Imagine an AI monitor that checked the sanity of pilot actions. In this case, it would be easy to determine that cutting off fuel would be irrational. In that case, the AI could voice a warning and ask for confirmation. The AI model would consider potential situations where a very quick fuel cutoff would be potentially reasonable. So, not an AI that directly affects actuators, but one that just voices sanity checks.

One more possible choice. Have the human perform the complicated, high risk procedures but relegate the lower risk tasks to the robot. This is sort of like an airplane autopilot mode, where the human pilot can relax during the low-risk coasting part of the flight to have more stamina and concentration during the high-risk takeoffs and landings. Of course, this only works if most of the procedure time is low risk.

It doesn't necessarily have to be a binary choice. Maybe the robot can perform the surgery with a human watching every step, sort of like a Waymo with a human driver with hands near the steering wheel at all times. Or the human can perform the surgery, but the robot can be performing the same virtual surgery and issuing real-time warnings for potential problems.

The definition of inefficiency is whatever you need is not really needed and is therefore inefficiency and whatever I want is needed and therefore not inefficiency. This has always been true regardless of party or ideology. Look at what DOGE has axed and squint to see if there's any logic or reasoning in what was cut and what was not. Part of what DOGE cut was based on pure ideological whims and part was based on cutting something regardless of what was cut.

True. However, the main thing is income. Women in these high income countries can get jobs that allow them to be independent and either choose to remain single or to be in a relationship without getting married or having children. Most importantly, women are still significantly penalized in their careers for getting pregnant, despite all the self-congratulatory proclamations to the contrary. So, these economies essentially heavily motivate women to not have children. Some countries offer pittances to encourage having children, but the career penalty is so overwhelming that all the government programs are practically insignificant. This penalty is true at both high and low levels of income.

This correlation between macroeconomic measures of economic vitality and individual quality of living is an interesting discussion. There's probably some correlation, albeit not necessarily a strong correlation, between the two. Furthermore, income inequality may cause a strong national economy to lead to great lives for the few rich and struggles for the poor masses. Maybe the most interesting question is whether strong macroeconomic numbers are needed for sustainable broad quality of living, or if weak macroeconomic numbers necessarily lead to an inevitable deterioration of broad quality of living.

Exactly. How is this a robot? To me a robot either needs to move itself around or move other objects around. This thing does neither.

Yes, this works for private equity and those that want a quick artificial way to pump up stock prices or valuation in the short term.

All companies have deadweight to some extent. The big challenge is identifying that deadweight. Somehow, new management always has the hubris to believe that they have the ability to identify the deadweight. The one big ace in the pocket of these managers is that it's impossible to determine in the short term the accuracy of that identification. So, no one can tell them that they're wrong until long after they've been let go.

It used to be that Republicans and conservatives would complain about the liberal bias in news media. Then businessmen like Murdoch realized that he could make a lot of money by pandering to the right wing folks. But at least there were sources across the political spectrum.

What changed recently was that super rich guys like Musk and Bezos realized that they could use their money to outright buy and control these news sources. Instead of just being annoyed by hearing stuff they didn't like, they realized that they could bend the news to their liking and force it upon others. Not just censoring news and people, but actually creating the news itself.

But did the lawyer intentionally lie, or was she lazy and incompetent, delusionally trusting in the AI info? Of course, neither deceit nor incompetence are good, but they perhaps deserve different punishments and remedies.

Since there's no generally accepted definition or concept for AGI, why do we care? Different companies, research groups, etc. are working in different AI fields and on different use cases, and very few of them work on AGI, whatever that might mean.

Perhaps the one practical definition of AGI is the concept associated with AI that is intended as clickbait.

The Tensor SoC chip is very different from the TPU. Totally different architecture and different target domain.

There are multiple dimensions to helping the environment, and sometimes what's good for one dimension can be bad for another. This is one of those cases. Yes, avoiding the carbon dioxide emissions incurred for manufacturing a car is a good thing, since the increased efficiency of a new car won't offset the car manufacturing for a long time and maybe never.

However, that's just from the perspective of greenhouse gases. There's also the separate issue of air pollution and its health effects. For that, it's clear that getting rid of old cars is a clear win. Which issue is more important? If you live in New Delhi, air pollution is a huge problem, among the worst in the world. The air pollution is so bad that it's 10-20x worse than WHO guidelines and impacts both health and longevity. Old cars is a contributor to the air pollution crisis.

Climate change is important, but it's a long-term problem with varied worldwide effects. This type of air pollution is localized and immediate. If it were me and I lived there, I'd want to address the health issue first. The only local pushback is from considering family finances and not climate change.

"But some said the government support for Hangzhou's tech scene had scared off some investors. Several company founders, who asked not to be named so they could discuss sensitive topics, said it was difficult for them to attract funds from foreign venture capital firms, frustrating their ambitions to grow outside China."

Why do Chinese companies need non-Chinese funding? There are many Chinese venture capital firms. These firms can also fund expansion outside of China. The main problem with foreign/Western funding of Chinese companies is not Chinese government control or influence. Rather, the core problem is the instability of control and ownership in China. The law is defined by whatever the current thinking of the government is, much more than what the written law actually says. The government can favor certain companies. Situations like ARM China can happen. This risk is in addition to the inherent risk in picking startup winners.

So the Indian company seems to be just an outsourced manufacturing plant for the Chinese company. This could be seen as a problem of a poorer country being taken advantage of by a richer country. However, this is the same situation that China was in a few decades ago. The Chinese used the outsourcing opportunity to learn, then accumulate income, then copy, then innovate, and then finally compete. Why can't India do the same? This could be viewed as sort of a different take on the Innovator's Dilemma.