that AI will get a lot better in the next few years. If that happens, building a clone of Cursor/windsurf should be trivial. The only reason you would buy windsurf today is to either pump up the bubble OR use it to increase your market share of developers by taking users away from claude
Most but not all companies are bottlenecked by organizatial issues not speed of completing a jira ticket. A lot of those companies have moats or sales issues that prevent competitors from easily taking market share.
So instead of seeing mass drop in job openings you will see companies that are not bottlenecked by org issues start to move very fast. In general that will create new markets and have a positive effect on kobs
Agree. For interrupt work for my team, I often spend more time writing the story and prodding coworkers for PR review than I do writing code. GH Copilot has been our tool of choice and it's been a great, advanced tab complete. Haven't used it for much more.
I'm not following the logic here. There are tons of free tier AI products available. That makes the world more fair for people in very poor countries not less.
Lots of models are free, and useful even, but the best ones are not.
I'm not sure how much RAM is on the average smartphone owned by someone earning $5/day*, but it's absolutely not going to be the half a terabyte needed for the larger models whose weights you can just download.
It will change, but I don't know how fast.
* I kinda expect that to be around the threshold where they will actually have a smartphone, even though the number of smartphones in the world is greater than the number of people
This is the nature of social media. If you see 10 examples of something happening in another country there is a good chance you will alter your behavior.
The reverse scenario is happening too. There are americans that refuse to go to europe because of the examples they seen of immigrants committing crimes and ruining neighborhoods.
As someone who spends a lot of time on both locations, I know that both scenarios are rare, and can logically overcome the emotional response after seeing examples online. But for people who don't travel much I understand.
I travel quite a bit, and there are only 2 places on Earth where I was close to a shooting happening in real time: São Paulo, Brazil, and Oakland, California.
As much as the scenarios can be rare, there is an undeniable sense of everything hanging by a thin thread when traveling around the USA, which I've only experienced in Latin American countries in all my trips.
Giving that I'm originally from Brazil even though with Swedish citizenship, I won't be traveling to the USA anytime in the near future. I have no idea what could happen, might be a completely rare occurrence to be profiled at the border, jailed for no cause, etc., but there's nothing in the USA worth enough to make me even more paranoid at crossing its borders. It's more like the straw that finally broke the camel's back, it's been brewing for a while, I've been stopped by CBP for holding both a B-1/B-2 visa on my Brazilian passport as well as an ESTA on my Swedish one, I do not want the potential issues that another interrogation by CBP at present times could create, like being sent to some jail for 20 days instead of just being refused entry and put on a plane to get back to the EU.
I travel even more than you and only got robbed twice in 40 years on earth: in Johannesburg, South Africa and Stockholm, Sweden.
You can claim it's rare, but it happened on 100% of my trips to Sweden. Therefore, I will never travel back to Sweden. It's just not worth it and I have no idea what could happen (stabbing? murder?). Nothing in Sweden is worth the absolute fear this country provokes.
How do you know? It doesn't look like OP specified how much they've traveled.
> I only got robbed twice...in Johannesburg, South Africa and Stockholm, Sweden...it happened on 100% of my trips to Sweden
That sucks, I'm sorry that happened. It sounds like it happened once, and so that "100%" is just one trip by one person? Unfortunately, that result is within the realm of randomness, though I'd understand if, to you, it felt bigger than that.
That said, I don't think it compares to arbitrary (or worse, politically- and personally-motivated) government detainment (or otherwise harming) of innocent people.
Please don't come here, it's definitely dangerous and you shouldn't ever step foot here again to avoid any danger to your life, it's a hellhole that no one like you should ever attempt to visit :)
It's not rare, it will happen to you. Do not come.
> This is the nature of social media. If you see 10 examples of something happening in another country there is a good chance you will alter your behavior
That's actually the nature of bayesian probability: if something is happening a higher proportion of the time vs. before, it's more likely to happen now vs. before. If that something is bad, that means higher risk. It's expected that a rational actor would act to minimize risk to them.
> As someone who spends a lot of time on both locations, I know that both scenarios are rare
Precisely how rare is it, over the last couple months, for US immigration officials to detain someone (perhaps "for further questioning", perhaps to a prison) who hasn't violated any laws? Claiming "it's rare" isn't very useful. Remember, the expected probability is ~0.
Particles don't exist. We just perceive waves with high decoherence rates as particles. Things we call objects effectively have a 100% decoherence rate. Things we call waves like light have low decoherence rates.
Decoherence is the process that makes it impractically difficult for an experiment to be designed that makes your observations the two interfering possibilities in some kind of double-slit experiment.
