I'm aware of that. It was detailed in TFA. I'm actually tickled pink that the part I got wrong wasn't the scam but the very idea that any of these companies would in any way invest in actual hardware development. I am both chastened and amused by that fact. I should have known better. It's just 100% scam instead of 99% scam.

In a previous post on AI wearables, I already spelled out why this isn't just being sold as an app. Link

Let's start with the fact that this "AI" assistant will work about as well as Siri or Alexa. Meaning, not very well at all.

To answer another question posed further up, this isn't an app because it uses a combination of custom chipset to accelerate LLM processing, combined with the fact that the actual purpose of this device is to harvest data to sell. It's not an app so Google and Apple don't get to wet their beaks.

If you look at the IPCC reports from, I believe it was 2006, they lay out the path necessary to keep warming below 1.5C. It stipulated that GHG emissions would need to be halved by 2010 (they weren't). Then halved again by 2020 (they weren't). And halved again every decade until the 2050s (they won't be). At which point, using unknown and undeveloped technology working against entropy, GHG emissions actually go negative at increasing rates until the end of the century (this is the modern equivalent of where medieval scholars would say "and then a miracle will happen"). Simultaneously, GHG emissions from agriculture would need to drop to zero by 2050 (they've increased). And all this in service of a target of 1.5C of warming which is a nightmarish scenario on its own.

Because we didn't hit those earlier targets, the targets we would have to hit now are even more draconian in nature. If you project forward from what we've actually done, we're looking at 3.5C of warming, which will be utterly apocalyptic. That's not hyperbole; we'd face a collapse of the biosphere similar to the PT boundary event, known colloquially as "The Great Dying" in which something like 90% of species went extinct and the oceans became toxic. Even if everyone met all proposed and agreed upon limits (which no one except the likes of Bhutan has done), we'll still hit 2.5C of warming.

You need to understand three facts about 2.5C. First, the last time the Earth was 2.5C warmer than it is now, the oceans were 27m (89ft) higher than they are now. Second, a quarter of the human population, and nearly half of world industrial capacity, lies on the coasts within 27m of sea level. Third, humans can't breathe salt water.

We're not going to reduce our GHG emissions to zero, let alone take them significantly negative, while simultaneously relocating a quarter of our population and half of our industry. None of the proposed climate actions are even of a scale to that challenge. The Inflation Reduction Act was the largest GHG-reducing piece of US legislation, bar none. Biden does deserve credit for that much. If we did five more of them today, it wouldn't be half enough to hit 1.5C.

Per capita, Americans emit more CO2 than the rest of the G7, also. (Although the oil sands combined with relatively small population size might drive Canada higher.) So, what's your point? The climate cares about the absolute amount of GHGs in the atmosphere, not the relative emissions.

My point was that, certainly in the case of the US, and most likely in every other case as well, there is no enforcement mechanism required by this agreement and that any enforcement mechanism would be unacceptable to all parties as they all presume to violate these targets at the earliest convenience.

America is the largest emitter of CO2 in the G7 by far. In order for this agreement to be binding in the US, it has to be ratified in the US Congress. The lower chamber, the US House of Representatives is currently controlled by the GOP. They will not ratify this. It also will not be able to pass in the Senate due to the narrow majority the Dems hold there. Two defectors from coal producing states like Pennsylvania, Illinois, or West Virginia, or coal consuming states like California or Michigan will result in a failure to ratify.

In the US, the ability to sue the government (standing) is surprisingly limited. It's also proven an incredibly ineffective means of enforcement in the past.

I'm admittedly less familiar with other G7 countries' legal systems.

Highly unlikely to happen within Western nations, either. The decrease in coal-fired powerplants in the West is almost entirely due to cheap and abundant natural gas undercutting the price per kWh of electricity generation. It's cheaper to move natural gas because you can use a pipeline instead of railroads. It's cheaper to mine natural gas because once the well is drilled, labor costs plummet. This means if you have to money to start a coal plant, starting a natural gas plant instead will put you in the black faster. But as you can see after Fukashima and Ukraine took other sources offline, coal was turned to quite readily.

