I sincerely doubt this will work. Sure, they'll come up with something and advance some sort of new architecture, but mimicking the brain is going to take a ground-up new approach at a hardware level.

The human brain once you really start to research it is so far beyond our bare bones concepts of systems architecture. One might look at it's complexity as a continuous, non-stop R&D process that has emerged over hundreds of millions of years, but unlike human based R&D projects it is fully capable of trying any and all permutations at once and nips in the bud (through survival) any failed experiments or changes.

To put in actual English, which is insufficient for describing the vast complexity of the human brain, the human brain is a massive parallel processing system with dynamic, inter-changing connections, is capable of self-repair and cross wiring, combines both memory and processing into a single system, and while the brain is generally a centralized system of neurological operation, quite a few of it's functions are distributed in highly specialized systems throughout the entire body.

But most importantly, all of our computer logic is built on a fundamental building block of the silicon transistor, which has an "on" and "off" state, 1s and 0s. Thus it's basic computing function consists of 2 states. A neuron on the other hand is an analog processor that can have up 10,000 concurrent variables operating at once. At any given millisecond, every neuron you have is receiving several thousand inputs simultaneously. When computers store information in bits, neurons store by adjusting the connection strength of individual synapses. Thus while a transistor stores a 1 or 0 and can be connected to 2 to 3 other transistors each storing ones or zeros creating data, a single neuron across it's 10,000 connections can have up to 26 distinct synaptic "weights" or states of being. Neurons integrate chemical gradients, electrical pulses, and physical structural changes that they can make on their own. In essence a single transistor can store 1 bit of information, whereas a single neuron can store roughly 4.5 bits per synapse x 10,000 synapses per neuron, and it uses that storage as processing as well.

I don't think it's possible to mimic the human brain with our foundational concept of silicon transistor architecture. People have tried this on neuromorphic chips and while you can kind-of manage it, the most we've found with these experimental chips is just how little we understand about how it works.

This concept of releasing sterilizing males to control insect population goes back to the 50's; it's not a new science. What failed before was the sorting process to find the correctly bred insects at sufficient volume to make a dent in the process. Verily, Google's life science entity, leveraged Google's expertise in robotics, computer vision, AI and industrial automation to build a system that could identify and sort the correctly bred mosquitos at scale so they could be bred at a high enough volume to make this strategy practical.

In essence the real issue isn't about the science; that was solved. The problem was scale and sorting through large volumes of produced eggs (like data on the internet) to return sufficient results in volume and reasonable cost, so this actually leverages Google's technology focus over the last 10-15 years to solve the key bottleneck.

It won't do anything. This has been studied extensively.

Adeptus aegypti, the mosquito that carries so many diseases and tends to bite humans a lot, is an invasive species all over the world; removing it would actually create more space for native species. The predators that feed on Adeptus aegypti like bats, birds and dragonflies are generalist predators; they can eat other things. Adeptus aegypti is actually a poor pollinator, so this would create space for better pollinators like bees and butterflies. And there are 3,500 species of mosquito known; most fill the same ecological niche but only a dozen or so tend to carry the bad diseases, and only Adeptus aegypti prefers humans over animals for biting.

This has been studied extensively, and by all science we have the world could do without this species and get along fine. And these "sterilizing male" strategies are highly targeted to just the one species so they are quite good solutions.

I know I'll get modded down for this given the anti-capitalist slant here, but no it doesn't.

1) Chasing profit isn't a bad thing, it's the actual correct thing. Building a business that can provide a new useful function to society and doing it profitably means it can provide that valuable capability without investor or government welfare. Profit is the goal, because it means the value it provides is sustainable.

2) Being a public company doesn't make you beholden to shareholders. It already is beholden to shareholders. It's founders, Dario and Daniela Amodai, are the primary shareholders. Several private equity funds like Altimeter, Dragoneer, Sequoia Capital, Fideilty, and others are major shareholders, as well as Google, Amazon and NVidia. Those entities are already shareholders and already have influence on the company. The IPO document is private but it's expected they'll sell 5-10% of the company, so any new public investors won't have much voting power even as a collective class.

