66 karma · joined March 28, 2013
Disappointed in the VC move but I don't have any complaints at the moment. I have had good experiences with Cloudflare for other services so feel predisposed to them.
> Anthropic is also quick to link this new research to its previous work on sycophancy, noting that “sycophantic validation” is “the most common mechanism for reality distortion potential.”
I commented on Ars that I see the same sort of mechanism at work in politics. So is the really an AI problem or a case of AI surfacing inherently human characteristics? It's a sincere question.
I don't think I'm alone here when I say I frequently gain understanding of things by writing code to model them. I know I understand something when I can write it in code. So after ingesting the lesson about a topic I'll see if I can write code to do what I just learned. I have spent a lot of hours stuck at a juncture of not being entirely sure my code, which works on the simple examples, is also working in the complex ones. The difference in results may not be evident. Am I troubleshooting my understanding or my code? I need a known result to actually know.
This is one place being able to copy something from your textbook, for example, paste it into ChatGPT and ask questions is a huge benefit. ChatGippity doesn't always get it right but usually the interchange is much more useful than being stuck.
Insurance rates are much higher because of the extremely high repair costs for even minor collisions. Consequently EVs are 'totaled' much more easily than equivalent hybrid or gas vehicles. Eastern Europe is doing a booming business in refurbished EVs shipped over from the US.
Then there are the issues of battery production and disposal which can do a lot to mitigate the environmental benefits of an EV.
These details, aside from charging, weren't readily apparent at first. But they are real and they are expensive.
What's changing is the scope, scale and access. It's no longer necessary to be a savant or have a special talent - all you need are hundreds of banal data points easily scooped from the meta data wake we all leave. Not to mention it can be programmed and commoditized instead of being a special attribute of another human. The realization feels a bit like climate change in that some folks saw it coming, pointed it out, articulated the issues, underestimated how fast it would actually arrive and how profound the impact would be.
But back to the point at hand you can't legislate people not see what is publicly before them. You might criminalize recording that data for a while but that doesn't seem to work long term thought it's brutal in the short term. If you can record that data you can analyze that data. From analysis comes inferences and decisions and actions.
I think the more salient question is about equality. I have some ability to obfuscate my meta data wake but as an individual it's pretty limited. Apple, google, Amazon have much more ability to do so despite leaving a gazillion more bits of data laying around. Tools individuals can use to obscure our meta data trail will do more to maintain privacy than trying to tell someone observing me what they can and can't do with their observations.
It was users.
I was living in Santa Fe in the '70s and was at that first screening, it was the world premier, at the Lensic Theater. I still have the stub. I was there because Godfrey Reggio came into the sandwich shop I was a partner in and because I was friends of friends of some of the guys on the camera crews shooting footage in the Southwest and working with Alton.
I recall it as a pivotal moment of my 22 year old life. It was like nothing I'd experienced. The lack of 'anything happening' lasted a few minutes. But I recall starting to 'get' the symbolism and the narrative the images convey. Much like the author I was floored by the music. I was a prog rock fan with an appreciation for classical music. Well, Beethoven at least, so the concept of a long form piece was engaging. The epic sweep of the imagery and music gripped me and held my attention throughout. Since then many of those images have come to mind at various times. And the final sequence of the Saturn V leaving the launch pad in slow motion coupled with the beautiful, haunting score to it blew me away in the moment and I can still recall its impact now.
Koyaanisqatsi is a rare work of art. The article brings me hope. The world is quite fucked up right now and while it's completely possible for us, all of us, to change things to mitigate the situations there is little cause to believe that will actually happen. Then I come across someone articulating such a profound resonance with a work actually closer to 50 years old and it sparks that hope.
From the business side you get a notice of the chargeback along with the statement the customer provides when they file it. You then have some period of time, a week or two, to respond. Businesses get scammed by charge-backs too, you know. So you have the opportunity to submit your paperwork showing why you are entitled to the funds. A surprising number of the times the business doesn't even bother.
The credit card company looks at what the customer says and what the business says. They may ask either party for clarification or more documentation. They always prefer the vendor and the customer to settle things themselves so they look to see if the customer reached out to the vendor for some help and resolution first. So if you have a problem you really want to do this first. Documenting emails, phone calls, chat-bot transcripts you, as the customer, have with the vendor is extremely useful here. What's also useful are screen shots of a website that doesn't offer any way to contact the vendor and initiate a refund.
They look to see what the vendor did to try to resolve the issue. Then they may look at the actual evidence. And they may have internal data, like a vendor has a lot of other complaints and charge backs for similar reasons. Or that a customer has a lot of charge backs that weren't upheld.
All things being equal the one with the better documentation generally prevails.
My personal experience is some cards are more likely to stick up for their customers. These tend to be the cards that pay high rewards because the customer pays them directly (the annual fee). Some, like USAA, seem to just be more oriented to helping their customers. But this is totally anecdotal.
One thing I've never heard of is anyone getting 'banned' for making a charge back. If someone has personally experienced this, not just heard it happened, please leave a comment.
And to what end? To be a dick? To piss off the false-positive customers your system inevitably flags incorrectly? There are reasons it's hard to have an accurate, effective no-fly list and that list costs a lot to maintain.
Plus OP has an excellent story to post as a review since as a 'paying' customer clearly has standing in the matter.
AirBnB has gotten so much bad press over their nit-picky extra charges, fees and required cleaning by guests I would not be surprised if they encourage the 'host' to reconsider their position.
Alternatively gin and tonic but for crap sake use a good gin.
Brainstorming: Tequila. Just try it. Margaritas if everyone is under 23.
Structure and Design: Red wine. Deep, rich, moody cabernets, old style zinfandels that roar out of the glass. This is where you pop for the expensive bottles so you do it right the first time.
There is an argument to be made for single malt scotch or 12 year old bourbon here. Either is a solid choice but avoid mixing them in the same session.
Bug Fixes: Beer. Hoppy micro-brews but avoid the high alcohol content ones. Also avoid the hazy, unfiltered fadish brews. You want to see and think straight and be sharp.
Dull, tedious refactoring 5 year old crap code: Rose. Hear me out- it's about creating a mind set here. Rose is just that - it's a mind set more than an actual drink.
Anything having to do with Mongo db: Grand Marnier, Ouzo, Pernod - something at least 80 proof with a strong flavor to compensate for the bitter taste in your mouth.
Bottoms up!