For these devices to be useful at all so many things would have to go right.
The emergency must be reported, the van driver informed, he then has to drive (presumably without breaking any traffic laws, which EMS would do) to the affected person and correctly connect the device.
The driver needs to be somewhat trained (although the devices are largely automated), but he also has to be prepared for some serious emotional distress. Having a person die right in front of you while you are fiddling with the medical device that could save his life is a terrible task.
This is hard on EMS personell, but they are trained for this, have experience and self selected into that role. A van driver has none of that.
All in all this seems of dubious value. From the article:
"Amazon confirmed that more than 100 contract drivers took part in the experiment, with several receiving alerts from citizen responder apps and arriving on site, where rescue services were already treating the victims."
Meaning there was exactly zero value in the program.
> Meaning there was exactly zero value in the program.
if your customers die, they stop being customers. definitely some long term value there, but youre right, the psychological damage to the drivers could offset those gains.
I think the point that the parent commenter is making is that if the drivers are always arriving after the actual emergency workers, they're not providing any value. It sounds like they've yet to save a customer with this experiment, so the value you're talking about is still hypothetical.
Amazon is the one performing the experiment; if the sample size isn't big enough to be useful, that's a mark against their experimental method, not the feedback of the people on this forum. The claim that this is actually valuable in some way should be the one with the burden of proof, and it's unreasonable to argue that flaws in an experiment showing one result somehow imply that the opposite is more likely because of it.
Also "Amazon driver" is being phased out as a human occupation. While their replacements are humanoid perhaps those can be of some emergency assistance. At least as a flotation device.
Initial targets for humanoid bots are in the $20k range to purchase. Human Amazon drivers currently are around $40k per year to rent. At even half the performance and the same (40h/168h) 24% duty cycle, the humanoids will have to have very high maintenance and power costs to be more expensive than humans. And they don't unionize.
Hilarious. You really think a humanoid robots time is so valueless that it would be wasted on going door to door with packages? Why? You would obviously use them where their economic activity is the most valuable.
This is peak human narcissism. It won't be robots doing the jobs humans do not want to. It will be humans doing the jobs, that are so valueless that robots won't bother with them.
>Human Amazon drivers currently are around $40k per year to rent.
Now. Yes, NOW. Consider what will happen to the unskilled labor market when there are humanoid robots competing for these jobs. Going from place A to place B with a package is of so incredibly low economic value that it will always be cheaper for a human to do than for a robot.
Consider this. A robot at its lowest will cost around 10k with 2k to 5k yearly maintenance. Human labor at its lowest is around 3 dollars a day. Robot labor will never be cheaper than human labor.
In other words. A human consumes 2k kilo calories a day. How many calories does it take to produce, charge and maintain a robot? Without a doubt many orders of magnitude more.
To think that robots will ever be cheaper than humans is such a warped western perspective. You are just neglecting how cheap labor is around the world.
> You are just neglecting how cheap labor is around the world
You can't globally arbitrage local labor, unless the labor is performed remotely, which the upcoming humanoid bots won't require much of in the delivery role.
None of my argument has to do with global arbitrage. Local competition with robot labor is enough to drive wages to near zero.
Robot labor will first overtake where the greatest economic gain is. All human labor below that skill level will now be pushed down to lower levels.
Robot labor costs will be a hard upper limit for human wages for unskilled labor. There will be like 40% of the population in direct competition with robots. Human wages will drop hard and very fast. Thankfully things like package delivery are always cheaper for humans. Again minimum labor cost for humans is like 5 dollars a day. Robots can not compete with that, so the easiest unskilled jobs will remain human.
If I was having a heart attack, I'm sure'd I'd be fine if Barney the Dinosaur showed up with a defibrillator. But from where I'm sitting right now, bereft of heart issues (that I know of), I get kind of icked out, knowing that the guy who shows up with a jump start kit could be a stressed out Amazonian who still is expected to do 200 more deliveries that day.
If they save my life, do I get a discount on Prime? Does that defibrillator deliverator get a bonus 15 minute break? Are they allowed to use a real bathroom instead of a bottle as a reward?
(I looked up what dual use meant right after I posted this, and apparently it has military implications usually?)
(The thing I'm trying to point at here is that we have limited resources, and whenever we can use a single thing for more than one purpose, it's good.)
(Maybe the term I was looking for was dual-purpose? But even that doesn't really seem to map quite right.)
> “This type of initiative is always a plus. If someone has a cardiac arrest, the more people trained and defibrillators available, the better,”
I suspect there is a limit to this ideology. It's not an area where I'd expect more to automatically equal better. Moreover, if you want AEDs in residential areas, why not just install them there? Make it part of code for new constructions.