Rather what being public does is it raises the standards by which they account for things and have to disclose, in exchange for access to public markets which have roughly 25X as much funding as private equity markets. So everyone will know a lot more about Anthropic since they have to disclose a lot more. In the meantime for the current investors, owning shares that have a public market for them vs. owning shares that were all private equity changes the nature of their shares; they can borrow more easily against them raising money for other things, or sell the shares or deliver the shares they own directly to their own LPs, generating liquidity for those LPs. THat's a good thing; many of them will make liquid money on this deal, money they'll most likley recycle back into the venture and private equity ecosystem to enable more companies.

So ultimately it's a good thing, and the only real difference for Anthropic is now they have more regulations to live up to via the SEC, but it doesn't really change their governance in any way.

It is garbage. I use Copilot all the time. It's good at some things but not others.

I write a lot of papers, from marketing posts to technical papers. Copilot is not very good at creating the content, but it is very good at editing the content and making suggestions, even with PowerPoint slides. It's also very good at taking a rough idea and making a reasonable outline that I can hten turn into a first draft and have it review and edit. Coding? Not really; at best it can write a macro for excel for you but that's about it. But like all the tools out there, or any tool really, just know what it's good for and use it for what it's good for.

Callum Turner was great in Masters of Air, but the nerdy news is he's been cast as Case in an Apple TV series of Neuromancer. That's something to look forward to for sure.

Fermi isn't a startup. It's a Real Estate Investment Trust (REIT); basically a mutual fund but with real estate behind it rather than stocks or bonds. They did this because REITs offer a very significant benefit: they don't incur a corporate tax, instead all income gets distributed to shareholders directly for their own income purposes, and most importantly you can depreciate the value of the underlying real estate and use that as a deduction against the REIT, and pass that deduction on to the shareholders in the REIT. This is important, because real estate can have some pretty hefty deductions relative to their gross income generated, but depreciation is a non-cash expense. So in essence you can provide cash income that carries with it significant income deductions for tax purposes; in practice that means you get a cash dividend on the REIT that is taxable, and a "return of capital" which is a cash payment representing the deductions. I make that distinction because startups are usually associated with new technology of some form; this is more of a financial services company.

Their plan was to buy real estate to be dedicated for data centers, and paired with that they would provide "behind-the-meter" power; in essence they would not only lease the shell buildings as data centers but also sell power generated. They were talking about nuclear, but I don't think they were that stupid. A modern stick built nuclear reactor can take 5-8 years after a multi-year permitting process, so this just wasn't going to happen in any reasonable time. They were also planning on putting on-site natural gas and solar, which are vastly quicker to set up. So part of the distribution to investors was also in the form of energy sales direct to the client AI companies taking over.

The CEO was fired for cause per the Board's own statements, but the big issue is he had no clients and now 6 months after going public they still had no clients, so even if they started building they had no one to act as a customer.

So in the end, it's a simple act of a guy with good pitch to investors but bad at execution, with an idea that maybe on paper was good (leasing land with dedicated power to data center places), but came down to bad math. He was never going to be able to provide the 11GW he proposed without an insane amount of financing (estimates $120B was needed to build that out), he lost a deal with Amazon when he couldn't' give them confidence the power would be there, and he couldn't get favorable terms to make it happen. he should have bent over backwards to get Amazon in there as an anchor client to prove the model and go from there.

Your point is valid and a key problem with HSR in California. Where HSR excels is where the urban population density is high, like in China or Europe. The US has a far more suburban population, where point-to-point travel doesn't have the same economics. I was involved in a study looking at the United States Marine Highway Project, which is still ongoing. Given the US' massive amount of coastline and navigable rivers, in theory water transport should be the best because it's basically free train tracks; the water's already there. The problem came in with what's called drayage, or the last mile of transportation. No matter what with point to point transportation, you still need trucks and cars to get people to their final destination, and from a freight perspective it was still over 50% of the cost; you barely save anything from our current model right now.

California's issues are eminent domain and the cost of land, environmental laws, and litigation of all of this which is hugely expensive. But the primary issue is just that our population is highly distributed and in the end the economics just aren't there for point-to-point travel.

You must not have passed middle-school reading comprehension. I'm using Bren as an example, not to empathize with, but to show how someone like him will respond to this tax, and how the knock-on effects will lead to fewer businesses, lost tax revenue and people being laid off.

