I ended up buying a used Lenovo Mini PC for $200 from a local seller, found on Facebook. It's an i7 something or other with 32GB RAM, I think I got pretty lucky with the find. Much cheaper than a Raspberry Pi considering I didn't need the pins anyway.
If you do need something like a raspberry pi I think the Arduino Uno Q is a solid choice. The 4 GB RAM variant is iirc $80 or so.
My work still forces password rotations. Everyone I know jokes about just changing the last character of their password each time, eg horse-staple-battery-1 to -2 to -3 etc.
Secretive should do this. It generates a key inside the secure enclave, and you can set it to require TouchID, so you know whenever it's used. Keys can't be exported either.
What I don't understand is how I can, allegedly (I've never tried), ask siri to dig through my photos and iMessages to look for stuff, but when I ask it to summarize the website on the screen it says it can only see literally what's on the screen and can't scroll down or grab the entire article that's already been loaded.
When my phone is sitting on my countertop and I say "hey siri, add eggs to my shopping list" I expect it to add eggs to my shopping list, as it has done so for years, instead of saying "you'll need to unlock your iPhone first."
When I'm driving my car merging onto the highway and I see furniture on the ramp, and I say "hey siri, call the local non emergency police department number" so that I can report it, I expect it to say something like "I have the [city name] non-emergency phone number, want me to call now?" Instead of it saying "calling the New Jersey police department" which is halfway across the continent from me, and then giving me no way to hang up the phone hands free forcing me to grab my phone from the console and quickly try to hang it up while I'm driving at highway speeds.
- When I'm driving and I say "Hey Siri, navigate to Walmart" I expect it to begin navigating and not show me a button which I must press to begin navigating.
- When an alarm is going off on my phone and I say "Hey Siri, turn off the alarm," my laptop or HomePod should not answer to say there is no alarm.
- When I say "Hey Siri, add eggs to my grocery list," my HomePod should not think for 20 seconds and then tell me to check my connection, despite both it and my phone being on very strong wifi.
- When I say "Hey Siri, add eggs to my shopping list," it should add them to my "Groceries" list and not tell me there is no list named "shopping."
Another example from yesterday: I’m using CarPlay. A text message notification pops up. I ask Siri to read the message. Siri says there are no recent messages. More notifications pop up. Siri still has no idea.
Yeah, the “new” Siri described in this article is not the Siri I’ve been interacting with. Basic speech to text is still flawed; I have to enunciate slowly and clearly with no background noise or it will make absurd mistakes that would be obvious if the context of the prompt and the phone were considered. Adding a reminder to a list is now slower and more error prone than with the old Siri, like you said. I have to sit and delicately hold the phone or watch to avoid disturbing it while it thinks, lest it decide that I’ve moved on and give up on its task. I tried to push it a little by asking for the title of the last podcast episode I listened to; it started talking about a podcast I’d never heard of nor even pulled up on the phone before garbling the second half of its sentence and then going silent.
I suspect it's regressed because they actually don't have good e2e evals comparing the two experiences to test the regression and UX. Like the simplest evals you want is just a shit ton of STTs across human speech accounting for dialects, accents etc and then figuring out how the distribution of errors has moved or increased on the axis you care about. That they likely have. But then you have to integrate that into the product to capture the entire funnel of errors, which I suspect is the part they don't have. The stacks are not "vertically integrated" to test quickly.
Dictation being so terrible is truly wild to me and I don't understand it. On my M5 MBA literally any of the vibe coded local dictation apps that come up (and there are like, hundreds of them!) are better than the built in macOS dictation. They are much faster and the AI text cleanup is so much better than whatever the Apple foundation models are doing.
I have a list in Apple/iCloud Reminders called, literally, "Shopping". Pre-Siri AI saying "hey siri add eggs to my shopping list" puts it into the correct list. Post-Siri AI saying the same thing gets a "you'll need to unlock your iPhone" response. This isn't specific to a "shopping list", it's just any named list in general. I have a list called "Chores" and "Errands" and saying "add get cash from the atm to my errands list" gets the same refusal.
