Yeah, and in a way it's even worse than that, since there's another layer of cognitive illusion: "It" doesn't exist.
The LLM algorithm is an ego-less document-generator, often applied to growing a document that resembles dialogue between two fictional characters.
So when your human-user character is "asking" the AI assistant character to explain its intentions, that's the same as asking a Count Dracula character to describe what it "really feels like" to become a cloud of bats.
You'll see something interesting, but it'll be what fits trained story-patterns rather than what any mind introspects or perceives.
It seems the key here isn't—or shouldn't be—what kind of service the defendant used, but whether something special happens when a service is involved in preparing a message to his lawyer.
IMO if the "for my lawyer" purpose/intent is not in dispute, then it shouldn't matter whether the service is a search-engine, an LLM, a browser-based word processor, or the drafts/sent folders of a webmail client.
The reverse direction is much clearer: Imagine a client receives an obviously-privileged email from their lawyer, and uses a cloud text-to-speech service to listen to it. Should that audio/text be admissible as evidence? Hell no.
Hold up, there are two very different kinds of "free" that pop up, from mutually incompatible ends of a spectrum.
In the first, it's a "free" market because everything moves smoothly and efficiently in ways that make economic mathematicians happy. There's perfect information, all transactions and prices and actor-identities are are known, and it's generally assumed there are no monopolies or monopsonies.
In the second, individual actors are "free" to construct secret deals with special secret prices, non-disclosure agreements, act anonymously through proxies, etc.
At least in US politics, the equivocation is common: Someone appeals to the benefits of the first, while promoting the contradictory mechanics of the second.
Pish tosh, my dear sir, it's simply common-sense that there are oodles of people out there with secrets that would be completely ethical to distribute and would undeniably better all humankind, but they're sitting on them purely because they haven't figured out how to make a profit from it. /s
In other words, the overlap between these is too small to justify the idea that prediction markets are a net-benefit by default:
1. Is valuable
2. Not already known
3. No current reward mechanism exists (e.g. patents)
> I suggest that you think rationally about what you just said. Why wound anyone
That doesn't read like a good-faith response. Parent-poster brought up an archetypal scenario asserting that certain people are ignoring malice/harms because they are not personally affected, exhibiting a latent logical contradiction or double-standard.
Instead of addressing the actual point (the existence/seriousness of the harm category) you've begun nitpicking that there aren't enough irrelevant operational details about the hypothetical arsonist and hypothetical attack.
_______
"They are correct that rain is natural and normal, this isn't a big deal."
"If it was your house being flooded by neighbors' failure to control runoff flooding created by impermeable construction, you wouldn't say that, you would recognize the harm!"
"Pish tosh! I suggest you think rationally about the fact that my house is on a hill! For what possible reason would I put myself in such a situation?"
The existence and severity of the harm, as argued by OP, relies on incentives. But incentives don't exist on a vacuum, nor are they individually absolute over human behavior. Simply stating that something has a bad incentive and that'll result in evil, on its own, isn't much better than assuming spherical cows.
The reason I nitpicked on the arsonist example is to elaborate on that: the arsonist does indeed have an incentive to put fire onto someone's house due to the bet, but the bet itself increases the probability of the fire being prevented, the culprit being caught, etc., which are all disincentives that may negate the original misaligned incentive. When we escape from this particular scenario onto others, this pattern remains. If the disincentives didn't prevent said harm, there are only two causes:
- culprit is unreasonable beyond saving: this one is trivial - they are beyond saving. Banning gambling would simply cause them to commit crimes in a different way;
- the mechanisms behind the disincentives are malfunctioning or aren't enough: that warrants an investigation onto said mechanisms, not a ban on prediction markets.
Perhaps this has made it clearer that, at least from my perspective, these "irrelevant details" aren't irrelevant.
> bettors are waging using their own property [...] To prohitbit such a thing is to violate their property rights
This is technically true but doesn't deserve top-billing. Any fraudster or embezzler (or mugger or drunk-driver) will incidentally be risking some of their own property while exercising their right to control it.
> informed decisions about an event by proof-of-stake
A large chunk of the problem here occurs when people (politicians, judges, police, CEOs, etc.) are wagering assets and outcomes that aren't actually theirs, but things they control in (violated) trust. In other words, the personal "stake" of their overt bet is actually far too small.
> The latter is functionally the same as policemen/judges/prosecutors/prisons having an incentive to create more criminals. [...] we increase punishments
So... repercussions like "is is a crime for those people to possess a private account on the anonymous bribery website"?
In regards to first point, I might not have made it clear, but betting, on its own, does not harm others, contrary to mugging or hitting a car. Property rights come with the associated duty of respecting others' properties afterall. Failure to do will result in the duty to repair and pay to restore the "correct" state to best of their abilities.
Onto the second, I don't think the people you mentioned actually wager stuff that they don't own. To be more precise, the problem is that they use criminal means to "rig" the bet in their favor. That, at least, seems to be what OP is talking about with incentives.
Finally, regarding the last, I agree that roles that require trust should have more norms and rules to enforce that trust. In case of strictly private roles, I personally believe they should be done via contracts and cultural pressures. For governmental roles, they should be enforce by laws (such as the one you suggested). Since this is a top-level, abstract, description, we can get into more about more specific cases (such as CEOs).
Worse: It won't even be your own bet that does the damage.
It'll be bets collected (or "lost" bets which are actually payments) by other parties that have an interest in an outcome which involves whether you lose your house.
Yeah, and in a way it's even worse than that, since there's another layer of cognitive illusion: "It" doesn't exist.
The LLM algorithm is an ego-less document-generator, often applied to growing a document that resembles dialogue between two fictional characters.
So when your human-user character is "asking" the AI assistant character to explain its intentions, that's the same as asking a Count Dracula character to describe what it "really feels like" to become a cloud of bats.
You'll see something interesting, but it'll be what fits trained story-patterns rather than what any mind introspects or perceives.
reply