> “And for Amazon, it valorizes their image.”
At the expense of their employees. Do your delivery drivers get a pay bump? Do they get hazard pay? These scenarios can get pretty hairy and there's a decent chance they may end up watching someone die. Do they get time off? Counseling? Anything like what other first responders are expected to get?
> The program’s backers also speculated that Project Pulse could improve driver retention.
Managerial nonsense. They're trying to solve public perception problems of congestion and driver quality of life issues by having them run around with AEDs? This is pure corporate insanity.
> It's not an area where I'd expect more to automatically equal better.
It really is. I've been to over a dozen cardiac arrests where the patient was surrounding by a gaggle of able-bodied adults, doing nothing except looking confused.
AEDs are about 1500 to 2000 generally here in the US at a quick glance and they're not a set and forget device you have to maintain them regularly or their battery will be dead when you go to use it. They only last about 2-5 years before they need replacing.
Also, you need a despatch service and vehicle to get your home defibrillator to the stranger who called emergency services on the other side of the estate. I've heard of promising pilots with defibrillator drones, but using existing delivery infrastructure makes a lot of sense.
You just described the nr. 1 reason why people hate fire alarms.
Even though we all agree they are prudent, you rarely meet people for whom checking battery health is part of the routine is they go on holiday. The audible chirp (and eventual full out false alarm) is just horrible user interface for not-so exceptional use case.
It's not just the batteries that need regular servicing, the pads do too. They expire on a similar timescale to the batteries but most AEDs don't alarm for expired pads.
It's not uncommon for AEDs to also need firmware updates, although this typically doesn't effect the usability of the device.
There are prescription AEDs and non-prescription AEDs. In most of the non-prescription ones I've seen the pads are part of the self check and you will get an alert if they're out of tolerance or past expiration. All the Philips HeartStart AEDs, which is what we have in our offices, have this feature.
> AEDs to also need firmware updates
There's a set of release notes I'd very much like to read.
My point being requiring them to exist in every home doesn't guarantee they'll be available and usable where having them on the trucks and as part of a program they're more likely to be in decent shape. The other main thing I see is how often will people know where their required AED actually is in time to use it?
These same flaws exist in every safety device we deploy. In practice it's obvious that having a consistently deployed base mostly solves these issues. There are corner cases, to be sure, but how is the greatest amount of good served?
Having these on delivery trucks ferried by unskilled drivers in non emergency vehicles who might just happen to be close enough?
Or just having them in almost every house on the block even if some number of them fail to work on first use?
I strongly suspect the latter is the better choice. Just replace AED with fire extinguisher and we could literally repeat this entire conversation.
More pharmacy options is fine by me. The Mark Cuban branded pharmacy that everyone was excited about can be a good option if you have no insurance and you’re careful to check the grand total (inclusive of shipping fees) against your local pharmacy.
The few times I checked, it was net cheaper for me to go to the local pharmacy though. Many generic drugs are $2-5 these days and ordering through an online pharmacy with shipping fees is both slower and more expensive.
If Amazon could integrate cheap generics into their shipping infrastructure it would solve the shipping problem. If they made the effort to work with insurance they’d be a great option.
I’m perfectly fine with a race to the bottom for medication dispensing middlemen.
I have no doubts that you end sentence is likely scenario.
BigTech can operate their products at a loss to kill competition, capture market, lobby gatekeeping legislation, and then raise prices to meet shareholder expectations.
Sometimes I wonder how progress would have looked like if we by some means had a hard cap how big companies can get.
I am quite skeptical of how viable this is.
For these devices to be useful at all so many things would have to go right. The emergency must be reported, the van driver informed, he then has to drive (presumably without breaking any traffic laws, which EMS would do) to the affected person and correctly connect the device. The driver needs to be somewhat trained (although the devices are largely automated), but he also has to be prepared for some serious emotional distress. Having a person die right in front of you while you are fiddling with the medical device that could save his life is a terrible task. This is hard on EMS personell, but they are trained for this, have experience and self selected into that role. A van driver has none of that.
All in all this seems of dubious value. From the article: "Amazon confirmed that more than 100 contract drivers took part in the experiment, with several receiving alerts from citizen responder apps and arriving on site, where rescue services were already treating the victims." Meaning there was exactly zero value in the program.
> Meaning there was exactly zero value in the program.
if your customers die, they stop being customers. definitely some long term value there, but youre right, the psychological damage to the drivers could offset those gains.
