That's fair, and I genuinely wasn't trying to nitpick, it is a very good question. If I try to answer that question as written, I'd say that any time I see a probability estimate with on-the-order-of-hundreds of zeroes, when I know that event actually happened (at least) once in Earth's past light cone, I'm going to assume there is an error in the model that generated the estimate, whether I know what it is or not. So what I way trying to point to is that if a catalytic cycle of many (much smaller) RNA strands was sufficient for an abiogenesis event, that could lower the probability estimate enough to make such events more likely by enough that there could have been multiple even just on Earth without straining credulity, and the world today would likely look basically the same either way since the more-competitive biochemistry would have long since reach fixation (and/or the lineages could have merged in some analog of later endosymbiosis events).
I'm not aware of an argument that there was only on abiogenesis event on Earth, just the observation that all known surviving lineages come from a universal common ancestor fairly early on. In principle that would be compatible with any number of initial events. It's just that once a given lineage evolved enough adaptions/improvements, it would spread and take over, and then no new lineage would be able to compete/get started.
Also, your scale for probability seems to be starting from assuming a single long self-replicating genome, but that isn't strictly necessary to bootstrap the evolution of a basic self-replicating metabolism. There are much shorter RNA strands (<200 base pairs) that have some catalytic activity including synthesizing additional RNA (though not copying themselves, AFAIK). Something like that could locally generate large numbers of shorter RNA strands, many with some form of catalytic activity of their own, collectively comprising some form of catalytic cycle that includes making more of all of them. Such a system would also be better able to cope with low copying fidelity b/c the individual strands that need to be copied correctly are shorter.
As far as going from bare RNA to a bacterium, I admit I don't know how this happen(ed? happens?). My naive initial thought is some RNA arising in this environment that could produce fatty acids, which could form a lipid layer around a cluster of RNA molecules spontaneously. Repeat and replicate enough times, and add in some endosymbiosis events and you're not too far off?
Which kind of impossible-to-solve do you think alignment is, and why?
Do you mean that there literally isn't any one of the countably infinite set of bit strings that could run as a program on any mathematically possible piece of computing hardware that would "count" as both superintelligent and aligned? That... just seems like a mathematically implausible prior. Even if any particular program is aligned with probability zero, there could still be infinitely many aligned superintelligences "out there" in mind design space.
Note: if you're saying the concept of "aligned" is itself confused to the point of impossibility, well, I'd agree that I'm at least sure my current concept of alignment is that confused if I push it far enough, but it does not seem to be the case that there are no physically possible futures I could care about and consider successful outcomes for humanity, so it should be possible to repair said concept.
Do you mean there is no way to physically instantiate such a device? Like, it would require types of matter that don't exist, or numbers of atoms so large they'd collapse into a black hole, or so much power that no Kardashev I or II civ could operate it? Again, I find that implausible on the grounds that all the humans combined are made of normal atoms, weigh on the order of a billion tons, and consume on the order of a terrawatt of chemical energy in the form of food, but I'd be interested in any discussions of this question.
Do you mean it's just highly unlikely that humans will successfully find and implement any of the possible safe designs? Then assuming impossibility would seem to make this even more likely, self-fulfilling-prophecy style, no? Isn't trying to fix this problem the whole point of alignment research?
These are my first two thoughts as well (although I think the second is partly a subset of the first - many great inventors had teams of unnamed helpers, or were just the last and luckiest of a long line of forgotten inventors who didn't quite get there, or both).
My third thought is - maybe it's just really hard for most people to feel amazement, in a world so filled with wondrous things, when you don't yourself know how any of it works? Like, an LED is amazing compared to an incandescent lightbulb, but if you aren't armed with a good understanding of physics and/or chemistry, it's all just a light bulb. Landing a rocket on a barge is incredible, but if you're not familiar with the relevant engineering, does it seem that much more incredible than landing even part of the space shuttle on a runway? Joy in the merely real is hard without a Feynman-level understanding of "mere."
Sometimes I have to remind myself how amazing so many things are that I encounter everyday, let alone the things I see coming in the next handful of years. For me one of the few things that consistently induces wonder is the field of metamaterials. I've read enough papers to know at root how they work, but still, I now live in the world where physicists can make things they call "illusion devices" that can block or alter the transmission of light through open air! There are arrangements of pillars, or trees, or rocks, or tunnels that could make a building or ship (or city?) invisible to tsunamis and earthquakes! But to most people in my life, this just gets dismissed as magical thinking along with fusion and lots of other things they assume belong to fiction or the far future (a set which usually includes a large number of things that have already been known, done, or used industrially for decades).