I know that the concept of unintended consequences might be foreign to a person such as yourself, but that doesn't change the fact that this tax will be more damaging to California's economy and the people in it than is proposed. That might be too complex for you though, so here's a ball for you to play with. Perhaps you'd like to bounce it.

Well we're about to, as it's poor people voting for this.

Very few so-called "billionaires" have billions of dollars in cash. Most of it is tied up in non-liquid ownership of their companies or their real-estate. Let's take on billionaire in California that can't "flee" like the tech billionaires, Donald Bren. Bren owns the Irvine Company completely; it's a private entity that owns around 125 million square feet of apartments, retail centers, and office buildings. Irvine Company is private, but his estimated net worth is $19.2B because of his ownership in the company and the real estate assets that underpin it; it's not cash.

There are two major problems with this. The first is "estimated" net worth; what is that? What is the value of his holdings? One could argue the value of the real-estate, but they fluctuate all the time due to supply and demand. If there was a sudden increase of real estate on the market, that would drop the value; what's it worth now?

And why would there be a sudden increase of property on the market? Simple, because of the second problem. If we assume his net worth is $19.2B, then he would need to pay $960M. He doesn't have that cash on hand, so he has to do one of two things: sell property in his portfolio (many of the targeted people for this tax are real estate developers/owners), increasing the supply of real estate on the market and depressing prices all around for everyone, or increase his rents to his tenants, who are all businesses who employ people. If the rent goes up, expenses go up in all businesses, meaning they will then lay people off or pass on the higher expenses in prices to consumers of their products. Most likely they will do both.

And absolutely believe these billionaires, who can directly influence companies, prices, and the like, will make any tax pain they feel punitive to all California residents, and for one simple reason: the tax is only one time, until the next time. If it can get voted through once, it'll be voted through again because the ballot initiative is clear, it's a "one-time tax paid over 5 years" to pay for "healthcare and public services" which are by nature on-going expenses; when the 5 years are up those expenses will still be there, requiring another 5% asset tax.

So you'll see companies lay people off, you'll see services go down, you'll see companies close, because the billionaires will have to sell assets, which are always companies that employ people, just to pay the tax. Hence, because of poor people, poor people are going to be laid off.

Actually, China is a signatory to the Berne Agreement (As of 1992) and the TRIPS Agreement when they joined the WTO. They are required by those agreements to follow the injunction according to their own domestic law to enforce it. I don't know what that means in CHina's specific case, but if they have laws protecting Chinese entities copyright, they must apply the same protections to the US entities' copyright. Now China may not do that, but then they would be in violation of the treaties they signed, meaning the other member countries could immediately allow for copyright infringement of Chinese entities' copyrights in their nation, so China would have to be very selective about how or why they would not follow the injunction.

So that's factually incorrect. Sweden IS a member of the Berne Convention, and they are a WTO member and the TRIPS Agreement requires a certain minimum set of standards of enforcement of copyright including banning domains that are actively engaged in the distribution of copyrighted material and providing for monetary damages for said distribution, and the material meets the definition of copyright because it is copyrighted in the US and the Berne Agreement requires them to recognize that.

Hello Sunday morning lawyer!

While a US judge does not directly have the power to order other countries around for an injunction, it doesn't mean that this injunction can't be forced, and it also doesn't mean that just being in another country leaves you free to just steal any copyrighted material from another country with no consequence. In fact, the Berne Convention goes back to 1886 and has 180 national member signatories that says that every member country must extend the same copyright protections to the creations of those in other nations as though they were their own citizens. Essentially it means that all signatory members must automatically recognize any US copyright protection, so by the judge ordering this injunction and stating that Anna's Archive violated copyright, by extension every member nation of the Berne Agreement recognizes the copyright of all of the plaintiffs and in essence would recognize that Anna's Archive violated their copyrights.

Further under the TRIPS Agreement in the WTO (1994), all members must provide the same copyright protections to foreign citizens that they would provide to their own citizens, along with a minimum set of standards such as no illegal copying and distribution of copyrighted works.

By ordering this injunction, he's declaring that under US Law Anna's Archive violated the US entities' copyrights, and under Berne and TRIPS at a minimum (and probably other treaties), every member nation of the WTO and the Berne Agreement would provide the same protections to those US entities against illegal copying.