I think the point that the parent commenter is making is that if the drivers are always arriving after the actual emergency workers, they're not providing any value. It sounds like they've yet to save a customer with this experiment, so the value you're talking about is still hypothetical.
i give to thee the small sample size song:
https://m.youtube.com/watch?v=dw9qqvm-LT8
Amazon is the one performing the experiment; if the sample size isn't big enough to be useful, that's a mark against their experimental method, not the feedback of the people on this forum. The claim that this is actually valuable in some way should be the one with the burden of proof, and it's unreasonable to argue that flaws in an experiment showing one result somehow imply that the opposite is more likely because of it.
Also "Amazon driver" is being phased out as a human occupation. While their replacements are humanoid perhaps those can be of some emergency assistance. At least as a flotation device.
>While their replacements are humanoid perhaps those can be of some assistance.
Their replacement is automated transfer between van and package station.
Robot labor is far too valuable to waste on going door to door with boxes, humans will always be cheaper.
Initial targets for humanoid bots are in the $20k range to purchase. Human Amazon drivers currently are around $40k per year to rent. At even half the performance and the same (40h/168h) 24% duty cycle, the humanoids will have to have very high maintenance and power costs to be more expensive than humans. And they don't unionize.
Hilarious. You really think a humanoid robots time is so valueless that it would be wasted on going door to door with packages? Why? You would obviously use them where their economic activity is the most valuable.
This is peak human narcissism. It won't be robots doing the jobs humans do not want to. It will be humans doing the jobs, that are so valueless that robots won't bother with them.
>Human Amazon drivers currently are around $40k per year to rent.
Now. Yes, NOW. Consider what will happen to the unskilled labor market when there are humanoid robots competing for these jobs. Going from place A to place B with a package is of so incredibly low economic value that it will always be cheaper for a human to do than for a robot.
Consider this. A robot at its lowest will cost around 10k with 2k to 5k yearly maintenance. Human labor at its lowest is around 3 dollars a day. Robot labor will never be cheaper than human labor.
In other words. A human consumes 2k kilo calories a day. How many calories does it take to produce, charge and maintain a robot? Without a doubt many orders of magnitude more.
To think that robots will ever be cheaper than humans is such a warped western perspective. You are just neglecting how cheap labor is around the world.
> You are just neglecting how cheap labor is around the world
You can't globally arbitrage local labor, unless the labor is performed remotely, which the upcoming humanoid bots won't require much of in the delivery role.
None of my argument has to do with global arbitrage. Local competition with robot labor is enough to drive wages to near zero.
Robot labor will first overtake where the greatest economic gain is. All human labor below that skill level will now be pushed down to lower levels.
Robot labor costs will be a hard upper limit for human wages for unskilled labor. There will be like 40% of the population in direct competition with robots. Human wages will drop hard and very fast. Thankfully things like package delivery are always cheaper for humans. Again minimum labor cost for humans is like 5 dollars a day. Robots can not compete with that, so the easiest unskilled jobs will remain human.
If I was having a heart attack, I'm sure'd I'd be fine if Barney the Dinosaur showed up with a defibrillator. But from where I'm sitting right now, bereft of heart issues (that I know of), I get kind of icked out, knowing that the guy who shows up with a jump start kit could be a stressed out Amazonian who still is expected to do 200 more deliveries that day.
If they save my life, do I get a discount on Prime? Does that defibrillator deliverator get a bonus 15 minute break? Are they allowed to use a real bathroom instead of a bottle as a reward?
https://archive.is/2025.04.10-095446/https://www.bloomberg.c...
Dual use!
(I looked up what dual use meant right after I posted this, and apparently it has military implications usually?)
(The thing I'm trying to point at here is that we have limited resources, and whenever we can use a single thing for more than one purpose, it's good.)
(Maybe the term I was looking for was dual-purpose? But even that doesn't really seem to map quite right.)
sounds similar to what Jewish dense communities have https://en.m.wikipedia.org/wiki/Hatzalah I hope that amazon's been consulting with them to get some tips
> “This type of initiative is always a plus. If someone has a cardiac arrest, the more people trained and defibrillators available, the better,”
I suspect there is a limit to this ideology. It's not an area where I'd expect more to automatically equal better. Moreover, if you want AEDs in residential areas, why not just install them there? Make it part of code for new constructions.
> “And for Amazon, it valorizes their image.”
At the expense of their employees. Do your delivery drivers get a pay bump? Do they get hazard pay? These scenarios can get pretty hairy and there's a decent chance they may end up watching someone die. Do they get time off? Counseling? Anything like what other first responders are expected to get?
> The program’s backers also speculated that Project Pulse could improve driver retention.
Managerial nonsense. They're trying to solve public perception problems of congestion and driver quality of life issues by having them run around with AEDs? This is pure corporate insanity.
