The FCC wants to criminalize AI robocall spam | Hacker News Reader

The FCC wants to criminalize AI robocall spam

theregister.com

229 pointsmikece138 comments

If a carrier delivers a spam text or call, if they can’t definitively prove who sent it, it should be treated as though the carrier themselves sent it (and fined accordingly). Watch how fast telcos mandate authentication and other KYC measures after that.

I don't want that?

How is this a better world than getting spam calls

How is it not? It'll cut down on spam and other antisocial behavior (e.g., unsolicited intimate images).

I'm not sure why you want to connect that fine to kyc-style behaviour. Just apply the first part and fines. Essentially "you pay the fine unless you can forward it to the entity that originated the call". That would solve the issue just fine.

Telcos already know where the connections come from and who to bill for what. Someone just needs to mandate it.

I would structure that differently: the telco pays the fine immediately, but can collect with damages from the source. That ensures that consumers are compensated without delay and gives the telco a strong incentive to be serious about collecting and/or blocking anyone without a robust legal presence. Most of the problem goes back to the decision in the VoIP era that checking ID would slow revenue growth so that needs to be robustly corrected.

This might work for domestic calls, but what about international ones? Do you really want to force telcos to block all calls from countries that you cannot collect fines from?

They can collect the fine themselve, by billing it to the other country operator.

The US market is big enough to essentially cause its own "Brussels effect". (It already did in other areas) If the choice for some external telco is "deal with spam or lose access to call the US", any legit company will choose the first one. Or they can follow the same fine collection rules themselves.

> Do you really want to force telcos to block all calls from countries that you cannot collect fines from?

I'd suggest it be optional on a per-customer basis, perhaps with a whitelist.

E.g., the only international calls I ever want to receive comes from family members' cell phones when they're traveling.

I'm 100% okay with blocking all other international calls. If something is that important, they can send me something via mail, FedEx, etc., or send a message courier to my house.

Most countries will not have a problem with it since they want to have the correct data anyway and are not a source of SPAM. The countries that are not fine with it are countries most of the US doesn't want to deal with anyway.

Just make it so you can opt-in to an exception and nobody will care. If you live in the EU and want to call me (as if you know my number, and have reason to call me) you won't have a problem. If you live in Russia you might have a problem - but I have no reason to think anyone in Russia will ever call me so I don't care. The few people who do have friends in Russia can add an exceptions and those exceptions are not enough for spammers to bother operating at all anymore.

I'd love if receiving international calls only worked on an opt-in basis, maybe opting in for each originating country. I think most people in America would have no reason to opt-in. Of course, American corps using overseas call centers would be inconvenienced, but screw them anyway.

Yes. They don’t peer with every other company in the world directly so it would be written into the existing contracts between the major companies, who would collect from their smaller peers.

This already works: a staple of these threads is some European user who has sender-pays service expressing surprise that this is a problem because it’s so rare in their experience. Those countries can still receive international calls.

I imagine instead of blocking you'd see peering agreements with very stringent terms around allowable portion of calls fined as spam.

This is one of the views I'm coming to.

I'd also suggest that there be a rebate per spam call delivered. Say, $10/call to the customer.

Telcos could seek 10x that from the upstream who peered the call.

That is, there's a protection and credit to both the end-recipient and the telco provider, which might make the politics of such policy or legislation more widely palatable.

I want emails, phone calls, and texts sent to me to cost the sender some fixed amount of money… which is refunded only if I send a return token confirming it was not spam. For previous contacts or people in my organizations this refund would be automatic.

This is how linkedin cold outreach works.

>I want emails, phone calls, and texts sent to me to cost the sender some fixed amount of money… which is refunded only if I send a return token confirming it was not spam.

I've thought about that type of spam refund scheme many times and I like it but the potential problem I see with it is the unintended effect of even more spam that you can't ignore by the companies you already have a business relationship with that bill you monthly.

E.g. You get a monthly bill from Comcast to pay for cable tv. With a new hypothetical "spam refund" law, they now abuse it to their advantage by adding a new "Customer Information Updates" fee for $5 to the bill. (They won't call it "marketing fee" or "promotional fee" but something innocuous like "information fee".) The cable company can now can send you more spam by using the "spam refund" law against you.

But... the customer can get a "rebate" on that extra $5 fee by responding to spam and "refunding" it back to the cable company. If the customer chooses to ignore the spam and never refund it, the customer ends up paying the $5 extra fee on the monthly bill. The customer is now paying oneself to to receive more spam that they can't ignore.

A potential spam refund scheme needs protections to prevent abuse like the above.

As the commenter you’re responding to said, just add the cable company to your contacts and the refund is automatic. Opt out of any emails from them other than billing notices.