So in effect, yes, he can order a worldwide injunction to at least member states of the Berne Convention and the WTO because all of those nations would provide the same legal protection. The enforcement may vary from country to country, but none allow for illegal distribution of copyrighted materials.

Totally different business but exactly the same problem. Nordstrom generally has the latest trend clothes in fashion and pretty good quality; it's known for it. But when it had leftover inventory it knew there were people a step down from their target demographic that would love Nordstrom's quality products even if they're a season or two out of fashion for cheaper, so they opened Nordstrom's Rack to sell off the excess inventory.

Nordstrom's Rack got so popular they couldn't keep it stocked, and eventually started developing their own dedicated Nordstrom Rack brands, which sort of defeated the purpose of Nordstrom's Rack as it's entire value was Nordstrom's quality, late season, at a discount, but now it's discount quality with the Nordstrom's name on it.

Law of unintended consequences I guess.

I spent my first working years in retail finance and financial planning. Generally people contributing to 401ks are already earning well; on average people who earn over $150k contributed around 15% to them, and overall contribution goes down the lower income you go. You also see white and Asian demographics contributing the most, whereas Hispanic and African American contributing the least. In general they have worked as intended, encouraging workers to save up for retirement, but they have had 2 unintended consequences that create an opportunity.

1) 401ks indirectly contributed, and i would argue was the single largest contributor, to the growth of private equity and asset classes like venture capital, which led to an explosion of startups and innovation. It works like this. VCs generally get the money they invest in startups from either wealthy families (family offices) or very large money managers like pension funds, insurance companies, sovereign wealth, and the like. Pension funds, from a dollar perspective, are around 85% of all funds going into Venture capital, so in essence workers working every day and contributing to pension funds for their later pensions are a key contributer. However 401ks did as well, because pension funds generally are required to allocate assets according to some strategy. It's usually something like 30% bonds (US treasuries), 30% blue chip stocks (DOW, S&P 500, NASDAQ Index funds), and then some other areas like foreign stocks, foreign bonds, etc. Virtually every one has at the bottom 1-2% in "high-risk, high return asset classes"; this is usually VCs. What happens is 401k contributions also buy into usually index funds, like the S&P 500, and that also happens every pay period. That raises the value of those funds (and the stocks, it's great to be on the S&P 500). As such, at the end fo the year, maybe the allocation is now 30% bonds, 33% stock, other stuff, and now .6% high risk, high return assets. So they sell out of the stock classes to reallocate, and more money gets allocated to give to VCs. The constant influx of worker's capital into pension funds and 401ks that go into index funds that raise the value of those funds directly leads to reallocation of money to venture, leading to more startups getting funded.

2) 401ks end up being a bit of a burden when you're retired. While some people use them for sure as an income source, because your gains are taxed as income, if you have other sources of income you want to draw on this as little as possible. As such, Rollover IRA money, from the 401k, just sits there in an asset class, not being used. Many of the clients I helped saw their 401k RMD as a burden; they never needed it but were required to do it.

And let's add one more element here. Private equity, and in particular venture capital, has much longer market cycles from the public markets, so over a 5 year horizon might not beat the public markets, but over 20 years the returns will crush the markets. For example the last 5 years VC has struggled to compete with public stocks, but over a 20 year horizon the category has achieved 14.3% CAGR compared to say the S&P 500's 7.5%. This means over a 20 year horizon, $1,000 would become $15,533 when in VC vs. $4,247 in S&P 500; roughly 3X the gain.

I say all that because 401k money is long-term horizon. Gen Z appears to be good savers and is contributing well, but they're not touching the money for 30 to 40 years. That makes the capital patient, and with a long-term horizon looks very attractive. And given the size of 401k contributions, this would create a whole new pool of funds available to invest in the next generation of startups and keep the innovation cycle moving.

I'm not a fan of crypto, but i have to admit my issues with it have led me to miss out on the upside. But opening up the US 401k market, even just a small fraction such as 3 or 5%, would create an influx of capital available for VCs and as such for startups, creating a whole new generation of companies. And 401k money is great for this because it's long-term horizon and patient.