Siri is not incapable of adding reminders when the phone is locked. A generic "remind me to buy eggs" gets a "buy eggs" item added to the default list. It just refuses to put them into specific lists, which then means I have to triage it later and i ain't got time for that.
> Instead of it saying "calling the New Jersey police department" which is halfway across the continent
Yeah, that's Apple's generally broken maps search. It regularly and consistently just brings up the wildest things from the farthest reaches of the Universe (well, usually restricted to Planet Earth, barely), when I am clearly looking for something local.
Stock trading isn't gambling. It's just owning assets that can change in value. Options trading is much closer to gambling, and I would say the biggest difference between it and Kalshi and similar gambling companies is that it's much more rigorously (if nowhere near perfectly) regulated.
Both have risk associated with them yes, one has oversight from the goverment and the other is a game / world event.
With stocks if I try to do insider trading I run the risk of Jail time. Or if I try to do a Pump and Dump scheme, I go to jail. If I am an person and I "place a bet on polymarket/kalshi" and then go and influence that outcome. It may not be illegal. Since these markets commodify literally any event, its impossible to control the items themselves. So the markets need to be controlled.
But oversight from the government could change tomorrow. The definition of something shouldn't hinge on what the government thinks it is. The government's definition should be based in something meaningful.
I could hedge it at the blackjack table too -- that doesn't make a blackjack hand a financial asset. I'm talking about whether the instrument itself is backed by something of inherent value.
You could buy a stock expecting asset appreciation, but the stock itself is a certificate of ownership in a business. Even a derivative of this, like options trading, is a contract to buy or sell or deliver said certificate of ownership in a business. Ultimately it is backed by something of tangible value.
"Predictions markets" are all of the mechanics of trading, but without any underlying value. It is literally gambling wearing Groucho glasses. The bet is simply for the sake of the bet.
I would not trust the AI to order food off of these websites for me because the labeled data itself is pretty much shit. Lots of dishes are unlabeled, have really bad pictures or no pictures at all, and the worse is the ones that have AI generated descriptions which are often wrong or not helpful. I'm totally fine with letting it just use my saved credit cards and delivery address (I assume the AI doesn't see them) but I can't trust it to order the food I actually want.
E: As an example (n = 1 etc): I am vegetarian. I have never ordered meat dishes, ever. Yet all of these food delivery apps suggest meat dishes for me.
I have never ordered meat dishes, ever. Yet all of these food delivery apps suggest meat dishes for me
This has been my experience too, with Uber Eats. It recommends meat dishes, every. single. time even though I have never ordered a meat dish, even once. This is the absolute basic, minimum help a food recommendation algo should do, and it still doesn't. I wonder why. Maybe vegetarians are too small a percentage of customers for these apps to care?
doordash is regularly audited for PCI, they almost certainly store raw CC#s and follow modern tokenization standards. The best the agent would see is the name you stored for the card if you have multiple cards it can pick from in the account.
You can pull the PCI compliance documentation from their trust center if you are a partner
https://trust.doordash.com/
For addresses, it's a almost certain the agent would see those since those are visible in the UI when ordering and required for selecting a destination for the order
It's plausible, but also, AI systems have a long history of being plugged in on the wrong side of security boundaries.
Remember when you were able to take over any Instagram account by asking the helpful support AI to send a password reset to an email address of your choice?
>Remember when you were able to take over any Instagram account by asking the helpful support AI to send a password reset to an email address of your choice?
Wow, first I've heard of that. Damages must have been immense for that level of incompetence occasioning financial harm. Or do those laws only apply if you can't hire very expensive lawyers to blame a computer?
It's better they don't because considering how Claude reacts when you give it a meaningless SSH login password for a local box, the moment it learns your CC# and CVV2 it will probably call your bank and cancel the card instead of ordering any food.
> AI, I asked you to check the docs for the latest version of the API, yet you didn't. Why?
> You are completely right! That was my miss, I have no excuse. I've now looked at the docs and implemented it.
Now imagine how triggered you'd be when the AI decides to get you a vegan burger because the chicken wrap is on the second page and it know you're a health conscious person.
reply