We should ban all representations of computers as human; all computer-generated (including AI-generated) communication needs to identify itself as such.
One way to think of it: Why not, unless you are trying to trick someone?
> Why not, unless you are trying to trick someone?
On social media, there's no good UI/UX for communicating something is AI-generated without it being too verbose and defeating the point. It sounds silly, but it's the truth.
Meta's requirement for AI-generated media to be disclosed on FB/Insta has been the only push toward social media support.
> On social media, there's no good UI/UX for communicating something is AI-generated without it being too verbose and defeating the point. It sounds silly, but it's the truth.
It is silly. Of all problems in the world, I bet that one could be solved.
Until we properly integrate LLM into culture, people can always test by making off-color remarks that trip up commercial LLM filters. Or by asking strangely off-topic questions. There are quirks that we can use to spot them.
Does this also ban generated voices when they self identify as such? IMHO, if someone is not trying to deceive, it should be allowed. E.g. if the call starts out as "this is ai generated voice from xyz, ____". There are likely useful use cases for that.
A useful use case for the unsolicited caller. I don't believe there is a single useful use case for an unsolicited robocaller for the receiver to begin with, regardless of the voice being human or not.
We are discussing robocalls. I don’t know what you are trying to achieve with a comment like that. It is obviously not in good faith. A call to a business with the express purpose of working with them is exactly why they have a phone number.
> Consistent with our statements in the AI NOI, we confirm that the TCPA’s restrictions on the use of “artificial or prerecorded voice” encompass current AI technologies that resemble human voices and/or generate call content using a prerecorded voice. Therefore, callers must obtain prior express consent from the called party before making a call that utilizes artificial or prerecorded voice simulated or generated through AI technology.
So that disclosure won't work, unless (IANAL) you have a checkbox in your signup flow that says "Yes, I consent to allowing voices generated by AI call me."
Some people record their calls. Businesses often have to per compliance in most direct to consumer sales situations. From the recording, if not algorithmically, a court of law could easily determine an AI voice case by case.
A real call center would have a record of which employee made which calls when. The court subpoenas those records and the phone company's records. If they don't match, there are problems. Unless the company wants to commit perjury by inventing fake employees and call records.
Enforcement is difficult, but tracking complaints back to the source telecom / source customer and taking them to court, generally.
Automated voice messages were already restricted, this ruling just affirms that AI generated voices fit the categorization of automated voice messages.
Here's some relevant text from the ruling:
> II. BACKGROUND
> 3. The TCPA protects consumers from unwanted calls made using an artificial or prerecorded voice. See 47 U.S.C. § 227(b)(1).
> In relevant part, the TCPA prohibits initiating “any telephone call to any residential telephone line using an artificial or prerecorded voice to deliver a message without the prior express consent of the called party” unless a statutory exception applies or the call is “exempted by rule or order by the Commission under [section 227(b)(2)(B)].” 47 U.S.C. § 227(b)(1)(B). The TCPA does not define the terms “artificial” or “prerecorded voice.”
and later
> III. DISCUSSION
> 5. Consistent with our statements in the AI NOI, we confirm that the TCPA’s restrictions on the use of “artificial or prerecorded voice” encompass current AI technologies that resemble human voices and/or generate call content using a prerecorded voice.
> tracking complaints...and taking them to court, generally
Incredibly prejudiced judicial procedure, given the power, size, globalization, and ease of automated calling systems vs the normal people they most affect. Multiplied by an already burdened court system.
> Automated voice messages were already restricted, this ruling just affirms that AI generated voices fit the categorization of automated voice messages.
This is helpful. This isn't a tip-of-the-spear ruling, then, just something that affirms another ruling. But regardless, it sounds easy but in fact necessitates a huge amount of burden.
> Incredibly prejudiced judicial procedure, given the power, size, globalization, and ease of automated calling systems vs the normal people they most affect. Multiplied by an already burdened court system.
Well sure, the FCC should mandate a code to dial after a call that induces an electric shock into the most recent caller; I think *ZAP should do it. But we have to work with what's available :P
This came up on a thread the other day, and I think a good counterpoint is emergency evacuation orders for the elderly. My mom doesn’t use a computer, cell phone, tablet, etc, and a robocall to her land line would be the only way to notify her.
Is it? I mean, the scammer scammers can still use AI to answer the phone. They just can’t initiate calls en masse using AI, which I don’t see them doing.
Some of them actively call in on known spammer numbers, like the numbers found on a fraudulent Norton invoice. Often the scammers wait for you to call.
The "poor political campaigns" already exempted themselves from needing to adhere to the "do not call" list. So were they to make robocall's illegal, the politicians would likely again exempt themselves from the "robocalls are illegal" law.
With the result that (assuming the existing robocallers all quit) the only robocalls one would get would be politician robocalls.
In any case, most all of the current robocalls are already "illegal" under one or more existing laws/regulations, yet they still occur because the ones making the robocalls face few (if any) penalties for violations.
cryptographic signatures are going to have to start becoming necessary for all kinds of things, like even your average JPG image, otherwise nobody can tell what is "fake" or not, court evidence will start to become useless.
STIR/SHAKEN is already required for VOIP providers and intermediate carriers. The FCC is working it's way through the system to implement this, there is in fact a way but it takes a while.
> STIR/SHAKEN is already required for VOIP providers
I'm not convinced that STIR/SHAKEN even works properly. Recently, I migrated a DID from one VOIP provider to another. I set the outbound caller ID on the new provider, and it was showing up Verified with a checkmark to mobile devices before I had even submitted the port request to the old provider.
Depending on your new provider, they might just see that they have a contract with you and sign the call on your behalf with B level attestation - indicating that they "know" the end user, but not that they have the right to use the number.
As long as they managed to attach the identity header to the sip invite correctly, and are not considered to be a shady actor - downstream providers such as carriers probably have no reason to label it as spam. Spam labeling is typically done via analytics, outsourced to third parties like First Orion.
Attest levels are not in themselves proper tools for spam detection. The real meat of stir shaken is the origid in the identity JWT claim which is an opaque identifier that can be traced back to a particular user/customer/network equipment.
STIR/SHAKEN being sold as the one and only solution for spam calls was a mistake as it is only one iteration in the right direction. You have a handful of RFCs and ATIS specs that the FCC told operators to implement in a phased approach, and ultimately some gaps were uncovered in practice that reduced its effectiveness.
IDK, my iPhone will show me the live transcription of the callers message without me answering it. And then if I want to speak to them, I can answer the call in the middle of the message being left and talk to them.
Sounds cool, but this concept isn't at all new. Anyone who used answering machines did exactly this. You would listen to the message being left in real-time and pick up if you actually wanted to talk to them.
If people can't be bothered to leave a message, then that's their problem.
My phone shows a live transcription of the message being left.
If I see that it's an important call, then I can pick up and answer right there mid-voicemail.
That's what I was referring to. They start talking when they leave the voicemail.
This is how we did it for a long time with home answering machines too. Except instead of reading a live transcription, you listened to their live recording, and could interrupt it and answer if you wanted to talk to them. It's not a new idea.
Callers can't (well, shouldn't) expect to be able to reach you immediately by calling you. There's a lot of valid reasons to not answer your phone. You might be driving, you might be in the bathroom, you might be getting lunch in a noisy place, you might be in the middle of a different important conversation, etc.
At which point the caller needs to realize that the "professional" thing to do is leave a message if they want to be called back. (Or try calling again later.) Because there's enough junk calls that expecting people to call back every missed call that didn't leave a message is just unreasonable.
It's not a solution for you, but you're one of a shrinking group. Phone calls as a way to communicate with unknown people are on the way out, no one under 40 uses that method except under extreme duress.
I've been on-call for decades and 24x7 caregiver and it's not an issue, even in emergencies:
1. For non-emergencies, just use social media or email, which have better anti-spam filtration.
2. For most true emergencies, "hang up and call 911" just like every doctor's office recording says.
3. For urgent non-emergencies, either accept the consequences of waiting until your can reach the person via option #1 above, or get creative. Contact friends of the person and ask if they can get ahold of them... or someone IRL near them to get their real-world attention.
4. Consider what happens if you lose or break your phone. Responsible people let a reasonable group of people know how to reach them, and the rest contact a member of that group.
25 year odl here, I prefer phone calls as my primary method of communication, and often place calls as my first method of contact with previously-uncontacted entities. Please check your assumptions :)
Better to abandon that technology all together (for normal phone calls). It should be used exclusively for emergency calls and similarly vital functions. Let everything else operate over cell networks and require explicit opt-ins before party A can call party B.
