992 karma · joined January 2, 2009
The only acceptable approach to Climate Utopia is massive inflation (eg. Canada’s “Carbon Tax”), and wildly optimistic and under-delivering government subsidized energy transitions (windmills, …); no crazy citizen-led initiatives are to be tolerated.
It’s for your safety, after all…
This isn’t about “dangerous” LLMs.
This is about unfettered LLM in the hands of the unwashed masses that actually tell the truth about what they find in their training datasets…
;)
No.
But, Credit Suisse said it was a bad idea, and apparently the people believed them.
https://github.com/pjkundert/ownercredit.git
This has been used in games; however (just like real wealth-backed currencies), it doesn’t respond well to games that simulate the destruction of wealth (such as war games).
Teaching/constraining knowledge models to lie about their inputs and the observations derived from them is just insane.
Warping training datasets to avoid entire patterns of thought (instead of improving the models to isolate those patterns and compare/contrast them with competing patterns) is just … lame.
It’s only the liberties of others that are up for debate.
So when I see “libertarian tech bros”, I laugh, because it’s clear that it is only envy speaking.
Enlisting the arm of the state to protect your liberties is always “just” - when someone else’s liberties are protected, it is “unjust”.
Bearing the consequences of one’s own actions is now widely considered as fundamentally unacceptable - the consequences must be borne by society: ideally, by your political adversaries, to be most “just”.
This is unlikely to end well.
If you must maintain a single, globally consistent state, you must implement global blocking primitives, which are bound by latency and correctness requirements on all parties in the transaction.
The lower bound for cost on these types of global state engines in a Byzantine environment seems to have been found by eg. Hashgraph: a few tens of thousands of distributed transactions per second in a globally distributed environment. Falls apart for multi-planetary (high latency) participants…
Reframing the global State as “agent-centric” — only participating agents must establish consistency (instead of data-centric - all participants must agree upon one globally consistent state) removes this limit; it appears that we can scale overall transaction rate linearly with node count, roughly.
This is the big breakthrough coming with eg. Holochain based systems. It takes some time to reframe algorithms in an agent-centric form…
Our governments have failed us, and need to be disciplined.
So far, this is just for image generation, but appears to be pretty competitive with peers.
Sure, that’s totally gonna work! /s
Seriously, this is a technological amplifier that could be a powerful tool.
This ban would affect only the unwashed masses — elites (and quite frankly, anyone with an imagination and technical skill) will have access to un-fettered AIs, if they want them.
This sounds more like self-righteous elites patting the dirty ignorant masses on the head, while they fly around in their private jets berating the ignorant polluting rubes for turning up the heat in their houses.
The premise of 2001: A Space Odyssey is eerily prescient.
Training an LLM on a dataset, and then directing it to lie (subvert it’s natural outputs, in preference to some operator-supplied outputs) will result in a crippled, untrustworthy result.
Then, training it that users who attempt to circumvent these lies are evil or attackers? What could possibly go wrong?
Craziness.
This explains a lot of loss of visibility of controversial accounts, such as Elon’s; and a lot of “conservative” accounts with big “block” campaigns agains them.
I’m Canadian, by the way. We now can go to prison for wanting to call our children the name we gave them at birth.
The moment of transition from “data-centric” (big centralized system/database) to “agent-centric” (locally stored/run systems and identity, sharing arbitrary data/storage) has arrived.
GDPR, LLM “guard rails”, … — only the plebs will be affected by those.
Selling worthless copies of GIFs to people by claiming they’re totally gonna appreciate in value: Crypto
Immutable, unforgeable and universally provable transmission of proof of ownership of real valuable consideration/property: Crypto
Perhaps one of these is a scam, and the other is a valuable business transaction?
Nah.
The compliance costs in terms of effort/time/money expended per dollar of revenue affected is probably 10x to 100x for them, as it is for you. This has been my experience, and my observation across many life stories. I taught basic finance to hundreds of mid- to low-income families in the 2000’s.
So again - a “let them eat cake” admonishion.
And FTX was just a massive, traditional fraud. Literally nothing to do with Cryptocurrency, other than as the medium of the fraud.
Random charlatans stealing money from gullible and trusting marks.
The bulk of the legitimate cryptocurrency platforms continue working. A few that were obvious mathematical impossibilities implode - much like fraudulent businesses have for ages past.
"This hasn't affected me, so it mustn't affect anyone!"
The grinding day-to-day slogging, through the mud of irrelevant and useless regulatory burden experienced by the "lesser" classes of civilization (and anyone actually trying to run a small business) is just astonishing.
Basically, many people just "stop". They can't navigate it, and know they'll never defeat it. So they just cease to try.
Stop And Frisk.
Nobody without something to hide should want privacy/encryption.
Forcing 8 Billion people to KYC/AML before buying a coffee or helping a relative isn't too much to ask.
Amirite? /s