>, just add the cable company to your contacts

No, people don't want to pollute their contacts listing by adding every company they have a billing relationship as a new entry to "manage spam". It's visual pollution to have "Comcast" as a useless entry alphabetically in between "Charlie" and "David" just to offset a new spam rebate scheme because Comcast is abusing the law. And if Comcast has multiple identities such as creating "Comcast Updates" and "Comcast New Channels", etc to further abuse the spam rebate scheme for more customer "engagement", you're doing even more digital housekeeping and adding more entries to the contacts listing.

Another problem is that the smartphone's "contacts listing" is a special area that has downstream interactions with other smartphone settings to manage/filter notifications and sounds. E.g. "silence notifications not in contact listing" or "block calls not in contact listing". By adding Comcast into the contacts listing, they can become even more invasive in your life by polluting alerts on your home screen and making your phone ring.

The spam rebate idea can have some weird unintended side effects if such a scheme (or law) is not crafted carefully to prohibit abuse and make life worse.

It would be simple to have a ‘whitelist’ for contacts that are free but low priority and get muted, put in a special folder, or auto deleted. It’s cold contacts from anonymous senders that are impossible to reliably auto screen.

Noting that the way we currently do things would be insufficient to embrace a new idea is not controversial or an idea killer.

Yes, we would need to update how we handle transactional emails. No, this would not need to be difficult or even frequently user facing.

Email providers/phone should have hierarchical contact lists. Comcast can go in the lowest rung, maybe call it subscription services, and your spouse is in the highest rung, in the family group. Each rung comes with its own abilities with respect to notifications, calls, sounds, etc. Have a button at the bottom of every email to add the sender to a group if not already there, and that would handle your spam rebate scheme automatically.

I'd argue strongly against your premise: people specifically do want to note their established vendor relationships (subscription services, doctors' offices, schools, government services, utilities) if only so that they can distinguish legitimate from illegitimate contact attempts, but also to manage different groups of known contacts differently.

This presumes a few advances, such as reliable caller-ID systems (this is presently not the case for North Amercian dialing systems). Friends who are innundated with robocalls noted that they'd received a call spoofing an entity with whom they do have a relationship, however it was clear from the call characteristics that it was not a legitimate call --- among other factors, the caller identified themselves as being from a different entity, which is a pretty low bar.

The problem of mega-services (financial, comms, federal government, etc.) being subject to spoofing simply because so many people have interactions / relationships with them is an extant problem. But odds are pretty good that your local water / sewer / trash / gas / electric service will be less universal, and knowing that they're calling when they do in fact call is useful.

It seems you're associating a contacts list with a friends list. That's not the case. Your contacts are, well, your contacts, and different contacts have different roles. Among other factors, you might set, say, different contact rules, priorities, and ringtones for, say, immediate family, work, casual social contacts, and business entities. The latter would generally not be permitted to call outside regular business hours, and you might specifically restrict them around mealtimes or other inconvenient times of day (redirecting to voicemail or another messaging service, say).

The problem with this is the monopoly position your cable provider Comcast has over you.

I strongly dislike this idea, because it would make people I want to hear from extremely hesitant to use email, phone calls, or texts.

I don't think so. The idea is that you could set the fee so low that a normal person would never ever encounter it, but when multiplied against the number of bulk emails a spammer sends out becomes a real cost. And at that point you could allot everyone a small credit sufficient for even vigorous personal use.

You could also have a periodic forgiveness of de minimus amounts.

The Public Access to Court Electronic Records (PACER) system charges 10¢ per page. Every month, balances under $10.00 (edit: apparently it's $30.00 per quarter) are forgiven. It strikes me that a similar model could be employed. Are there people who send more than 100 unsolicited texts per month? More than 200?

I didn't know about that, thank you, and yes exactly. The difference between normal use and spam is so marked that any system of charges could be set such that no normal user was ever even aware of them.

And to be clear I'm not talking 'normal' as in 'median' I'm talking 'human using their thumbs'. Even if you spend all day every day hunting down craigslist bargains and coordinating swaps

True this, and a far lower ding, closer to say, $0.01 than $10, could still be quite effective.

Keep in mind that egregious robocall providers have been making billions of calls annually (55 billion in 2023: <https://www.techdirt.com/2024/01/16/americans-received-55-mi...>), whilst a typical instance was unable to pay a $10 million fine (<https://therecord.media/ftc-settles-with-company-that-facili...>). At that rate, a per-call ding of only $0.0002 (2 hundredths of a cent) might prove sufficient. A penny-per-call penalty should actually be relatively effective against the lowest of the bottom-feeders, though a higher rate would afford additional protections.

Honestly? This

The American phone pricing model, charging for "airtime" is absurd

It means that a spammer can pay $10 for 1000s of calls while the same customer receiving the call pays much more than that per minute

It is frankly ridiculous

They would just jack up prices to account for the additional costs and then they would never actually pay you that $10

percentage of total revenue rather than flat fee, so there is no price shifting to the user, and its cheaper to just comply with no malarchy

> if they can’t definitively prove who sent it,

They already can. There's an internal "billing" code for every call so carrier knows who to bill for having carried that call.