No, it's not a one or the other thing. Phone calls don't work like web apps. Hell, most land lines aren't actually copper either , they're essentially VoIP. A phone call is not just a socket connection. Look up SS7 and PSTN. It's quite literally impossible to change any of this stuff, it's far too embedded.
It's privately-owned infrastructure, for the most part. And if the companies could, they'd charge you simply for the privilege of existing in the same universe as the infrastructure even if no one ever used it, and just send that money to their shareholders.
> And if the companies could, they'd charge you simply for the privilege of existing in the same universe as the infrastructure even if no one ever used it, and just send that money to their shareholders.
Of course. If I could I'd draw a salary from every employer on the planet. People be peoplin'.
It's going to the FCC, of course. If they ever solved robocalls, what would be there for them to do? Literally, this agency has been trying to solve spam calls for half a century now. They are the most incompetent people in history.
The old Bell companies are largely already in compliance with SHAKEN/STIR. It is mostly smaller shady companies that are not, because they know their customers don't want them to comply.
What should be done is something else entirely. Apple and Google should offer, as part of their standard software, a personal "phone robot". When you get a new phone, you spend 15 minutes recording various phrases, and from that point on you just have the robot answer for you.
When the robot talks to these spammers and telemarketers, it will try to keep them on the phone as long as possible. A minute would be good, 10 minutes would be better. As the spammers tried to avoid this, Apple and Google could improve the robots to counter.
And, within a few months of this, at most, that industry would just be dead. It can't afford to spend a half hour on each call trying to determine if they've got a real live knucklehead who will start sending cash to Nigerian princes, or just bad software tricking operators who don't speak English as a first language. Their margins would drop, their need for more sophisticated AI to try to determine if they were talking to a real person or not would skyrocket, etc. It just wouldn't be economically viable to continue.
So that it would be standard, and could tap into the "setting up my new phone".
Just looked up Robokiller...
>Robokiller is a phone app that blocks 99% of spam calls and texts with predictive analytics and audio fingerprinting.
Doesn't look like what I'm talking about at all. We don't want the calls to be blocked, we want them to linger on forever. I'm not sure why that's so difficult to understand.
Sure. So, let's see what's wrong with that... it's one feature of many, and they focus on the wrong one. Not big enough to make it ubiquitous or even a standard. Can't tap into the "everyone sets this up" level of authority the other two companies have.
You seem to think I was saying that I have this neat idea for an invention, and you're rebutting with "someone already thought of that".
I was describing "this needs to be a policy, if only a soft one, and only these two gigantic companies have the sway to do that". So you've totally misread things. It didn't click for you. That happen to you much? I guess I shouldn't ask, you wouldn't know even if that were the case.
Sometimes they're bad at communicating. Other times, they're at the zoo near the chimpanzee enclosure. If HackerNews ever has an interactive crayon drawing canvas, I can try again I guess.
Computer time isn't that expensive; I'm relatively certain that the calls I get are either fully driven by voice recognition, or by someone in the third world or in prison, pressing buttons that activate pre-recorded statements by a script.
The former is cheap enough that yes, they would engage for 15 minutes. The latter are smart enough to understand what's going on so that they'd hang up.
> The former is cheap enough that yes, they would engage for 15 minutes.
No, they wouldn't. This isn't "hey, when they call some random number and talk to a grandma that will never buy their stuff/scams, is wasting 15 minutes that once a big deal for them".
It's 15 minutes on every call, or enough that they can't filter down to those who will end up sending money.
> The latter are smart enough to understand what's going on so that they'd hang up.
That's debatable. But even if they are smart enough, please describe what logic you think they're using that they can tell pre-recorded voice responses from a live person? What exactly would go on in one of those calls? Did his "oh sure, uh huh" sound a little too much like the last one?
It seems like they're targeting the symptom instead of the problem.
One of the biggest problems with robocalls is that it's really impossible to know who's calling you, and that SPAM reporting tools don't have much teeth.
IE, when I have an incoming call, I should be able to see who's liable for the call. IE, "[phone number] is registered to [Person or corporation]", and that reports of spam should impede that party's ability to use the phone network.
Do most people actually care about being able to place phone calls and be anonymous in 2024? If I call someone it's either someone who has my number already or someone who is going to ask who it is (like a business) and I'm going to tell them who I am.
There are many valid reasons for making anonymous calls in 2024, including but not limited to being able to suss out information without exposing ones on identity.
Me and most people I know have stopped answering the phone completely if we don't recognize the number, because the ratio of spam to useful calls is so huge. Since this screening renders your use case for anonymous calls completely moot, the benefit of allowing them (very small, in my opinion) has to be weighed against the costs of the current system. Just to pick a random one, political polling is completely fucked at the moment, because so many people don't pick up pollster calls.
Edit: actually the more I think about your comment, the less sense it makes. What information could be gained by an anonymous phone call? Please walk me through this scenario, because I don't see it at all. Who is giving away sensitive information to an anonymous caller that they wouldn't give if there was caller ID?
> Doctor's offices and schools are notorious for using the caller ID "blocked." I let them hit voicemail.
My doctor's office won't leave messages, and appears to have about 20 minutes a day where they pick up the phone, so, if I don't pick up when they call, then I can't talk to them. (I know, I know, get a new doctor. But this is my third try to find a specialist who's willing to go beyond "here are some easy suggestions that you've already told me don't apply to you," and there are only so many battles that I can pick before I just run out of specialists entirely.)
You can thank HIPAA for that. Under the Privacy Rule medical information has to be guarded. While I have seen some practices let you indicate on the patient forms that you allow brief or full voicemail, many won't do it as there's no one to confirm their name and DOB. Even the fact that you are a patient at a clinic can be protected health information (for example getting a call from a women's health clinic or drug rehab center that doesn't block caller ID can be compromising).
Yes. What happened to that? It's interesting that we became more private in that regard while gushing personal information from sensors worn on our bodies 24/7.
They stopping printing phone books because everything is online.
Google your name and you’ll likely find much more information than the white pages ever had. I found an old email address of mine from the 90s that is long gone, every place I’ve ever lived, relationships to various family members, my parent’s address dating back decades, even my grandfathers last couple addresses and he’s been dead for over 20 years.
About 10 years ago someone on eBay tried to pull something on me and I was trying to figure out what I was dealing with. Within 45 minutes I had his name, parent’s names, phone number, and their address. I didn’t do anything with it, but it wasn’t that hard to find, with nothing more than a username or email address.
Scale I guess. No one but people nearby will have your local phonebook. And there would be no way to go through all the information even if someone had all phonebooks. The world used to be far more disconnected.
They still do. If you’ve made any public transaction (like buying a home), Whitepages will publish your info. That’s not the only reason for it, either. My 90 year-old relative was listed, and she doesn’t own anything.
I think what I would is a level playing field. If I get a call like that I cannot trace, I would expect that I should be able to do the same. If I am held to a standard that is not conducive to privacy, so should the person on the other side of that call.
But.. there is money on the line. Clearly, money from telemarketers/scammers/whoever is using this tech is enough to make telecoms hesitate from actually doing something about it.
Make it an option. I should be able to block my number from the receiver of the call if I choose. The receiver should be notified the number is blocked and can choose accordingly. The fact that numbers can be spoofed is what should be illegal. Any company making calls should have to identify themselves to the person receiving the call.
When I visited the FCC many years ago, one fo the reasons they give for allowing anonymous calls is the the protection of domestic violence victims. Eg they may need to call their abuser to talk about child support payments. They shouldn't need to reveal too much information away, particularly if it could be used to find their address (eg a phone number)
I’ve never personally been involved in this type of situation, but it seems like if the relationship is such that there is a safety issue from information potentially slipping during a phone call, maybe the court should be dealing with that communication if there is an issue with child support payments not getting made.
Wonder why my husband is 20 minutes late to drop my kid off for the weekend. Let me call up my lawyer and he'll get a date on the judge's docket next month to find out what's up.
If he is dropping the kid off, he already knows where you live, so having him figure it out via a phone number is kind of a moot point.
The example given was child support, which is financial, not visitation. I’m assuming this person would be an ex-husband, and that abuse, leading to assurance that he can’t track you down, means visitation with the kid is off the table.
Be careful what you wish for. No reason why governments might decide they want the same thing for the Internet and domain names. Requiring a license to own domains… who are we kidding, they’d do it for the tax revenue.