If there are no concrete and costly penalties for not hitting the target and no enforcement mechanism in place, then this is just words. No party to this agreement would allow such penalties or enforcement mechanisms to exist. The reason the won't allow them to exist is that they have absolutely no intention of actually keeping to the targets. They just want the positive press saying they are tackling climate change without the pain of actually having to do something meaningful.

For those of you who don't realize this, all that super fast delivery options Amazon has been giving you for a low monthly fee are almost entirely subsidized by investor funds. It's part of a monopoly bid to drive competitors out of business before the bill comes do so they can leverage their monopoly to jack up prices and service the debt they've accrued. Amazon isn't actually innovative. It's the Sears catalog with a website. Lots of places do drop shipping. Hell, Ali Express exists. But they all make you pay the true cost of shipping. Amazon is borrowing huge sums of money (and viciously exploiting their workforce) to build out the network of warehouses and processing centers needed to situate the products people want close enough to them to deliver them in two days. You simply can't do that economically. Amazon also won't be able to drive Walmart, Ali Express, and Ebay out of business before they burn through the cash reserves they've gotten through debt. As interest rates tick up above 1% (which is still historically quite low), the money tap from investors will slow, and suddenly it's like a plate spinner who put up one too many plates; you don't lose the one plate too many, you lose all the plates.

For those of you who don't know the reference in the title to this post: The first rule of hole digging is that if you find yourself stuck in a hole, stop digging. Instead, Amazon (and Uber, and Google, and Facebook, and Tesla, and Twitter...) have broken out the excavation equipment. After all, they must be closer to the other side by this point, right?

Let's start with the fact that this "AI" assistant will work about as well as Siri or Alexa. Meaning, not very well at all.

To answer another question posed further up, this isn't an app because it uses a combination of custom chipset to accelerate LLM processing, combined with the fact that the actual purpose of this device is to harvest data to sell. It's not an app so Google and Apple don't get to wet their beaks.

This whole decline started in earnest with Jack Welch at GE. None of his many sycophants ever acknowledge the object fact that Jack Welch's ideology has never been anything but a naked failure in practice. The man ran GE, one of the largest and most powerful companies on Earth, into the ground such that it is no little more than a name brand bought to slap on goods made by some Chinese third party.

The same management philosophy drove MacDonald Douglas into the ground and is currently killing off Boeing.

The first time one of these reports hits an actual court containing one of LLM's inevitable hallucinations, it's going to get bounced and probably sack the whole case. Technically submitting one of these LLM-generated documents will be regarded as perjury, and the whole system will collapse overnight. And no one will get their money back.

It's all vaporware and hype.

If nine people sit down at a table with a Nazi, there are ten Nazis at the table.

If a person sits down to an interview with Tucker Carlson, there are two Nazis doing an interview for Twitter.

I didn't see anything in the article to suggest that the reactor has actually successfully produced a fusion reaction. I'm not talking about net power gain. I mean I didn't read anything to suggest that they've even induced fusion in the reactor at all. The off the shelf permanent magnets don't seem to have enough power to actually confine plasma to fusion temperatures. I'm pretty sure the only thing this "reactor" has done is show that the magnetic field they've generated is actually consistent with with their mathematical models, meaning it's not actually a reactor at all as there is no reaction taking place inside it.

The reason electromagnets, usually superconducting magnets, are used in fusion devices is that permanent magnets simply can't generate enough Teslas of magnetic flux to confine a plasma of fusion temperatures. Proving that you can make a quasisymetrical field with permanent magnets for a fusion reactor is pointless if you can't make permanent magnets with enough strength to hit net energy gain in a fusion reaction. This is maybe an interesting theoretical paper and a good thesis project for some grad students or post-docs, but it's not really advancing fusion power research.