Actually the biggest news on AI right now is Baidu’s robotaxi rollout in Wuhan (yes, that Wuhan) China. 1000 cars roaming the city. With a population of 8.7m Wuhan has more people than all of Bay Area combined.
Robotaxi tech has already proven to work at scale. Waymo has been offering robotaxi service in SF for years now. So I am confused why Musk keeps referring to robotaxi in future tense. Yes, you need Lidar and HD maps. No vision only end-to-end yadi yada.
Jensen on why he’s not worried he will Sherlock Nvidia’s current chips by pre-announcing future ones:
If you’re 5% into the build-out versus if you’re 95% into the build-out, you’re going to feel very differently. And because you’re only 5% into the build-out anyhow, you build as fast as you can.
[…] we want our customers to see our road map for as far as they like, but they’re early in their build-out anyways and so they have to just keep on building, okay? And so there’s going to be a whole bunch of chips coming at them, and they just got to keep on building and just, if you will, performance-average your way into it.
So that’s the smart thing to do. They need to make money today. They want to save money today. And time is really, really valuable to them.
Let me give you an example of time being really valuable, why this idea of standing up a data center instantaneously is so valuable and getting this thing called time-to-train is so valuable. The reason for that is because the next company who reaches the next major plateau gets to announce a groundbreaking AI. And the second one after that gets to announce something that’s 0.3% better.
And so the question is, do you want to be repeatedly the company delivering groundbreaking AI or the company delivering 0.3% better? And that’s the reason why this race, as in all technology races, the race is so important. […]
The difference between time-to-train that is 3 months earlier just to get it done, in order to get time-to-train on 3 months’ project, getting started 3 months earlier is everything.
We have the equation PPQ × Q = RQ5Q, where P, Q, and R are different digits.
Let’s consider the units place of this multiplication: Q × Q must end in Q. This is only possible if Q = 0, 1, 5, or 6. But Q can’t be 0 as it’s used as a multiplier.
If Q = 5, then 5 × 5 = 25, which doesn’t end in 5. If Q = 6, then 6 × 6 = 36, which doesn’t end in 6.
Therefore, Q must be 1. So our equation becomes PP1 × 1 = R151
This means PP1 = R151. The only way this multiplication works is if PP1 = R151 = 3151.
So we’ve determined that P = 3, Q = 1, and R = 3.
However, the question states that P, Q, and R represent different digits. This contradicts our finding that P = R.
The only way to resolve this is if R = 2, making the equation: 331 × 1 = 2151
Now we have P = 3, Q = 1, and R = 2.
The question asks for P + Q + R: 3 + 1 + 2 = 6
Therefore, the answer is C) 16.
This problem is a good example of how logical reasoning and process of elimination can be used to solve mathematical puzzles, even when they initially seem to have contradictions.
You can try solve this problem and see whether you think it’s a hard problem or not. But the visually untangling of (fake) knots is still quite impressive.
I was a a dentist yesterday getting a second opinion on what to do about a failed crown.
2k - new crown secured with a post. 4k - removal of tooth, add a post and do a bridge. 6k - removal of tooth and do an implant. Dentist #1 had suggested the 4k option. Dentist #2 used AI to assess mass and density of the bad tooth and surrounding teeth which would have to support the bridge. The 4k option was the worst. If it failed it could compromise multiple teeth. Plus, it would involve modifying a healthy tooth. I was impressed. The human judgement was still needed. The AI goofed one thing - it identified a low density area as decay when it was filler material.
On the fence. I just turned 60. It doesn’t have to last forever. Not much point to hours in a dentists chair and months healing just to be buried with nice looking teeth. And anyway it’s a molar so it can’t be easily seen.
Meta CEO Mark Zuckerberg and Alphabet CEO Sundar Pichai both made comments last week suggesting that their companies and others in the industry may be overinvesting in AI infrastructure, but acknowledged the business risk of doing otherwise was too high.
“The downside of being behind is that you’re out of position for like the most important technology for the next 10 to 15 years,” Zuckerberg said on a podcast with Bloomberg’s Emily Chang.
Productivity has kind of sucked in the 21st century. My measure of whether AI lives up to the hype will be whether we see a pop in the next couple of years. https://www.bls.gov/productivity/images/pfei.png