I think "SHAKEN/STIR" is supposed to fix this long term. I'm not sure why it's taking so long, but I believe phones will already indicate if the phone call has a verified caller id. Probably next step is to just block any non-verified caller. I'm assuming there's just a lot of migration work to happen.
I would say that money is the root of the problem. I think that most VOIP providers don't want to loose out on unencrypted traffic (both legitimate and spam).
Also, why do I seem to always get spam from a few providers? And why aren't we holding them accountable?
Money is always the problem. In the carrier world, the party accepting ("terminating") the call gets paid by the party originating it. This is why there are VoIP services that will give you a free inbound-only number and why others only charge for outbound calls.
If you're a carrier, it pays to terminate all calls -- spam or not -- by delivering them to your actual customer. You get paid by the originating carrier, and in a lot of cases you also get to charge your customer per-minute fees (or use up their prepaid minutes).
I also get very few spam calls, but I ended up buying Verizon's thing that prevents spam calls. It is all a scam but before signing up I got a ton of spam.
(What makes me sad is that I mostly use Google Voice; and that blocks spam pretty well. But people can still call my actual mobile number by guessing it, and they do.)
Google Voice has gotten somewhat difficult recently because some API-to-SMS services consider it "VOIP", and so they flat-out refuse to send text messages. Some places do this on purpose (Discord won't let me use it for 2FA because 2FA is really their anti-spam mechanism, not a security feature), and some places do it by accident (I couldn't add my Fidelity FSA debit card to Apple Pay because it simply won't send the verification code to my number on file). So some people have my "real" phone number now and it makes me sad, but that's why they call it the Internet Of Shit. (I don't even WANT SMS 2FA. Less secure than making your password 1234. Harder to use than a Nomad. Please let me use my Yubikey or a Passkey.)
> Signed traffic between Tier-1 carriers increased to 85% in 2023
We're getting there, just not soon enough. The whole world will have transitioned to never answering their phone before this actually is fully enforced.
This was my thought too. While I do think going after this kind of scam is a good first step, I don't see overseas operators not using this any less. Most spam calls I get don't follow the do not call list, why would they follow this either?
I think the FCC needs to step up and have a hard deadline for STIR/SHAKEN with fines for operators who don't comply. That is the only way, IMHO, that the VOIP operators will take it seriously.
As long as 1% or more of voters in Pennsylvania keep voting based on whomever talked to them last; and as long as Super PACs can continue to receive unlimited anonymous money; no media channel will be legally restricted from spamming people. Phone spam is too effective politically.
I don't think there's much evidence to suggest that robocalls produce material swings in elections at all, let alone 1%, a number commonly attributed to all campaign GOTV efforts put together.
I don't think there's much evidence that these fraudulent robocalls have much of an impact, if any, either. You can tell a plausible story that they have the opposite effect (they tend to target the Black vote, and the Black vote is relatively well organized compared to other US voting blocs, and is sensitive to suppression). The people running these campaigns tend to be complete chucklefucks, so it doesn't follow from the fact that people are taking the time to do them that they actually work.
> Personally I think the idea that fake news on Facebook, which is a very small amount of the content, influenced the election in any way — I think is a pretty crazy idea. Voters make decisions based on their lived experience.
Mark Zuckerburg, Sep 27, 2017
> The facts suggest the greatest role Facebook played in the 2016 election was... Campaigns spent hundreds of millions advertising online to get their messages out even further. That's 1000x more than any problematic ads we've found... After the election, I made a comment that I thought the idea misinformation on Facebook changed the outcome of the election was a crazy idea. Calling that crazy was dismissive and I regret it. This is too important an issue to be dismissive. But the data we have has always shown that our broader impact -- from giving people a voice to enabling candidates to communicate directly to helping millions of people vote -- played a far bigger role in this election."
Mark Zuckerburg, Sep 13, 2018
> When it comes to implementing a solution [to influence campaigns opposed by both parties], certainly some investors disagree with my approach to invest so much in security. [Read the 3,300 word description of concrete actions here https://www.facebook.com/notes/737729700291613/]
Do you know who the real "chucklefucks" are? The people telling Mark Zuckerburg "plausible" stories with first principle inductive reasoning about what is or is not important on Facebook. It was a huge mistake to listen to them between November 4th and November 11th, 2016, just when he issued his first erroneous comment. He controls all the data on Facebook and has the means to analyze it, so he had absolutely no reason to listen to those people at all. He should have just waited and found out what the real answer was.
You're making a good faith comment. But you don't really know what evidence there is. In fact you don't know anything about it at all. You have no reason to speculate, because campaigns and phone companies have all of the data needed to answer the question, and agitating them to answer it is the right thing to do. Mistakes happen from people conflating fast answers with correct ones. Even Mark Zuckerburg does. So your answer is good because it is fast and inductive and first principles, but it is also really, really bad because it requires no reading, no analysis and no real knowledge, just fuzzy-wuzzy podcast-and-pop-sci takeaways. Sucking the air out of the room with a fast and cheap answer undermines the people trying to investigate influence campaigns. So you can be sincere and co-opted at the same time.
based on the evidence around effectiveness of social media ads, his initial comment was likely right. there's a reason campaigns still mostly spend on tv, knocking, and phone.
'well organized' in the sense that there is a lot of GOTV organizing but that is to make up for a deficit, it doesn't mean that black folks are particularly resilient to these tactics.
Well-organized as in it's well-organized; for instance, it's significantly coordinated through Black churches (church participation is partly predictive of Democratic turnout performance in major Black districts). All this from White & Laird's book.
I mean, by all means send people who do this stuff to prison. I'm not saying it shouldn't be taken seriously. But I don't think it really works at any kind of scale.
That assumes a number of facts not anywhere in evidence, including that the people launching these idiotic fraud call operations are rational actors (the ones we've learned about so far manifestly are not), and that fraudulent calls would work algebraically against actual GOTV calls.
That's also the issue with swatting and fake calls to 911. When investigators trace it they'll hit a VOIP provider and it becomes near impossible to take it any further.
It's already possible to lookup the carrier of a number, and I'd love for the ability to be listed under their location on the incoming call screen. Makes a big difference if the call is coming from T-Mobile or some company you've never heard of.
It is maddening that the companies that provide the service appear to have thrown up their hands & pretend that they have no idea how they could possibly prevent spoofed numbers. Imagine if this was this easy to spoof IP #s. Perhaps it is.
> It seems like they're targeting the symptom instead of the problem.
I believe this is a quickly adopted band-aid in response to the recent political scam calls that pretended to be President Biden telling voters to skip voting in the primary.
I thought people's behavior these days was to ignore calls from numbers they don't know and let the phone screen it. I don't ever have problems with unknown numbers or SPAM calls on my Pixel
Twilio had some strict policies introduced that I think were industry wise for USA. Basically all voip numbers had to go through thorough checks, which even our legitimate company failed (go figure). So as long as all companies like Twilio introduce those checks then spam calls should dramatically decrease. I thought it was already the case for USA?
> Callers who use AI technology must get prior consent from the people they are calling, the FCC said.
The text of the ruling says "prior express consent" instead of unsolicited. That seems clear, but I wonder whether it is in practice. Is the one of those things where, by signing up for website A and agreeing to their terms by clicking a checkbox, I am agreeing to allow my phone number to be called by robits from companies B-Z, because of some line buried in the middle of the legal text I didn't read? I.e. "The User consents to contact for any purpose by Website A and our partners", and a partner is defined as anybody who buys their contact list from them?
That is a case where the nature of T&Cs and end-user agreements makes the words "express" and "consent" more abiguous than they ought to be, since they rarely match anyone's definitions except the law's.
> The FCC announced the unanimous adoption of a Declaratory Ruling that recognizes calls made with AI-generated voices are "artificial" under the Telephone Consumer Protection Act (TCPA).
So illegal in the sense that artificial robocalls are already illegal, then.
You can't possibly expect Congress to give executive branch agencies enough money to do a bare-minimum job of enforcing the laws Congress passes. Especially when there are political donors making sure that we deregulate things that society wants controlled so that they can rent-collect.
This ruling was driven by fake Joe Biden robocalls, but there are/(were?) AI startups trying to create AI customer support bots or political reachouts with consent from the parties involved to clone those voices.
