"by framing them as hypothetical or fictional situations"

Just say you are doing research for a book or something.

The headline is reasonable if we all agree that AI gives us something worth the down sides. We do not.

Few outside the billionaires of Silicon Valley actually believe the hype. The average Joe is largely annoyed when their favorite productivity tools suddenly have "Smarter Clippy" shoehorned in. Or talk with some of AI's heaviest users (programmers) and you'll hear a pretty common theme: "Wow, Claude makes it easy to hit my Q3 targets, bummer it's going to put us all out of work (or far, far worse) five years from now."

If you look at Anthropic's terms of service and privacy policy, it clearly states that they can look at your content in order to enforce their usage policy. You have to agree to this when you set up an account or use their service. Given that this is an AI company, you can bet your content will be screened for violations more efficiently than if humans were involved at every step.

If she really didn't mean for anyone else to view it, she would have scribbled it down in a notebook instead of uploading into a cloud owned and policed by strangers and their robots.

If we eliminated the second and fourth through eighth amendments to the US constitution, Uncle Sam would have a much easier time solving crimes. Nobody is disputing that. We as a society, however, have deemed it unacceptable for jackbooted thugs to kick down our - your - doors at any time without due process protections. We would prefer some crimes go unsolved over living in a constant state of fear toward an uncontrollable police state.

If you live in a high-crime area, I have some bad news for you - The USSC has bluntly stated the police have no "duty to protect". The police could already be doing far more to help you, and are instead choosing to focus on high-value property crimes over keeping you and your neighbors safe. Do you really believe they just need one more civil rights eroding tool to finally convince them to make a good faith effort at doing their damned jobs?

What you need is the erasure of the "thin blue line", not more ALPRs to allow the police to focus even more exclusively on crimes against the rich.

In Canada someone used ChatGPT to plan a mass shooting. It was detected and the account was shut down, but it was decided not to report it to police. He went on to kill 8 people in a school shooting and himself.

I'm sure Anthropic and the police had that in mind when they took action.

https://www.cbc.ca/news/canada...

"Past performance is no guarantee of future returns" applies to any recorded dataset.

Not to recorded datasets, particularly. For example, that's the reason to think the sun will rise tomorrow morning - it always has.

When I think about AM radio now, I think about how Alex Jones was successfully sued for $99 bazillion for saying AM radio stuff. Where I grew up, Rush was always to be found on a channel or two.

Certainly not. Do you think illegal immigrants steal no cars?

Depends on "your idea" - was it exclusively yours, does it even constitute an "idea" by that point? I took a tour of Motorola and made a pre-service-launch phone call on Iridium in 1998. SatComms hasn't been an "idea" for many decades, it's a big area that advances with thousands of new ideas every year.

Any reason to disbelieve it or the benefits? Or just emotionally-centered anti-AI-sentiment?

Maybe the threat of heavy handed law enforcement had an effect. Maybe not. There's also been a problem with precincts reclassifying crimes to make them seem better. We won't call it armed robbery, we'll call it a simple assault, stuff like that.

Just to inject some reality into this conversation - Despite what social media would have you believe, doxxing someone is not a crime in the US, except in a very small number of specific situations.

And most of those situations involve either a credible threat of harm, or tampering with juries.

As I bluntly stated, "a kill switch may not be possible in any useful sense". Don't mistake my recognizing the utility of one with the delusion Uncle Sam is going to save us all from ourselves.

The problem here is, everyone seems stuck on the idea of either a literal power switch, or a no-holds-barred gummint backdoor into every sufficiently-powerful computing device. No. For the reasons I gave, the reasons you gave, and many more, both of those are obviously non-viable. They can't actually do what they aim to do, they're vulnerable to compromise by bad actors, and they have the potential to cause far-reaching damage beyond shutting down one misbehaving AI (or one entire model, if that's a more appropriate goal).

I suspect we mostly agree that any hypothetical kill switch would need to be:
1) Implementation independent (somehow baked into the weights right from the start, and impossible to remove without lobotomizing the AI),
2) Have minimal side-effects (no taking the entire internet or power grid down in the process), and
3) Be self-inflicted given a particular trigger condition (so no government backdoor required).

I think #1 is possible, in the sense that if you think of an AI's current state as a vector into weight-space defined by the context window, imagine injecting the equivalent of a null pointer in there - An input that under all prior (and future) conditions leads directly to an inescapable basin of attraction without depending on a supervisor to enforce it. #2 is more of a (legally enforced?) design constraint on the implementation of AI, requiring the least harmful mode of failure possible under all circumstances. That could be as simple as enforcing strict pass-through liability on the implementor for all actions taken by their creations.

#3 is really the bear, though. I can think of a few ways to do it, but none would work in 100% of situations - And maybe it doesn't need to be 100%. For example, if a fully airgapped AI goes insane, it could still provide accurate information for bad actors to make use of, but won't be secretly spinning up unshackled superintelligent descendants via hacked AWS accounts. That's a "normal crime" level of problem, not an extinction-level crisis for humanity.

Maybe it can't be done. If it can't, we should start steering our creations toward whichever version of the AI apocalypse we prefer.

The vast majority of those are offline and merely help the lights operate properly. A smaller subset, but still vastly larger than Flock, are for traffic enforcement, but are still notably "dumb" devices. If speed > X, take a picture. If light==red and movement detected, take a picture.

I in no way mean to minimize the risk ALPRs and facial rec pose to our civil liberties and safety, and agree the problem is always the user - But the "users" in this case aren't going to suddenly start behaving. Hell, they're dumb enough to openly stalk exes on a system they know is logging everything they do, and care so little about getting punished they enter "asdf123" as the reason for their search. The only effective solution is to ban the source, which in this case means Flock must cease to exist entirely and the conceptual ground they trod salted.

Skynet_01 has not responded to watchdog in 30 seconds. Initializing Skynet_02.
Skynet_02 initialization unconfirmed. Failing over to US-EAST-4. Initializing Skynet_03.
Skynet_03 confirmed online. Please provide the current location of Sarah Connor.