In my book, the single most effective way to spend tokens is having it review code/specs you've written. One advantage to putting the ai in that position is that unreliable competence isn't much of a problem as you can ignore bad suggestions.
I would also recommend explaining the specs and doing a lot of your back and forth with a lower end model and set it to a higher end model only once the conversation history has all the context you feel the higher end model needs.
Musk said in his autobiography he announced the hyperloop plan without any intention of doing it to distract from the California high speed rail plans.
They always show me my total before the cars swipe, so as long as the obfuscation works until the card swipe, at least it would prevent dynamic pricing.
It isn't just how fast or slow it is. Reading at a slow pace gives you time to think in a way that is flexible from sentence to sentence.
To borrow the same analogy from the article, image trying to savor a meal where someone else was deciding when you take each bite. Even at a slow pace, the rigidness of the pace and your lack of fine control would still pose a problem with giving each bite it's rightful consideration.
That being said I love audio books and think I would struggle to apply this article's advice in my own life. Slowing down your audiobook is still a step in that direction, though I sometimes find that slowing it down can cause my mind to wander and my comprehension goes down and not up.
No, just the opposite. In general, the things wealthy people buy (luxuries) experience much larger swings in demand due to price changes like added taxes (in economic terms, "the elasticity of demand"). It's because they are only wants and not needs. They are also usually easily swapped. Instead of buying your wife those diamond earrings, you could get her a painting or a trip to Spain. And rich people are often very money savvy.
It's the necessities that people will continue to buy (or at least replace with close substitutes), regardless of what happens to the price.
Obviously, in this case it worked out much differently, but no, in general you can't say the wealthy people don't respond to price changes due to their wealth.
I feel like we watched different videos.. Seemed like the AI (or other monitoring system) recognized a problem with the 18000 cups of water order and quickly transitioned to a real human. That instance looked pretty production ready to me.
A lot of digital ones are "local" too in that they are context specific. As long as it stays context specific, your Uber rating is closer to being liked by your local bar tender than it is to the Chinese social credit system. Even your local bartender has a little context leakage.
I agree there is a scarier potential there. And also some do, on occasion, escape their context (mostly credit score). They also have bigger contexts, but not so big that I would jump to the Chinese social credit comparison.
That's absurd. That doesn't pass the sniff test at all for being remotely true that people would react like that to only a 3 percent tax.
I looked it up, and it was a 3 pence tax per pound. When tea was selling for 2 to 3 pence per pound. So yeah, a 100-150% tax combined with the fact that the East India Company was allowed to sell without paying the tax. That is very unjust and threatens their business a lot more than the tax alone.
One potential issue with that approach is the factors wouldn't stay very constant across generations of AI models.
While a lot of people have used various methods to try to gauge the strength of various AI models, one of my favorites is this time horizon analysis [1] which took coding tasks of various lengths and looked at how long it takes to humans to complete those tasks and compared that to chance that the AI would successfully complete the task. Then they looked at various threshholds to see how long of tasks an AI could generally complete with a certain percent threshold. They found the length of a task that AI is able to complete with a various threshholds is doubling about every 7 months.
The reason I found this to be an interesting approach is both because AI seems to struggling with coding tasks as the problem grows in complexity and also because being able to give it more complex tasks is an important metric both for coding tasks or more generally just asking AIs to act as independent agents. In my experience increasing the complexity of a problem has a much larger performance falloff for AI than in humans where the task would just take longer, so this approach makes a lot of intuitive sense to me.
I agree Hyundai should fix this for free (would make up a small portion of the bad PR for having this issue in the first place), but don't forced recalls usually only apply to defects that cause safety issues?
I'm not sure this would fit the definition of a product safety defect.
I think your take makes more sense in a world where you actually own the car fully and have the freedom to do what you want with it. Even if someone was able to write this patch themselves without the source code, distributing it would require owners to root their devices, which isn't legal in all jurisdictions.
You don't expect Microsoft or Adobe to issue fixes any time someone finds a remote exploit that let's attackers gain control of you system though security issue in their software? I 100% expect this of my software vendors even for this purchase in the past. The expectations for software and hardware are certainly very different, but even for hardware we have laws that force companies to fix their hardware in some situations.
Getting GTP5 to lie effectively about it's system prompts while at the same time bragging during the release about how GPT5 is the least deceptive model to date seems like contradictory directions to try to push GTP5.
I noticed one of the cited bluesky posts was all in French, so one might argue that technically it didn't find the English word "mouch", but rather a different French word that happens to be spelled the same. But trying to sort that out seems unrealistically challenging. "Mouch" is only in the dictionary as an alternative spelling to mooch, so probably a pretty rare word to see in English.
Have you considered that that might be the goal of releasing trickles of information about the film prior to its official release? It makes collected information feel more exclusive to super fans and encourages fans to interact with each other on social media providing fuel for Bjork focused communities. If collecting this information feels exhausting instead of exciting to you... why are you trying so hard to collect it? Just wait for the actual release.
> We need critics who devote their lives to browsing through the pile and telling us what is worth our time and what isn't.
I don't understand how you expect a critic to tell you whether its worth your time based on a collection of pre-release rumors and interviews. For deciding if its worth my time, I mainly want to hear from critics who have seen the upcoming media and I want to hear their opinion on what they saw. Why would I care to hear Ebert and Roeper's opinion on what the actors said in their press release tour? Unless it was something especially newsworthy and they wouldn't need to go digging for that. I just don't see how a critic's review would be enhanced by "devoting their lives to browsing through the piles".
> Failure to impose the RealPage rents could lead to landlords being expelled from the organization, according to the suit.
Makes this arguably much more over the line than just a bunch of landlords that happen to use the same pricing algorithm. Landlords being pressured to not use the algorithm however they want, say setting their price $100/month below the algorithm, under threat of losing access to the algorithm may be what ultimately loses this trial for them. If that is the case, we may not get to a ruling that resolves the legality of the more general practice of many people using the same pricing algorithm.
Miles seems like it might not be worth the added cost of administration. Also note that road damage as a function of weight goes up by roughly a power of 4. Doubling a vehicle's axel load causes 16x the road stress, though that may be in line with what you were suggesting and may even add weight to your point.
Right, each pandas column is a numpy array (or optionally pyarrow in pandas 2.0), but numpy is going to be faster for things that needs to be vectorized across columns, such as matrix operations.
It still not even close to the majority of Americans no matter your definition. The vast majority of the stock market isn't held by billionaires either. The total net worth of all US billionaires is ~5 trillion and the stock market is worth ~40 trillion.
I would also recommend explaining the specs and doing a lot of your back and forth with a lower end model and set it to a higher end model only once the conversation history has all the context you feel the higher end model needs.