From the declaratory ruling, any AI-generated voice call requires prior recipient consent:
> Consistent with our statements in the AI NOI, we confirm that the TCPA’s restrictions on the use of “artificial or prerecorded voice” encompass current AI technologies that resemble human voices and/or generate call content using a prerecorded voice. Therefore, callers must obtain prior express consent from the called party before making a call that utilizes artificial or prerecorded voice simulated or generated through AI technology.
So presumably the google assistant "feature" that can book a table at a restaurant for you is now illegal? IIRC that would place a call to the restaurant.
> AI startups trying to create AI customer support bots or political reachouts with consent from the parties involved to clone those voices.
This is where lawyers get to have fun. What is the line between a message in the public sphere copied and multiplied via broadcast, and a message consensually altered and multiplied via AI-then-broadcast?
The same law that bans artificial voices without prior recipient consent also bans recordings without prior recipient consent. So no difference whatsoever for phone calls.
Agreed. Once mobile data coverage is universal (via starlink et al, maybe), it's inevitable that the idea of a phone number will become antiquated. Either whatsapp (or one of its competitors) gets a sufficient monopoly and enables easily portable identities (to allow switching sims), or some other similar platform will come along. It may take a decade or two, but it will happen.
But it's never going to be universal. I felt very scared some weeks ago during a huge march against rightwing extremists in Munich, Germany. There were ~150k people concentrated on a few streets/km.
Now, how is this relevant? Well, the entire cell network was offline, at least for some providers. At first it wasn't possible to send/receive data. Calls were connecting, but my friend sounded like an alien. Then for one hour, 0 communication was possible.
So even though the most efficient (I think?) protocol was used, it came to a halt
> "State Attorneys General will now have new tools to crack down on these scams..." - FCC Chairwoman Jessica Rosenworcel
...How? How can this be enforced? What are the new tools? Based on the news release and documentation, fiat in this case means nothing but posturing, at most being hopeful some imaginary future tool will be able to bring execution to legislation.
It would be more correct to say that they have officially interpreted a current law (the Telephone Consumer Protection Act) to clarify that AI-Generated voices in robocalls violate that law, which seems reasonable.
Activist? You clearly don't have the same definition of activist that I do. Half the problem with these sorts of conversations is there is no agreement on definitions.
Please don't interpret my comment to mean that Supreme Court decisions can't be criticized, I just don't find the "activist" accusation to be particularly insightful.
Citizen’s United was “yeah, leaning pretty hard here.”
Flipping the hours d’ouvres table over on Roe v. Wade, a tense but stable compromise, that’s verging on activist. You don’t go knocking over fragile, workable standoffs that have held longer than an Ulster cease-fire if you can help it as a senior jurist.
It’s a pretty neo-neocon consensus to put it charitably.
It’s still the highest court in the land and it’s still binding, but I hope any thinking person is hoping for more a more“spirited but healthy” tension between major worldviews.
Citizen's United seems like a pretty clear cut case of individual rights from my point of view. If your definition of "activist" is strengthening individual rights and refusing to give power to the federal government and its giant bureaucracy then I guess I'm OK with an "activist" Supreme Court.
Roe v. Wade had been criticized for 50 years as an example of an activist judiciary and was held in place by rigid adherence to stare decisis.
Would you be as confident with stare decisis if we were talking about Plessy v. Ferguson, which held sway for 58 years before being overturned? Where the judges in Brown v. Board of Education of Topeka "activists"?
If "activism" is used to describe all sorts of political philosophy then it isn't really a useful term to bring to the discussion. I think it does have meaning though and that "activism" is not what conservative members of the court are doing.
Too much "Orwellian" language manipulation going on these days, IMHO.
I agree with most of what you said, but I disagree with this:
> If your definition of "activist" is strengthening individual rights and refusing to give power to the federal government and its giant bureaucracy then I guess I'm OK with an "activist" Supreme Court.
Activism is activism regardless of whether it "strengthens individual rights" or not. It would be ridiculous to argue Roe vs. Wade wasn't activist just because it "strengthened individual rights". What rights? Rights according to whom? You could justify pretty much any decision this way.
I'm not familiar with the details of the case in Citizen's United, but whether or not it constitutes activism depends not on the effects of the decision, but the reasoning by which it was reached.
Yes, the reasoning is definitely the problem with a lot of recent decisions. One thing that stands out in my mind is the majority opinion in the prayers in school case that stated that part of their reasoning to rule in favor of the coach praying was that he was doing so in private and players could voluntarily join him. When, as noted by the descent, the case itself included photographs of said coach, kneeling in prayer with his team at center court.
> Activism is activism regardless of whether it "strengthens individual rights" or not
My comment wasn't very clear, but I was trying to communicate that recognizing that the Constitution is centered around the idea of a limited federal government with explicit powers is not evidence of inappropriate "activism". Instead it is evidence of appropriate checks on federal overreach.
I don't know what to make of your statement "activism is activism". To be a useful term, "activism" needs to mean something other than "doing something" or "doing something that I disagree with".
My definition of judicial activism would be ruling in a biased manner to get a politically desired result rather than solely on the basis of the facts and the law as written.
You can rule in a biased manner in favor or individual freedom or against it. My point is that that's not relevant to the definition.
I do agree with you however that the current U.S. constitution is centered around the idea of a limited federal government with enumerated powers and that therefore an unbiased interpretation of the constitution as written will tend to result in rulings that support individual freedom in general, though that's not a hard and fast rule.
It can indeed be argued that Roe v. Wade was likewise activist, and it’s a failure of the legislature that a similarly workable compromise couldn’t have been done through proper channels.
But two wrongs don’t make a right and legislating from the bench for a net decrease in individual liberty via overturning a previous ruling runs contrary to the whole premise of stare decis (that’s Latin for “precedent” in case anyone missed that). Ruling from precedent has a lot of good properties, but maybe the best one is that it puts downward pressure on unbounded, escalating bench legislating.
Brown v. Board of Education was arguably overturning Plessy, though that was attached to specific language in the 14th Amendment, which is light-years from endless, muddy, subjective arguments around states rights like with Roe v. Wade. That question was settled by a war fought to a decisive military conclusion: red states don’t get to make draconian laws around individual liberty because we conquered them with guns. States rights “freedom” is subordinate to individual rights and freedoms via the landmark ruling in Union v. Confederacy. When it’s one interpretation of the Combined and Annotated Federalist Papers on one side and the armament of the high-GDP states on the other, well the victors make the laws. It’s a “happy accident” that the victors were on the right side of history.
Citizen’s United is the worst kind of judicial activism: tortured doctrines around corporate personhood being used to overturn a good law with good outcomes with broad bipartisan support (it was called the Bipartisan Campaign Reform Act for Christ’s sake) and it was a popular law that put downward pressure on the power of powerful institutions relative to the individual.
When the language of Locke and Rousseau is used to defend the life, liberty, and egalité of ruthless megacorps and and secessionist governors at the expense of the liberty of the common citizen: that’s Orwell hoss.
Without citizens united, a newspaper couldn't run an advertisement or an editorial about an biography of an active politician without running afoul of campaign finance laws- promoting a book that describes someone in positive light is clearly a contribution in kind, when column space in the paper costs money.
There's way more nuance that went into the decision than the "corporations are people" meme.
The Federalist Society, a political entity to alter the judicial branch, picked Neil Gorsuch while grooming many other federal judges who are then put in place by politicians. If you were put in place by activists, doesn’t that make you an activist judge?
No. Particularly when one of the primary goals of that activism is to produce judges who aren't activists:
> [The Federalist Society] is founded on the principles that [...] it is emphatically the province and duty of the judiciary to say what the law is, not what it should be.
You and I clearly have different ideas of what constitutes judicial activism. What would you consider to be "non-activist" then, if ruling based on what the law says rather than on your personal politics is itself activist?
Laws are not boolean logic that are cut and dry, otherwise we wouldn’t need human judges.
The federalist society is a well funded organization whose goal and track record is to install conservative judges to interpret the law in a very specific way, all the way up to the Supreme Court. I’d that’s not activism then I don’t know what is.
I don't disagree the Federalist Society is activist. I'm saying the judges they produce aren't, because that "very specific way" is "follow the law, not your political biases" which is, by definition, the precise opposite of judicial activism. It sounds like you disagree with that definition of judicial activism, but you haven't provided a better alternative.
To preempt the answer you've given elsewhere in this thread "they rule in a way that you don’t like" isn't a good definition, and seems from my perspective like an attempt to muddy the issue in order to allow you to put activist judges on equal moral footing with those who actually follow the law. You are correct that its impossible to eliminate all bias, but responding to that reality by throwing up your hands and saying "I guess everyone's an activist then, judges should just ignore the law and rule based solely on their personal biases instead" definitely isn't helping the situation. There's a spectrum here, and The Federalist Society has as one of its explicit goals encouraging judges to move towards the non-activist side of that spectrum.
