COVID-19 killed, idk, tens of millions worldwide rather than hundreds of millions.
But consider that an example of a (biological) virus takeoff of the order of months.
So the question for AGI takeoff .. death rate growing more rapidly than COVID19 pandemic, or slower?
Takeoff speed could be measured by e.g. the time between the first mass casualty incident that kills thousands of people vs the first mass casualty incident that kills hundreds of millions.
(This bit isn’t serious) “i mean, a days long takeoff leaves you will loads of time for the hypersonic missiles to destroy all of Meta’s datacenters.”
Minutes long takeoff...
[By comparison, I forget the reference but there is a paper estimating how quickly a computer virus could destroy most of the Internet. About 15 minutes, if I recall correctly.]
e.g. After the mass casuality incident...
“You told the government that you had a shutdown procedure, but you didnt, and hundreds of people died because you knowingly lied to the government.”
My personal view on how it might help:
Meta will probably carry on being as negligent as ever, even with sb1047
When/if the first mass casualty incident happens, sb1047 makes it easier for Meta to be successfully sued
After that.AI companies become more careful.
On the one hand, we encounter a lot of arguments about gender that seem, to me, to be philosophically bad. Maybe a good source of reasoning fallacies you might be able to spot in other contexts too,
On the other hand, the more I think about it, the less I care about the object level isue. It seems inevitable that there are going to be various sorts of statistical outliers and hard to classify cases, and really is it that big a deal?
I know one person who is intersex, and I know because they’re involved in right activism and they told me, Probably couldnt tell otherwise. Could well be xome other people I know are intersex and haven’t told me.Maybd they dont even know themselves, as it appears that this information was frequently witheld by doctors. Shrug.
Also: if you take gender as chromosomal sex, then (a) tye aforementioed person is totally genuinely both xx and xy because they have mosaic chromosomes; and (b) it seems really strange for your gender to sometimes be something that you, yourself, do not know,.
A financial conflict of interest is a wonderous thing...
“Okay, Beatrice. There was no alien, and the flash of light you saw in the sky wasn’t a UFO. Swamp gas from a weather balloon was trapped in a thermal pocket and refracted the light from Venus—Men in Black
The TV Series “Dark Skies” .. in which the US Government is orchestrating a coverup about the involvement of giant prawns from outer space in the Roswell incident, the JFK assassination, the shootdown of Gary Power’s US spyplane, erc.
I agree that Vernor Vinge’s A Deepness in the Sky is an example.
Almost but not quite an example: Edmund Cooper’s The Overman Culture. It is obvious to the reader from the outset that the characters cannot be when and where they think the are (evacuated from London during World War 2).Maybe not enough deceiver’s perspective to count.
Also not quite: Gene Wolfe’s The Book of the New Sun.
This is pretty much why many people thought that the term “Open Source” was a betrayal of the objectives of the Free Software movement,
“Free as in free speech, not free bewr” has implication that “well, you can read the source” lacks.
Yeah, many of the issues are the same:
*RLHF can be jail broken with prompts, so you can get it to tell you a sexy story or a recipe for methamphetamine. If we ever get to a point where LLMs know truly dangerous things, they’ll tell you those, too.
*Open source weights are fundamentally insecure, because you can finetune out the guardrails. Sexy stories, meth, or whatever.
The good thing about the War on Horny
probably doesnt really matter, so not much harm done when people get LLMx to write porn
Turns out, lots of people want to read porn (surprise! who would have guessed?) so there are lots of attackers trying to bypass the guardrails
This gives us good advance warning that the guardrails are worthless
Also note that Open Source precludes doing this …
The basic Open SOurce deal is that absolutely anyone can take the product and do whatever they like with it, without paying the supplier anything.
So
The vendor cannot prevent the customer doing something bad with the product (If there is a line of code that says “dont do this bad thing”, then the customer can just delete it
The vendor also cannot charge the customer an insurance premium base on how likely the customer is to do something ba with the product
… which would suggest that Open Source is only viable in areas where there isn’t much third party liability.
With a nod to the recent Crowdstrike incident …. if your AI is sending out packets to other people;s Windows systems, and bricking them about as fast it can send packets through its ethernet interface, your liability may be expanding rapidly. An additional billion dollars for each hour you dont shut it down sounds possible.
If your AI is doing something that’s causing harm to third parties that you are legally liable for .. chances are, whatever it is doing, it is doing it at Internet speeds, and even small delays are going to be very, very expensive.
I am imagining that all the people who got harmed after the first minute or so after the AI went rogue are going to be pointing at SB1047 to argue that you are negligent, and therefore liable for whatever bad thing it did.
If quantum computers really work, for more than 3 qbits, then I think I will believe in infinite worlds interpretation.
On the other hand, if there turns out to be some fundamental reason why quantum omputers with many qbits cant exist then maybe not.
The version where you only have 3 qbits is kind of unsatisfactory (look, there are exactly 8 parallel universes and no more...)
If American citizens who can vote from Trump are arguing over whether hr’s a bad guy, there is arguably a point to it .. though I can also the case against.
But Swedish guys … who arent even allowed to vote for Trump if they want to .. aruing over Trump? What are you doing?
(Presumably, if there is any point at all, the argument is not against Trump specifically but the global phenomenon he is part of, and whoever the Swedish Trump equivalent is).
To be truly dangerous, an AI would typically need to have (a) lack of alignment (b) be smart enough to cause harm
Lack of alignment is now old news. The warning shot is, presumably, when an example of (b) happens and we realise that both component pieces exist.
I am given to understand that in firearms training, they say “no such thing as a warning shot”.
By rough analogy—envisage an AI warning shot as being something that only fails to be lethal because the guy missed.
An organisation such as the Catholic Church primarily wants to perpetuate its own existence, so of course the official doctrine is that they are The One True Church.
An individual Catholic, on the other hand, might genuinely believe that the benefits of religion are also available from other suppliers.