They don't want to "prove who sent it" because all those billing transactions for carrying those calls add up to significant profit, and they don't want to kill that profit if they don't have to.

This is literally the 'carrier liability exemption' that section 230 is based on, HN gets all up in arms whenever carrier liability is threatened to be taken from internet sites...

Should just apply to Telcos over a certain subscriber size. So, the Big 3 only. Don’t want to squash new competition into the market with new regulations.

Can they make it criminal to send spam SMS as well? I get multiple sms from political candidates (that I didn't sign up for, in fact if they address me by name it's the wrong name) every day. They always come from a new number so blocking doesn't work. Responding "STOP" doesn't work either since they hop numbers constantly and ignore stops. There is no way to stop the barrage of text messages and it's driving me insane.

In australia, political messages are exempt from anti spam and anti junk mail laws.

Politicians think they’re above everyone else, which evidently in this case they’re literally above the law.

I’ve had Amnesty International contacting me at work, using the “exemption for humanitarian emergencies” to say “We can’t remove you from mail lists, it’s a humanitarian emergency.” Highly annoying to be doxxed by who you’ve been donating to.

Actually, there is a solution: Say that you are back in Europe and a EU citizen. The GDPR applies to them too.

You actually just need to say you're an EU citizen, GDPR applies globally.

Nothing in GDPR says anything about EU citizens. Everyplace it talks about data subjects it talks about data subjects "in the Union".

If he is not "in the Union" and the processor or controller processing his data is not "established in the Union" then GDPR does not apply. See Article 3 [1] for details.

[1] https://gdpr-info.eu/art-3-gdpr/

I have an OG mac.com account (got it about five minutes after Steve announced it).

Because of that, a lot of folks enter my email address as theirs; often accidentally (they forget the numbers, afterwards, or somesuch).

One lady made donations to her local ASPCA, and used my email address.

They sold me to some of the craziest, most radical-left organizations on earth.

I get hundreds of emails, every day, from these outfits. Some, are absolutely barking mad. Many, are fake emails from politicians.

Even I can't stomach some of the crazy in these emails, and I lean left[ish] (I'm quite centrist, which means leftists think I'm MAGA, and rightists think I'm commie).

Needless to say, unsubs only make it worse.

All because one lady goofed, when entering her email, for a local animal assistance org.

> Say that you are back in Europe and a EU citizen. The GDPR applies to them too.

Any idea how GDPR works for EU citizens who reside in the USA?

E.g., suppose I want Facebook to forget about me, but I'm using a U.S.-based ISP to tell them that. Does Facebook have any legal justification for rejecting my request?

If you reside in the US, then Facebook can probably argue US laws apply. Over course if you go to the EU to visit family for Christmas or something (very likely) and ask Facebook to stop then, EU laws apply and they need to forget everything even though you will return.

Note that data generated only from your time in the US about your activities when in the US probably can still remain.

> Any idea how GDPR works for EU citizens who reside in the USA?

It doesn't. The cases it applies to are given in Article 3 [1]. It applies to these cases:

• Processing of personal data by entities that are "established in the Union", regardless of where the processing takes place.

• Processing of personal data of "data subjects who are in the Union" by entities not "established in the Union" if it is related to those entities offering goods and services to data subjects "in the Union" or those entities are monitoring behavior of those data subjects that takes place "in the Union".

• Processing of personal data by an entity not "established in the Union" is they are somewhere where Member State law applies "by virtue of public international law". (Anyone happen to have a list of such places?).

GDPR does not even mention "citizens". Every place it talks about data subjects it uses "in the Union".

Note that this works both ways. People who are not EU citizens are covered if they are "in the Union".

[1] https://gdpr-info.eu/art-3-gdpr/

Same in the US: politicians get an exception from the Do Not Call list and similar. Complete garbage.

I looked this up yesterday. The regulations are around the use of robodialers but the political messages generally aren’t using robodialers. Instead they use an army of people with burner phones to send messages person-to-person to avoid the regulations.

No, they are using robodialers, these days. Also, I have gotten a few robovoice calls from local pols.

One issue that corporations should be concerned about, is if they contract a dialer company to initiate their calls, and that company moonlights to scammers, they get blocked, so the legit corporation gets blocked for being a scammer.

Why do you think they are using robodialers for text messages? AFAIK, that’s still banned by the FCC:

https://www.fcc.gov/rules-political-campaign-calls-and-texts

They spoof the sources. The carrier still lets them through, but sending STOP results in a delivery error.

A certain second place candidate who has associated my wife with my phone number is doing this to me. I forwarded them to the FCC.