But they’re not “following the law, not a political bias.” The political bias is contemporary conservatism, which in reason years has meant making abortion illegal, allow corporations as many rights or more than citizens, allow infinite secret money in politics, remove as much regulation and federal agencies as possible, reduce civil rights particularly for gays/trans/people of color, get rid of affirmative action, extend “religious freedom” into new and every aspect of life including as a precursor to ignore any existing law, prevent any kind of mandatory public health response, etc etc.
Everything I mentioned is not only a stated goal but has already happened. They are getting the results for which they continue fund raising. Now conservative Supreme Court judges are inviting new areas, such as when Thomas said he’s hoping for a new suit to shut down gay marriage after killing roe v wade. Trump even said he appointed judges specialty that will kill Roe v Wade. How is this not activism?
I would posit that the vast majority of that is you reading court decisions through the lens of your own political biases.
Roe vs. Wade wasn't reversed because the judges thought abortion was wrong (though they may indeed think that). It was reversed because Roe vs. Wade was a ridiculously convoluted ruling by an activist court that created a constitutional right to abortion out of thin air when a plain reading of the text of the constitution makes it obvious it contains no such provision. But if you're looking at this solely through a political lens and see it as the court "making abortion illegal" (which it actually didn't even do, it just reversed the previous ruling that was preventing states from enforcing their own longstanding laws on the subject) then its understandable why you might (incorrectly) see that as activism.
In contrast, a right to "religious freedom" very obviously does exist in the constitution, in the very first sentence of the bill of rights, and racial discrimination (affirmative action) is banned by the 14th amendment. There's room for debate as to exactly how broadly those freedoms were originally intended to be applied, but its not obvious to me on its face that those cases were decided incorrectly due to "activism" either.
The rest of your accusations are so vague that it's not clear to me what they're even referring to.
Judges are “activist” as a political label only when they rule in a way that you don’t like. It is a meaningless label generated by politicians to get them to vote for them and put in different judges who will vote the way they think you want! And that’s not activism?
Now you are just pushing "activist" towards meaning "having a legal philosophy". And in practice it means having the wrong legal philosophy with respect to the person who labels you an "activist" as opposed to having a particular philosophy.
It isn't a particularly useful term because no one agrees on what it means. This has been illustrated quite nicely by the comments to my original comment.
They have been hunting for cases to pursue their political agenda. It's probably the most activist court we've ever had. What is your definition of activist?
Your comment illustrates the problem. Do you think that everyone agrees that "activist" means "hunting for cases"? What does "hunting for cases" actually mean?
The term "activist" is often interpreted as "legislating from the bench" where the judiciary usurps the role of the legislature. Some people actually want that. Other people don't want that.
Refusing to solve a problem and instead requiring Congress to clarify the law is another judicial philosophy. Is that being an activist?
Deciding that the federal government has no authority and that state authority or individual rights are more paramount is also a course of action that some people agree with and some people don't. Supporters probably don't call that "activism" but detractors might.
So I think the term is mainly used to slur your political opponent as opposed to being a succinct term for some particular judicial philosophy.
This is just stating an opinion that you disagree with the philosophy. Do you think leftist judges also don't have a philosophy but just aim for outcomes?
There is no philosophy, it's just partisan politics. Some examples,
They are supposedly "originalist" except in their 2nd amendment rulings they ignore "A well regulated Militia" because guns is a Republican religion.
In banning Biden's student relief they ignored the text of the law and legislated from the bench just saying that it was unfair because the size was too "significant". Activism.
Last year they invented out of thin air the "major questions doctrine" to override the Clean Air Act and help polluters. Activism.
In 2022, without citing any principle, they said OSHA couldn't protect workers from covid. Activism.
When they gutted the Voting Rights Act they invented an "equal dignity of the states" doctrine. Activism.
To conservatives judicial activism is only a problem when liberals do it.
I don't feel strongly about "activism" being inherently good or bad (and don't agree with all of the outcomes in those cases) but I think your reasoning isn't great. For example:
> In banning Biden's student relief they ignored the text of the law and legislated from the bench just saying that it was unfair because the size was too "significant". Activism.
I think the majority opinion explains this quite clearly:
> The HEROES Act, Roberts emphasized, gives the secretary of education the power to “waive or modify” laws and regulations governing the student-loan programs. Congress’s use of the word “modify” means that the Biden administration can make “modest adjustments and additions to existing provisions,” Roberts wrote, “not transform them.” But the debt-relief program, Roberts stressed, instead “created a novel and fundamentally different loan forgiveness program.” The plan “modifies” student-loan laws and regulations, Roberts suggested, “only in the same sense that the French Revolution ‘modified’ the status of the French nobility — it has abolished them and supplanted them with a new regime entirely.”
(from scotusblog)
the current court solidified itself as an activist court by taking on a litany of controversial, yet already decided cases one after another that were all lined up by the same organization that not only lobbied for their placement on the court, but even went so far as to line up a billionaire buddy system to make them more comfortable financially so they wouldn't retire from the court during a democratic administration.
Usually I challenge people when they call a court stacked or activist, because it’s just so rarely true: this is as close as you’ll (hopefully) see to a 1-bit high court.
It’s the masterpiece, the magnum opus of the Magnus of parliamentary politics. Nicollo Machiavelli doesn’t have shit on Mitch McConnell at that chess game.
I’m pretty indifferent to which color bumper-sticker late capitalism is sporting when it pushes the newest round of formerly “looking forward to better” working people below the waterline, it’s not a partisan thing.
The other team have plenty who match Mitch on evilness, but zero on skill.
I think it's pretty unlikely that the Chevron doctrine would be overturned completely. The specifics of the case before the Court involve a case where the NMFS has interpreted a fisheries act to require fishers to pay the salaries of government monitors, simply because the act does not specify who should pay the salaries. The more reasonable objection is whether "reasonable interpretation" under Chevron should be limited to prevent the creation of affirmative powers out of thin air. As Wikipedia puts it:
> Whether the Court should overrule Chevron or at least clarify that statutory silence concerning controversial powers expressly but narrowly granted elsewhere in the statute does not constitute an ambiguity requiring deference to the agency.
The initial "overrule Chevron" seems like a DITF [1] and the latter is probably what the plaintiffs are hoping to achieve. Granted, I find it hard to trust a Court that overruled PP v. Casey, but most of this Court's other rulings, at least, have not been as extreme.
the case from the fishermen currently in Supreme Court, precisely will nullify unelected agency officials from interpreting laws like this to legislate, then enforce, rules outside of the mandates and powers granted by the populace
With enough evidence, operators are compelled to provide data and are given an opportunity to correct their action. If they refuse, FCC will eventually issue an order to all other providers to not accept calls from the bad actor.
Yeah because other activities they have deemed illegal have totally stopped.
I predict a season of AI generated robocalls for the elections.
From all sides.
This message brought to you by .......
For now this could be seen as an incentive for TTS solution providers - build a product that is hard to distinguish from an actual human calling. In many cases the results are already convincing.
And what about the future. Please scan your retina to initialize the phone call? Please solve a captcha to start a phone call? Your workplace registered 12948230 calls in the last 24 hours, but employs only 3 workers registered as humans, pay fine now? Interesting times.
They describe this as giving "State Attorneys General across the country new tools to go after bad actors behind these nefarious robocalls." The way that I read that is that there are these scams out there that states are already trying to bring lawsuits against, and this simply makes their job a bit easier in some of the cases they're ALREADY bringing.
An antispam idea in bitcoin circles is to require payment to open an email from an unknown source. So if I want to send you an advertisement, it will only reach you if I add a payment invoice that meets your threshold. It makes spam costly and forces advertisers to focus on a narrower range of ads to people who more likely want the product.
The trick is in how invoices can be configured in bitcoin. You would not be obliged, but you would not receive payment, and the payer would be able to reclaim those funds.
701 comments
[ 4.3 ms ] story [ 474 ms ] threadOne way to think of it: Why not, unless you are trying to trick someone?
On social media, there's no good UI/UX for communicating something is AI-generated without it being too verbose and defeating the point. It sounds silly, but it's the truth.
Meta's requirement for AI-generated media to be disclosed on FB/Insta has been the only push toward social media support.