> It's not an area where I'd expect more to automatically equal better.
It really is. I've been to over a dozen cardiac arrests where the patient was surrounding by a gaggle of able-bodied adults, doing nothing except looking confused.
Speed, even over quality, is of the essence: https://my.clevelandclinic.org/health/diseases/21736-cardiac...
Even shitty CPR is better than no CPR.
This is not opinion it is evidence-based medicine.
https://www.ems1.com/ems-products/cpr-resuscitation/articles...
AEDs are about 1500 to 2000 generally here in the US at a quick glance and they're not a set and forget device you have to maintain them regularly or their battery will be dead when you go to use it. They only last about 2-5 years before they need replacing.
Also, you need a despatch service and vehicle to get your home defibrillator to the stranger who called emergency services on the other side of the estate. I've heard of promising pilots with defibrillator drones, but using existing delivery infrastructure makes a lot of sense.
> you have to maintain them regularly or they battery will be dead
The ones we have act like smoke alarms in a low battery state. They produce a visual flash and an audible chirp to draw your attention to the device.
> They only last about 2-5 years before they need replacing.
The batteries are replaceable.
You just described the nr. 1 reason why people hate fire alarms.
Even though we all agree they are prudent, you rarely meet people for whom checking battery health is part of the routine is they go on holiday. The audible chirp (and eventual full out false alarm) is just horrible user interface for not-so exceptional use case.
It's not just the batteries that need regular servicing, the pads do too. They expire on a similar timescale to the batteries but most AEDs don't alarm for expired pads.
It's not uncommon for AEDs to also need firmware updates, although this typically doesn't effect the usability of the device.
> but most AEDs don't alarm for expired pads.
There are prescription AEDs and non-prescription AEDs. In most of the non-prescription ones I've seen the pads are part of the self check and you will get an alert if they're out of tolerance or past expiration. All the Philips HeartStart AEDs, which is what we have in our offices, have this feature.
> AEDs to also need firmware updates
There's a set of release notes I'd very much like to read.
My point being requiring them to exist in every home doesn't guarantee they'll be available and usable where having them on the trucks and as part of a program they're more likely to be in decent shape. The other main thing I see is how often will people know where their required AED actually is in time to use it?
These same flaws exist in every safety device we deploy. In practice it's obvious that having a consistently deployed base mostly solves these issues. There are corner cases, to be sure, but how is the greatest amount of good served?
Having these on delivery trucks ferried by unskilled drivers in non emergency vehicles who might just happen to be close enough?
Or just having them in almost every house on the block even if some number of them fail to work on first use?
I strongly suspect the latter is the better choice. Just replace AED with fire extinguisher and we could literally repeat this entire conversation.
This would indeed be great - IF they widen out the experiment and it won't stay a single PR stunt.
Amazon seems quite interested in expanding into healthcare.
- One Medical, for primary care
- Amazon Pharmacy, for prescription medication
- This first responder experiment could potentially be commercialized somehow
I think I’m fine with it so long as they don’t verify that you’re a Prime subscriber before treatment a la Cyberpunk.
> I think I’m fine with it so long as they don’t [...]
Why not just keep it short and write I am not fine with this.
I really don't want their gig economy dystopia to leak into health care.
Amazon prime now includes 1 free van redirection in case you or someone from your family* needs immediate assistance
*includes spouse and children (up to a family of 5 people)
Why would they write that they aren't fine with it when they are?
Kind of like biting into a plastic apple.
More pharmacy options is fine by me. The Mark Cuban branded pharmacy that everyone was excited about can be a good option if you have no insurance and you’re careful to check the grand total (inclusive of shipping fees) against your local pharmacy.
The few times I checked, it was net cheaper for me to go to the local pharmacy though. Many generic drugs are $2-5 these days and ordering through an online pharmacy with shipping fees is both slower and more expensive.
If Amazon could integrate cheap generics into their shipping infrastructure it would solve the shipping problem. If they made the effort to work with insurance they’d be a great option.
I’m perfectly fine with a race to the bottom for medication dispensing middlemen.
They need to fix their counterfeit problem before I’d buy any medicinal from them.
I have no doubts that you end sentence is likely scenario.
BigTech can operate their products at a loss to kill competition, capture market, lobby gatekeeping legislation, and then raise prices to meet shareholder expectations.
Sometimes I wonder how progress would have looked like if we by some means had a hard cap how big companies can get.
This could go badly if it ends up in the mass adoption-enshittification pipeline.
I could see them putting volunteer responder lighting kits and avoiding parking tickets by doing this. I’m sure they’d never do such a thing.