I know at least one person whose doctor agreed they might want to avoid getting vaccinated for health reasons, but refused to put it in writing, even knowing they were going to lose their job if they didn't either get vaccinated or get an exemption from a mandate.
I'm not sure this is actually evidence. Or at least, it's only very weak evidence
Obviously, witnessing someone leave the simulation this way would be strong evidence, but anyone who themselves conducts the test wouldn't be around to report the result if it worked.
Alternatively, you have no way of knowing what fraction of the people you encounter are NPCs, for whom the phrase wouldn't do anything.
Plus, for you to experience a faultless simulation, where you can't detect you're in one, you would need to not become aware of other participants leaving the simulation. Plausibly, the easiest way to do that is to immediately insert a simulated replacement to fill in for anyone who leaves. (Although, if simulated NPC people are themselves conscious/sentient/sapient, a benevolent Matrix lord might respect their requests anyway - and create a real body elsewhere in addition to continuing their simulation here. (Other variant situations, like trying to use the code phrase to escape torture, might require a deeper change to the world to remove a person in a way no one notices).
For myself, I suspect my being in a simulation at all, if it's voluntary, would only happen if 1) conditions outside are worse than here, and/or 2) my death or sufficiently bad torture here would result in automatically leaving (restored to a recent save point of my mind-state when I would consider it in good condition). Relying on being able to pick up a code phrase and trust I'll be able to say it at the right time would be truly terrible UI design if it were the only way out.
A woodchuck would chuck as much wood as a woodchuck could chuck if a woodchuck could chuck wood.
For one thing, why would I believe any manufacturer that made this claim?
It's one thing for very old companies that already are known for longevity and also provide very long warranties, like, say, Le Crueset or Cutco. That's a believable signal built over generations.
But for anyone else, claiming to have increased durability is cheap talk unless it's accompanied by a long-duration and thorough warranty, and a reputation for not making the use of existing warranties unpleasantly difficult and frustrating, and some way of assuring me that that state of affairs will continue. (As a personal example: I have a GE gas stove bought in 2014, and whatever the situation was before, GE sold their appliance division to Haier in 2016, and customer warranty service got dramatically worse, as attested by several of the 11 customer service agents and 3 technicians I talked to in 2018 to get a replacement for a busted igniter (which is extremely easy to diagnose and fairly quick to replace, I knew what was wrong before the first call and the first customer service rep assured me the first tech would have the part with him when he arrived, but alas).) In any case, a credible signal of durability isn't just a design/engineering/QA expense, it's a deeper corporate infrastructure and organizational expense, and one that is really hard to believe in a world where companies constantly buy and sell divisions of each other, and change executives and strategies, and deal with constant external shocks of various sorts.
Actually, there is one way I can think of top-of-mind that might convince me a company really had engineered for longevity: if the product automatically came with a fully prepaid long-term replacement cost coverage insurance policy from a highly regarded, long-lived third party insurance company. Maybe one that states clearly that if the product breaks (other than usual act-of-god etc. exceptions), and can't be fixed within, say, three attempts over a one month period, the policy pays out, no other exceptions. And it needs to be transferrable - what's the use of a long-lived stove if the next owner of my house can't benefit from the policy? For an actually durable product, this would be cheap to implement, otherwise not so much.
I have the same problem with clothing, but worse. Sam Vimes was wrong, or would be in the modern world. My clothes last about the same number of wash cycles, and have about the same chance of coming clean when I spill things on them, whether I buy a brand of shirt that has a $10 MSRP or a $100 MSRP. And with manufacturing so much cheaper than labor these days, repair of an expensive item, like resoling a shoe, can easily be several times more expensive than an entirely new but cheaper shoe. Especially if I am patient and wait a few weeks or months to buy it on sale when it predictably goes on sale. Actually, I buy most of my t-shirts these days at Michaels (the craft store): comfortable, no logos or tags, wide range of color options, cost less than $4, similar lifespan to every other shirt I have bought in the last decade.
In my work my team members and clients are scattered across Boston, NY, Texas, California, Amsterdam, and Singapore anyway. Most of my work is during normal business hours, but I regularly have meetings as early as 7 or as late as 9-10. Not having to commute, and being able to easily shift hours without worrying about bus schedules, or being able to split my day and take time in the middle to run errands and do chores, is a huge boost in productivity. That said, I am in your category of an experienced employee who is mostly engaged in reading and writing.
Informally, one thing I've noticed is that some people, especially some more junior team members, seem more willing to speak up in (especially internal) group meetings over zoom than in person. In meetings with clients, zoom meetings also make it easier for one team member to Slack an idea or a reference or a data point over to whoever is presenting or speaking at the time, or to propose a question to ask, if they can't or don't want to speak up themselves.