I used to have similar concerns as you — how can anyone truly know what other people are like? Unless we’re doing research with the scientific method, we can only speculate unscientifically, right? Without science, what we say is just our belief, not established fact.
But how do you explain people who intuitively understand things? Mathematicians, for example, intuitively understand math. Psychologists and experienced authors intuitively understand people. We gain intuition through education and experience, which in turn improve our understanding and sensitivity towards the truth. Expert mathematicians, for example, _can_ have a good sense of whether a theorem is true before they prove it. And in general, people who possess scientific knowledge can intuitively know things.
I do agree with your intent, though — we need to possess humility about the accuracy of our beliefs. The author can’t factually know what other people feel and think without asking them.
But we also owe some deference to wisdom. Being wise is like being an expert darts players: you’re better able to throw darts into the bulls-eye than most people. If we develop a wisdom worth trusting, we should trust it.
SoftBank sounds familiar… weren’t they a major investor in WeWork? With such poor investing acumen, why haven’t they gone bankrupt yet? Perhaps the $40 billion raised by OpenAI can only be spent on services from SoftBank’s other investments? Do investors ever restrict how their investments are spent?
Is “multimodal reasoning” as big a deal as it sounds? Does this technique mean LLMs can generate chains of thought that map to other modalities, such as sound and images?
Fair point. There’s a real difference in attitude between speculation and saving. Most people buy stocks to save for retirement, not to speculate that they’ll have enough money for retirement.
But most people aren’t investors, either. Typical people buy stock for their returns, not for their productive capacity. My point is that’s speculation.
Most people have a speculator’s mindset. They would be happy to make a windfall on the market. For example, who wouldn’t be happy to own a plot in downtown Detroit before an economic boom raised land prices? Or stock in a startup before an established company announced they would acquire it?
Although someone who puts money in the stock market for retirement may intend to save money, since a 401k seems like just any another asset, but it’s quite different from other assets: Bank accounts fund lending to companies for productive investments. Homes provide a residence. Bonds provide money to countries and companies for investment. But when individuals put money in the stock market, it’s because they hope the asset will appreciate.
• Character. Is the public trusting? Are they anxious?
• Framing. Is the institution described compassionately, for example, as a national champion or underdog? Or is it described as a villain?
• Sectarianism. Are there motivated partisans who want to disparage the institution?
• Evidence. What particular evidence exists for unethical behavior?
• Ethics. What unethical behaviors have the institution done?
I am skeptical that institutions have become less ethical today compared to the past (e.g., today we would be appalled if a European country waged war against another state to sell narcotics, but Britain literally did that 150 years ago), so I have a hunch that other factors are the primary drivers for declining trust.
You’re right—both directions are true. But my point is that only the forward (or reverse) direction of a fact may be worth knowing.
We frequently only posses recall for one direction of a fact. Why? One direction may be important (what is Tom Cruise’s favorite color?), but the other direction may not (whose favorite color is green?).
So if we inquire into whether LLMs actually possess intelligence, their asymmetric knowledge seems similar to human knowledge and hence seems consistent with intelligence, rather than problematic.
The author is drawing the wrong the conclusion-knowledge relationships can be asymmetric.
The forward relationship of a fact (What is Tom Cruise’s favorite color? Green) may be worth knowing, but the reverse relationship (Who’s favorite color is green? A billion people, including Tom Cruise) may not be worth knowing.
A good analogy is ice. We all use “freezer-grown crystalline water,” but prior to the invention of refrigeration, we harvested natural ice from frozen lakes and stored them in ice houses for use throughout the summer.
Sure, there’s some romance from hand-harvested ice, but you can’t beat the price and purity of ice from a freezer.
The US is difficult to compare with other developed countries because it essentially has (1) affluent core regions and (2) poor peripheral regions. The US is essentially bimodal.
The core regions resemble the developed world, with similar levels of education, wealth, and political perspectives.
The peripheral regions are much worse off. Entire states, such as West Virginia, are peripheral. Think low wealth, low life expectancies, and entirely different politics from the rest of the developed world.
If you average values from both regions, you won’t really be characterizing either of them. The value will be too low for everyone in the core, and it will be too high for everyone in the periphery.
So when comparing the US to other countries, you’ll get a much more accurate picture by providing separate summary statistics for the core and periphery.
To play devil’s advocate, we really should ban 30 year fixed mortgages, not 5/1 ARMs. I know it sounds contradictory, but hear me out.
One of the Fed’s primary methods to control inflation is adjusting economic demand by making loans cheaper or more expensive.
In our current regime of fixed mortgages, new home buyers disproportionately bear the cost of the Fed’s effort to reduce inflation, since only they need to pay the higher costs. But if everyone had ARMs, the cost of reducing inflation would be spread out over everyone with a mortgage, so the pool of people affected would be significantly wider and the needed interest rate increases to fix inflation would be significantly lower.
As an example of “bad influences” being an ancient topic, in the 1st century CE, the orator and educator Quintilian gave advice to parents in the Roman Empire wondering whether public schools would be a bad influence on their kids.
Although Quintilian agreed that public schools can be a bad influence, he argued that parents were the worst influence on their own kids.
“If only we did not ourselves damage our children’s characters! We ruin their infancy by spoiling them from the start. That soft upbringing which we call indulgence destroys the sinews of mind and body. If a toddler crawls around in purple, what will he not want when he grows up? […] No wonder [why children have bad character]: it was we who taught them, they heard it all from us. They see our mistresses, our male lovers; every dinner party echoes with obscene songs; things are to be seen which it is shameful to name. […] The wretched children learn these things before they know they are wrong. This is what makes them dissolute and spineless: they do not get these vices from the schools, they import them [vices] into them [the schools].”
Individual states’ pay transparency laws are already applying upwards pressure on salaries across the country.
My company has multiple groups in different states including Colorado, and in anticipation of needing to post salary ranges for our open positions in Colorado, my group (with no positions in Colorado) preemptively bumped up everyone’s salaries to the midpoint of their pay bands to avoid anyone becoming frustrated if they learned they were in the bottom half. Despite the preemptive adjustments, a colleague of mine became angry and quit when they found out their salary wasn’t at the very top of their position’s pay range.
So pay transparency laws are having a big impact—not only in the obvious cases of candidates negotiating salaries in the states that passed the pay transparency laws, but also for average employees in other states who didn’t even need to do anything except learn how much their labor was worth.
Agreed, I think it’s currently causing problems in the US.
In theory public media like PBS and NPR should address the gap, but those organizations tend to depend on donations from the wealthy, so their coverage skews towards their interests and conversational style.
I think the government should provide newspaper subscription vouchers to the poor.
PSA: Newspapers educate the public, and their services are essential for the functioning of our democracies. The best newspapers fund their operations through subscriptions, so please help society become well-informed by subscribing to your favorite newspapers.
> An individual who felt strongly about a single issue could hoard their credits until its moment arrives, then blow all of their saved credits on it. That would allow passionate minorities occasionally to outvote indifferent majorities.
In other words, hoarders and impulsive spenders would decide policy?
Ultimately political systems are judged by the quality of their policies. Why would passionate minorities produce better policy than median voters?