It is silly. Of all problems in the world, I bet that one could be solved.
A useful use case for the unsolicited caller. I don't believe there is a single useful use case for an unsolicited robocaller for the receiver to begin with, regardless of the voice being human or not.
So, if I call my vet to make an appointment, is that solicitated or unsolicited?
> Consistent with our statements in the AI NOI, we confirm that the TCPA’s restrictions on the use of “artificial or prerecorded voice” encompass current AI technologies that resemble human voices and/or generate call content using a prerecorded voice. Therefore, callers must obtain prior express consent from the called party before making a call that utilizes artificial or prerecorded voice simulated or generated through AI technology.
So that disclosure won't work, unless (IANAL) you have a checkbox in your signup flow that says "Yes, I consent to allowing voices generated by AI call me."
Automated voice messages were already restricted, this ruling just affirms that AI generated voices fit the categorization of automated voice messages.
Here's some relevant text from the ruling:
> II. BACKGROUND > 3. The TCPA protects consumers from unwanted calls made using an artificial or prerecorded voice. See 47 U.S.C. § 227(b)(1). > In relevant part, the TCPA prohibits initiating “any telephone call to any residential telephone line using an artificial or prerecorded voice to deliver a message without the prior express consent of the called party” unless a statutory exception applies or the call is “exempted by rule or order by the Commission under [section 227(b)(2)(B)].” 47 U.S.C. § 227(b)(1)(B). The TCPA does not define the terms “artificial” or “prerecorded voice.”
and later
> III. DISCUSSION > 5. Consistent with our statements in the AI NOI, we confirm that the TCPA’s restrictions on the use of “artificial or prerecorded voice” encompass current AI technologies that resemble human voices and/or generate call content using a prerecorded voice.
Incredibly prejudiced judicial procedure, given the power, size, globalization, and ease of automated calling systems vs the normal people they most affect. Multiplied by an already burdened court system.
> Automated voice messages were already restricted, this ruling just affirms that AI generated voices fit the categorization of automated voice messages.
This is helpful. This isn't a tip-of-the-spear ruling, then, just something that affirms another ruling. But regardless, it sounds easy but in fact necessitates a huge amount of burden.
Well sure, the FCC should mandate a code to dial after a call that induces an electric shock into the most recent caller; I think *ZAP should do it. But we have to work with what's available :P
With the result that (assuming the existing robocallers all quit) the only robocalls one would get would be politician robocalls.
In any case, most all of the current robocalls are already "illegal" under one or more existing laws/regulations, yet they still occur because the ones making the robocalls face few (if any) penalties for violations.
I'm not convinced that STIR/SHAKEN even works properly. Recently, I migrated a DID from one VOIP provider to another. I set the outbound caller ID on the new provider, and it was showing up Verified with a checkmark to mobile devices before I had even submitted the port request to the old provider.
As long as they managed to attach the identity header to the sip invite correctly, and are not considered to be a shady actor - downstream providers such as carriers probably have no reason to label it as spam. Spam labeling is typically done via analytics, outsourced to third parties like First Orion.
Attest levels are not in themselves proper tools for spam detection. The real meat of stir shaken is the origid in the identity JWT claim which is an opaque identifier that can be traced back to a particular user/customer/network equipment.
STIR/SHAKEN being sold as the one and only solution for spam calls was a mistake as it is only one iteration in the right direction. You have a handful of RFCs and ATIS specs that the FCC told operators to implement in a phased approach, and ultimately some gaps were uncovered in practice that reduced its effectiveness.
IDK, my iPhone will show me the live transcription of the callers message without me answering it. And then if I want to speak to them, I can answer the call in the middle of the message being left and talk to them.
Sounds cool, but this concept isn't at all new. Anyone who used answering machines did exactly this. You would listen to the message being left in real-time and pick up if you actually wanted to talk to them.
If people can't be bothered to leave a message, then that's their problem.
Not professional, not an option for some calls.
> If people can't be bothered to leave a message, then that's their problem.
It's actually my problem if I miss an important call. A message is great, sure, but I still missed the call.
If I see that it's an important call, then I can pick up and answer right there mid-voicemail.
That's what I was referring to. They start talking when they leave the voicemail.
This is how we did it for a long time with home answering machines too. Except instead of reading a live transcription, you listened to their live recording, and could interrupt it and answer if you wanted to talk to them. It's not a new idea.
And not everybody has a phone that'll do this live message transcribing.
And no, everybody who decides to not leave a message isn't "not worth your time" or something.
Callers can't (well, shouldn't) expect to be able to reach you immediately by calling you. There's a lot of valid reasons to not answer your phone. You might be driving, you might be in the bathroom, you might be getting lunch in a noisy place, you might be in the middle of a different important conversation, etc.
At which point the caller needs to realize that the "professional" thing to do is leave a message if they want to be called back. (Or try calling again later.) Because there's enough junk calls that expecting people to call back every missed call that didn't leave a message is just unreasonable.
That's easy to say when you're not looking for a new job. Or don't run a business.
You live in a tiny bubble if you honestly believe that.
"Eight-in-ten Americans say they don’t generally answer their cellphone when an unknown number calls" https://www.pewresearch.org/short-reads/2020/12/14/most-amer...
And that study is from 3 years ago, it's surely a higher percentage now than it was then.
I don’t know why this is the hill you’ve chosen to die on
1. For non-emergencies, just use social media or email, which have better anti-spam filtration.
2. For most true emergencies, "hang up and call 911" just like every doctor's office recording says.
3. For urgent non-emergencies, either accept the consequences of waiting until your can reach the person via option #1 above, or get creative. Contact friends of the person and ask if they can get ahold of them... or someone IRL near them to get their real-world attention.
4. Consider what happens if you lose or break your phone. Responsible people let a reasonable group of people know how to reach them, and the rest contact a member of that group.
A man can dream.
Of course. If I could I'd draw a salary from every employer on the planet. People be peoplin'.
When the robot talks to these spammers and telemarketers, it will try to keep them on the phone as long as possible. A minute would be good, 10 minutes would be better. As the spammers tried to avoid this, Apple and Google could improve the robots to counter.
And, within a few months of this, at most, that industry would just be dead. It can't afford to spend a half hour on each call trying to determine if they've got a real live knucklehead who will start sending cash to Nigerian princes, or just bad software tricking operators who don't speak English as a first language. Their margins would drop, their need for more sophisticated AI to try to determine if they were talking to a real person or not would skyrocket, etc. It just wouldn't be economically viable to continue.
Just looked up Robokiller...
>Robokiller is a phone app that blocks 99% of spam calls and texts with predictive analytics and audio fingerprinting.
Doesn't look like what I'm talking about at all. We don't want the calls to be blocked, we want them to linger on forever. I'm not sure why that's so difficult to understand.
You seem to think I was saying that I have this neat idea for an invention, and you're rebutting with "someone already thought of that".
I was describing "this needs to be a policy, if only a soft one, and only these two gigantic companies have the sway to do that". So you've totally misread things. It didn't click for you. That happen to you much? I guess I shouldn't ask, you wouldn't know even if that were the case.
> Anyone who says that they're great at communicating but 'people are bad at listening' is confused about how communication works.
[0] https://en.wikipedia.org/wiki/Kitboga_(streamer)
[0] https://www.youtube.com/watch?v=maP2DwgdBts
The former is cheap enough that yes, they would engage for 15 minutes. The latter are smart enough to understand what's going on so that they'd hang up.
No, they wouldn't. This isn't "hey, when they call some random number and talk to a grandma that will never buy their stuff/scams, is wasting 15 minutes that once a big deal for them".
It's 15 minutes on every call, or enough that they can't filter down to those who will end up sending money.
> The latter are smart enough to understand what's going on so that they'd hang up.
That's debatable. But even if they are smart enough, please describe what logic you think they're using that they can tell pre-recorded voice responses from a live person? What exactly would go on in one of those calls? Did his "oh sure, uh huh" sound a little too much like the last one?
They're not supergeniuses.
One of the biggest problems with robocalls is that it's really impossible to know who's calling you, and that SPAM reporting tools don't have much teeth.
IE, when I have an incoming call, I should be able to see who's liable for the call. IE, "[phone number] is registered to [Person or corporation]", and that reports of spam should impede that party's ability to use the phone network.
Edit: actually the more I think about your comment, the less sense it makes. What information could be gained by an anonymous phone call? Please walk me through this scenario, because I don't see it at all. Who is giving away sensitive information to an anonymous caller that they wouldn't give if there was caller ID?