Interpreting this in the many-particle case is more difficult, but the basic idea is that due to single-particle uncertainty, you can't have a definite number of particles indexed by momentum and a definite number of particles indexed by position at the same time. If I had 100 particles that were definitely at x=0, in terms of momentum they'd be spread out over the range of possibilities unpredictably.
Not exactly. The Heisenburg uncertainty principle doesn't apply to knowledge (actual observations), it applies to observables (things that could affect interactions in principle). That is, Heisenburg uncertainty is not merely a limit on how fine our measurement instruments could get, or even how much information about an interaction we could conceive of and store. It's a limit on how strongly those properties can affect an interaction at all.
That is, the future direction and momentum of an interaction between two particles can't depend very strongly on both the position where the interaction happened, and on the momentum the particles had before the interaction. If the interaction is a direct collision, so the position is heavily constrained, then the momentum the particles had before the collision will not really matter a lot for what happens after they collide.
If you were to "put yourself in the shoes of" one of the particles, you could say that, because it "knows" where the other particle is at the time of the collision with high precision, it can't "know" the momentum the other particle had with any precision, so it's future movement can't depend strongly on that. But this stretches the definition of "knowledge" far beyond the normal understanding of the word.
> It happens outside QM, and even outside physics. It’s not a physical attribute, it’s statistical.
This is not true at all. In classical mechanics, particles have fully definite properties. In the theory, if two particles collide, the position and momentum they'll have after the collision depend on their exact position and exact momentum before the collision, with no bound on precision.
Of course, classical mechanics admits that we can't measure things to any level of precision, there is some practical bound below which noise in the measurement will drown out the signal. But the interaction itself has no such bound, it happens with infinite precision. If the speed of one of our particles were higher by just 10^-100 m/s, its trajectory might be completely different.
This is not possible in QM. In QM, if the particles collide (they meet at an exact point in space-time), then their trajectories afterwards wouldn't change even if one of their speeds were 10 times higher: if they have definite position, their speed is extremely fuzzy, and it can't significantly affect their trajectories after the event.
And QM turns out to be right abput this, when you measure things precisely enough.
I was making the point that the uncertainty principle is not about QM. And it is not. It occurs in a wide range of even classical systems.
As for your comment about them not having defined proporties; this is also just one interpretation. You argue for violation of realism. That’s fine, but unnecessary.
Violation of locality or realism is only needed in the context of Bell inequalities, and this assumes there is no superdeterminism and you are “free” to choose your experiment, which is of course a rather strange argument to have to begin with.
To spare your struggle with language: commutator of hermitian operators is a physical property of mathematical origin, inherited from linear algebra, because QM is described by linear algebra, so physics inherits all aspects of linear algebra as physical properties.
How frequently a wave would go through just 1 of the slits. If you threw a baseball at a wall with two baseball sized slits it would basically always go through just one of the slits. You would never see an interference pattern.
This is because a baseball is interacting with other matter on the way to the slit. A photon on the other hand might not interact with any matter and it stays as a wave and you can see an interference pattern on the other side.
What use cases? We are still in the "throw everything at the wall and see what sticks" phase, which might end before anything of value is found (if previous fads serve as a lesson here).
I doubt diffusion algorithms are a good fit for the kind of real time computer vision processing that self driving cars depend upon, do you have good sources for that?
About writing assistance, based on our office/copilot recent rollout, I don't see that getting past the novelty effect and turning into something most people will want to pay for (besides for programming and niche use cases).
China is building an entirely independent semi conductor supply chain and if they are not competitive now, they will be in the near future. US sanctions forced them into turbo charging their efforts.
I have two SO-DIMM sticks of a domestically produced Chinese brand named FASPEED, with chips bearing logos and markings that I don't recognize from anywhere else. These sticks' mfg. datestamp is "44-24" and from what I've learned they cost little in China but come with a salted price tag when sold through channels aimed at Western customers. I'm not sure if they come in fast-enough variants to compete, and not sure about the quality or longevity otherwise. FASPEED makes SSDs, too, but I have no data on those. I also have an M.2 SSD from another Chinese brand called XINCUU which I previously had never heard about. The label of that SSD is in parity with expectations of Chinese business morals - it claims to be PCIe NVMe with "1000 MB/s speed" but is in reality a SATA device, and it does not perform even close to the ~550 MB/s limit of SATA 3.0. Both of these run unusually warm for DDR4 memory and M.2 flash storage, leading me to believe they are wholly designed and produced in China.
Regardless of how valid the study is, it is most likely useless.
These kinds of studies have been done for decades and type 2 diabetes rates have only gone up.
There has been clear evidence for decades that obesity and high carb diets increase risk of diabetes. Comparing tea to coffee or Skittles to m&Ms is a useless research project as far as diabetes goes. Because it is extremely unlikely that someone will discover that the cure for diabetes was a small change in lifestyle like that.