My doctor's office won't leave messages, and appears to have about 20 minutes a day where they pick up the phone, so, if I don't pick up when they call, then I can't talk to them. (I know, I know, get a new doctor. But this is my third try to find a specialist who's willing to go beyond "here are some easy suggestions that you've already told me don't apply to you," and there are only so many battles that I can pick before I just run out of specialists entirely.)
To reiterate, calls need to say who's calling. They don't need to come from a number that will be answered.
It's about liability, and making sure there are consequences for spamming.
And anonymity against your interlocutor is usually a very bad thing. Even though there are a few exceptions.
Google your name and you’ll likely find much more information than the white pages ever had. I found an old email address of mine from the 90s that is long gone, every place I’ve ever lived, relationships to various family members, my parent’s address dating back decades, even my grandfathers last couple addresses and he’s been dead for over 20 years.
About 10 years ago someone on eBay tried to pull something on me and I was trying to figure out what I was dealing with. Within 45 minutes I had his name, parent’s names, phone number, and their address. I didn’t do anything with it, but it wasn’t that hard to find, with nothing more than a username or email address.
But.. there is money on the line. Clearly, money from telemarketers/scammers/whoever is using this tech is enough to make telecoms hesitate from actually doing something about it.
The example given was child support, which is financial, not visitation. I’m assuming this person would be an ex-husband, and that abuse, leading to assurance that he can’t track you down, means visitation with the kid is off the table.
https://www.fcc.gov/call-authentication
Also, why do I seem to always get spam from a few providers? And why aren't we holding them accountable?
If you're a carrier, it pays to terminate all calls -- spam or not -- by delivering them to your actual customer. You get paid by the originating carrier, and in a lot of cases you also get to charge your customer per-minute fees (or use up their prepaid minutes).
This is the norm for standard carriers in Europe too.
(What makes me sad is that I mostly use Google Voice; and that blocks spam pretty well. But people can still call my actual mobile number by guessing it, and they do.)
Google Voice has gotten somewhat difficult recently because some API-to-SMS services consider it "VOIP", and so they flat-out refuse to send text messages. Some places do this on purpose (Discord won't let me use it for 2FA because 2FA is really their anti-spam mechanism, not a security feature), and some places do it by accident (I couldn't add my Fidelity FSA debit card to Apple Pay because it simply won't send the verification code to my number on file). So some people have my "real" phone number now and it makes me sad, but that's why they call it the Internet Of Shit. (I don't even WANT SMS 2FA. Less secure than making your password 1234. Harder to use than a Nomad. Please let me use my Yubikey or a Passkey.)
> Signed traffic between Tier-1 carriers increased to 85% in 2023
We're getting there, just not soon enough. The whole world will have transitioned to never answering their phone before this actually is fully enforced.
I think the FCC needs to step up and have a hard deadline for STIR/SHAKEN with fines for operators who don't comply. That is the only way, IMHO, that the VOIP operators will take it seriously.
Robocalls every election season go out to targeted communities telling them the wrong polling location.
I will leave as an exercise to the reader what political slant those communities almost always have. The impact of those must be very hard to measure.
> Personally I think the idea that fake news on Facebook, which is a very small amount of the content, influenced the election in any way — I think is a pretty crazy idea. Voters make decisions based on their lived experience.
Mark Zuckerburg, Sep 27, 2017
> The facts suggest the greatest role Facebook played in the 2016 election was... Campaigns spent hundreds of millions advertising online to get their messages out even further. That's 1000x more than any problematic ads we've found... After the election, I made a comment that I thought the idea misinformation on Facebook changed the outcome of the election was a crazy idea. Calling that crazy was dismissive and I regret it. This is too important an issue to be dismissive. But the data we have has always shown that our broader impact -- from giving people a voice to enabling candidates to communicate directly to helping millions of people vote -- played a far bigger role in this election."
Mark Zuckerburg, Sep 13, 2018
> When it comes to implementing a solution [to influence campaigns opposed by both parties], certainly some investors disagree with my approach to invest so much in security. [Read the 3,300 word description of concrete actions here https://www.facebook.com/notes/737729700291613/]
Do you know who the real "chucklefucks" are? The people telling Mark Zuckerburg "plausible" stories with first principle inductive reasoning about what is or is not important on Facebook. It was a huge mistake to listen to them between November 4th and November 11th, 2016, just when he issued his first erroneous comment. He controls all the data on Facebook and has the means to analyze it, so he had absolutely no reason to listen to those people at all. He should have just waited and found out what the real answer was.
You're making a good faith comment. But you don't really know what evidence there is. In fact you don't know anything about it at all. You have no reason to speculate, because campaigns and phone companies have all of the data needed to answer the question, and agitating them to answer it is the right thing to do. Mistakes happen from people conflating fast answers with correct ones. Even Mark Zuckerburg does. So your answer is good because it is fast and inductive and first principles, but it is also really, really bad because it requires no reading, no analysis and no real knowledge, just fuzzy-wuzzy podcast-and-pop-sci takeaways. Sucking the air out of the room with a fast and cheap answer undermines the people trying to investigate influence campaigns. So you can be sincere and co-opted at the same time.
I mean, by all means send people who do this stuff to prison. I'm not saying it shouldn't be taken seriously. But I don't think it really works at any kind of scale.
3-7 votes per 1k calls
I believe this is a quickly adopted band-aid in response to the recent political scam calls that pretended to be President Biden telling voters to skip voting in the primary.
It is going to be an interesting year.
Because I get 2-3 a day on my Pixel and they annoy the poop out of me, even though I don't answer them.
The text of the ruling says "prior express consent" instead of unsolicited. That seems clear, but I wonder whether it is in practice. Is the one of those things where, by signing up for website A and agreeing to their terms by clicking a checkbox, I am agreeing to allow my phone number to be called by robits from companies B-Z, because of some line buried in the middle of the legal text I didn't read? I.e. "The User consents to contact for any purpose by Website A and our partners", and a partner is defined as anybody who buys their contact list from them?
That is a case where the nature of T&Cs and end-user agreements makes the words "express" and "consent" more abiguous than they ought to be, since they rarely match anyone's definitions except the law's.
https://www.federalregister.gov/documents/2023/06/29/2023-13...
So illegal in the sense that artificial robocalls are already illegal, then.
From the declaratory ruling, any AI-generated voice call requires prior recipient consent:
> Consistent with our statements in the AI NOI, we confirm that the TCPA’s restrictions on the use of “artificial or prerecorded voice” encompass current AI technologies that resemble human voices and/or generate call content using a prerecorded voice. Therefore, callers must obtain prior express consent from the called party before making a call that utilizes artificial or prerecorded voice simulated or generated through AI technology.
This is where lawyers get to have fun. What is the line between a message in the public sphere copied and multiplied via broadcast, and a message consensually altered and multiplied via AI-then-broadcast?
Now, how is this relevant? Well, the entire cell network was offline, at least for some providers. At first it wasn't possible to send/receive data. Calls were connecting, but my friend sounded like an alien. Then for one hour, 0 communication was possible.
So even though the most efficient (I think?) protocol was used, it came to a halt
...How? How can this be enforced? What are the new tools? Based on the news release and documentation, fiat in this case means nothing but posturing, at most being hopeful some imaginary future tool will be able to bring execution to legislation.
Please don't interpret my comment to mean that Supreme Court decisions can't be criticized, I just don't find the "activist" accusation to be particularly insightful.
Flipping the hours d’ouvres table over on Roe v. Wade, a tense but stable compromise, that’s verging on activist. You don’t go knocking over fragile, workable standoffs that have held longer than an Ulster cease-fire if you can help it as a senior jurist.
It’s a pretty neo-neocon consensus to put it charitably.
It’s still the highest court in the land and it’s still binding, but I hope any thinking person is hoping for more a more“spirited but healthy” tension between major worldviews.
Roe v. Wade had been criticized for 50 years as an example of an activist judiciary and was held in place by rigid adherence to stare decisis.
Would you be as confident with stare decisis if we were talking about Plessy v. Ferguson, which held sway for 58 years before being overturned? Where the judges in Brown v. Board of Education of Topeka "activists"?
If "activism" is used to describe all sorts of political philosophy then it isn't really a useful term to bring to the discussion. I think it does have meaning though and that "activism" is not what conservative members of the court are doing.
Too much "Orwellian" language manipulation going on these days, IMHO.
> If your definition of "activist" is strengthening individual rights and refusing to give power to the federal government and its giant bureaucracy then I guess I'm OK with an "activist" Supreme Court.
Activism is activism regardless of whether it "strengthens individual rights" or not. It would be ridiculous to argue Roe vs. Wade wasn't activist just because it "strengthened individual rights". What rights? Rights according to whom? You could justify pretty much any decision this way.
I'm not familiar with the details of the case in Citizen's United, but whether or not it constitutes activism depends not on the effects of the decision, but the reasoning by which it was reached.
My comment wasn't very clear, but I was trying to communicate that recognizing that the Constitution is centered around the idea of a limited federal government with explicit powers is not evidence of inappropriate "activism". Instead it is evidence of appropriate checks on federal overreach.
I don't know what to make of your statement "activism is activism". To be a useful term, "activism" needs to mean something other than "doing something" or "doing something that I disagree with".
You can rule in a biased manner in favor or individual freedom or against it. My point is that that's not relevant to the definition.
I do agree with you however that the current U.S. constitution is centered around the idea of a limited federal government with enumerated powers and that therefore an unbiased interpretation of the constitution as written will tend to result in rulings that support individual freedom in general, though that's not a hard and fast rule.
But two wrongs don’t make a right and legislating from the bench for a net decrease in individual liberty via overturning a previous ruling runs contrary to the whole premise of stare decis (that’s Latin for “precedent” in case anyone missed that). Ruling from precedent has a lot of good properties, but maybe the best one is that it puts downward pressure on unbounded, escalating bench legislating.
Brown v. Board of Education was arguably overturning Plessy, though that was attached to specific language in the 14th Amendment, which is light-years from endless, muddy, subjective arguments around states rights like with Roe v. Wade. That question was settled by a war fought to a decisive military conclusion: red states don’t get to make draconian laws around individual liberty because we conquered them with guns. States rights “freedom” is subordinate to individual rights and freedoms via the landmark ruling in Union v. Confederacy. When it’s one interpretation of the Combined and Annotated Federalist Papers on one side and the armament of the high-GDP states on the other, well the victors make the laws. It’s a “happy accident” that the victors were on the right side of history.
Citizen’s United is the worst kind of judicial activism: tortured doctrines around corporate personhood being used to overturn a good law with good outcomes with broad bipartisan support (it was called the Bipartisan Campaign Reform Act for Christ’s sake) and it was a popular law that put downward pressure on the power of powerful institutions relative to the individual.
When the language of Locke and Rousseau is used to defend the life, liberty, and egalité of ruthless megacorps and and secessionist governors at the expense of the liberty of the common citizen: that’s Orwell hoss.
There's way more nuance that went into the decision than the "corporations are people" meme.
possibly. I don't think I know anyone who regrets it though.
SCOTUS is just one big game of political football. for centuries.
> [The Federalist Society] is founded on the principles that [...] it is emphatically the province and duty of the judiciary to say what the law is, not what it should be.
https://fedsoc.org/about-us
The federalist society is a well funded organization whose goal and track record is to install conservative judges to interpret the law in a very specific way, all the way up to the Supreme Court. I’d that’s not activism then I don’t know what is.
To preempt the answer you've given elsewhere in this thread "they rule in a way that you don’t like" isn't a good definition, and seems from my perspective like an attempt to muddy the issue in order to allow you to put activist judges on equal moral footing with those who actually follow the law. You are correct that its impossible to eliminate all bias, but responding to that reality by throwing up your hands and saying "I guess everyone's an activist then, judges should just ignore the law and rule based solely on their personal biases instead" definitely isn't helping the situation. There's a spectrum here, and The Federalist Society has as one of its explicit goals encouraging judges to move towards the non-activist side of that spectrum.
Everything I mentioned is not only a stated goal but has already happened. They are getting the results for which they continue fund raising. Now conservative Supreme Court judges are inviting new areas, such as when Thomas said he’s hoping for a new suit to shut down gay marriage after killing roe v wade. Trump even said he appointed judges specialty that will kill Roe v Wade. How is this not activism?
Roe vs. Wade wasn't reversed because the judges thought abortion was wrong (though they may indeed think that). It was reversed because Roe vs. Wade was a ridiculously convoluted ruling by an activist court that created a constitutional right to abortion out of thin air when a plain reading of the text of the constitution makes it obvious it contains no such provision. But if you're looking at this solely through a political lens and see it as the court "making abortion illegal" (which it actually didn't even do, it just reversed the previous ruling that was preventing states from enforcing their own longstanding laws on the subject) then its understandable why you might (incorrectly) see that as activism.
In contrast, a right to "religious freedom" very obviously does exist in the constitution, in the very first sentence of the bill of rights, and racial discrimination (affirmative action) is banned by the 14th amendment. There's room for debate as to exactly how broadly those freedoms were originally intended to be applied, but its not obvious to me on its face that those cases were decided incorrectly due to "activism" either.
The rest of your accusations are so vague that it's not clear to me what they're even referring to.
It isn't a particularly useful term because no one agrees on what it means. This has been illustrated quite nicely by the comments to my original comment.
The term "activist" is often interpreted as "legislating from the bench" where the judiciary usurps the role of the legislature. Some people actually want that. Other people don't want that.
Refusing to solve a problem and instead requiring Congress to clarify the law is another judicial philosophy. Is that being an activist?
Deciding that the federal government has no authority and that state authority or individual rights are more paramount is also a course of action that some people agree with and some people don't. Supporters probably don't call that "activism" but detractors might.
So I think the term is mainly used to slur your political opponent as opposed to being a succinct term for some particular judicial philosophy.
They are supposedly "originalist" except in their 2nd amendment rulings they ignore "A well regulated Militia" because guns is a Republican religion.
In banning Biden's student relief they ignored the text of the law and legislated from the bench just saying that it was unfair because the size was too "significant". Activism.
Last year they invented out of thin air the "major questions doctrine" to override the Clean Air Act and help polluters. Activism.
In 2022, without citing any principle, they said OSHA couldn't protect workers from covid. Activism.
When they gutted the Voting Rights Act they invented an "equal dignity of the states" doctrine. Activism.
To conservatives judicial activism is only a problem when liberals do it.
> In banning Biden's student relief they ignored the text of the law and legislated from the bench just saying that it was unfair because the size was too "significant". Activism.
I think the majority opinion explains this quite clearly: > The HEROES Act, Roberts emphasized, gives the secretary of education the power to “waive or modify” laws and regulations governing the student-loan programs. Congress’s use of the word “modify” means that the Biden administration can make “modest adjustments and additions to existing provisions,” Roberts wrote, “not transform them.” But the debt-relief program, Roberts stressed, instead “created a novel and fundamentally different loan forgiveness program.” The plan “modifies” student-loan laws and regulations, Roberts suggested, “only in the same sense that the French Revolution ‘modified’ the status of the French nobility — it has abolished them and supplanted them with a new regime entirely.” (from scotusblog)
It’s the masterpiece, the magnum opus of the Magnus of parliamentary politics. Nicollo Machiavelli doesn’t have shit on Mitch McConnell at that chess game.
I’m pretty indifferent to which color bumper-sticker late capitalism is sporting when it pushes the newest round of formerly “looking forward to better” working people below the waterline, it’s not a partisan thing.
The other team have plenty who match Mitch on evilness, but zero on skill.
> Whether the Court should overrule Chevron or at least clarify that statutory silence concerning controversial powers expressly but narrowly granted elsewhere in the statute does not constitute an ambiguity requiring deference to the agency.
The initial "overrule Chevron" seems like a DITF [1] and the latter is probably what the plaintiffs are hoping to achieve. Granted, I find it hard to trust a Court that overruled PP v. Casey, but most of this Court's other rulings, at least, have not been as extreme.
1: https://en.wikipedia.org/wiki/Door-in-the-face_technique
With enough evidence, operators are compelled to provide data and are given an opportunity to correct their action. If they refuse, FCC will eventually issue an order to all other providers to not accept calls from the bad actor.
Is this a humorous reference? Or is this supposed to be notable for some reason?
For now this could be seen as an incentive for TTS solution providers - build a product that is hard to distinguish from an actual human calling. In many cases the results are already convincing.
And what about the future. Please scan your retina to initialize the phone call? Please solve a captcha to start a phone call? Your workplace registered 12948230 calls in the last 24 hours, but employs only 3 workers registered as humans, pay fine now? Interesting times.
If not - why would advertisers pay for that? If yes, that feels like a job and not like my personal email account - I wouldn’t want that.