In the UK, there is a growing group of people who voted for Brexit who are now claiming that they were not informed of what was going to happen.

The UK government sent a white paper to every household with their opinion on it. If there wasn't a special case for this, many households wouldn't have received this. these things can be abused, but paraphrasing patio11 "the optimal amount of fraud(/abuse) is non-zero" [0]. The argument for this applies in this case to.

[0] https://www.bitsaboutmoney.com/archive/optimal-amount-of-fra...

Assume that if I have a ‘no junk mail’ sign, I either will be informed some other way that doesn’t involve unwanted propaganda, or I do not care.

Anyone with the mentality ‘this junk mail is more important that your junk mail sticker’ can go fuck themselves - similar behaviours in the physical realm include rape.

> I either will be informed some other way that doesn’t involve unwanted propaganda, or I do not care.

These people said they didn't care, and they're literally in the news saying "why didn't anyone tell me?"

> similar behaviours in the physical realm include rape.

There's stretching an analogy, and there's comparing junk mail to rape. This is a first for me.

> These people said they didn't care, and they're literally in the news saying "why didn't anyone tell me?"

Then they should either read the news or remove their no junk mail sticker? Attempting to suggest that political mail would have swayed brexit is stretching the analogy pretty thin, too.

The same laws benefit you as well as politicians. You can pick up the phone, call your neighbor, and discuss politics at any time without fear of reprisal from the authorities. Try doing this in Cuba or China, you will end up in prison. In my view, every citizen in a democratic country is not only permitted to but obligated to participate in the political process, which is why political activity is exempted. If you don't want to participate, if you don't vote, or inform yourself of issues, you still benefit from all the same freedoms & protections of the law as those who do participate.

This is an important point, because the exemption of political activity is due to the protections of political speech. Imagine if it were not this way, the party in power could simply fine or even arrest their opponents for trying talk to people over the phone. This happens every day in nondemocratic countries. This protection should not extend to AI though. AI doesn't vote. AI doesn't have a voice in a democracy. So allowing AI to benefit from this exemption should not be permitted. I tolerate PEOPLE calling me, because I am a voter and they are voters and we must all discuss the direction our laws should go or who our representatives should be (when we have time). But AI should have no place in the political process. Not for robocalls, not for choosing representatives, and especially not for lawmaking.

> the party in power could simply fine or even arrest their opponents for trying talk to people over the phone.

Yeah, that would never happen in a democratic country…

These laws don't benefit me in anyway. I hate recieving texts from politicians begging for money using a loophole they have written for themselves.

I also don't want random people to be able to call me and yell at me about politics.

Enabling phone spam is in no way required for a functioning political system (not that we have one of those.)

Phone spam doesn't inform people on issues, doesn't create meaningful debate on issues, and has no value in strengthening democracy.

Every single candidate that sends me spam loses my vote permanently.

Freedom of speech != freedom to robocall.

Calling your neighbour is different to being paid to text or call thousands of people daily.

Same in the US, unfortunately.

This has been driving me crazy too. I don’t know how but I got on some Democrat list and they won’t leave me alone. Is there some central database they use that I can ask to be removed from?

It’s so bad, and egregiously dishonest at times, that I half suspect it’s actually a campaign to earn a red vote out of spite.

There is a legal loophole campaigns use where if a text message was sent by a human it’s not considered spam. So of course a program exists that dispatches the message you want to send to a call center and they dial the thousands of numbers you want by hand.

I remember a time when I'd pick up every single call received... now it's the opposite, I only have it ring on recognized phone numbers + I skim voicemails, filter out the ones that leave the 3-4s of silence or autoplay text, and call back those who actually have business purpose with me.

If there exist AI robocalls that start spoofing my friends and loved ones... ugh, I can just imagine the hassle of doing due diligence for every voicemail going forward. (This does kinda remind me of the scams where someone pretends to be a kidnapped distant family member, cries in the background of the call, and asks for ransom money....)

Same. If you're not already in my address book, I'm not answering your call. I don't have voice mail set up either.

That's not realistic for many people. I don't have every doctor's office that I might come in contact with in my contact database. Nor every emergency service that may contact me because of an elderly relative or someone else. And presumably you don't have a landline either so you're simply choosing to make yourself not contactable.

Which isn't a reasonable option for a lot of people assuming there are alternatives.

That's fair. I don't have any elderly relatives I'm responsible for, but if/when that situation arrises I'd probably try to set up contact records for their caregivers and providers. It's a shame that scammers and spammers have ruined our telecommunications system and that the telecom companies haven't done anything about it.

You probably should have at least an option to go to voicemail.

I do get messages from offices I don't have in my contacts database that I generally want to receive. Ignoring total junk is mostly fairly painless at least for me.

The fallback is that those callers should leave a voicemail.

At present, junk callers virtually never do, though that of course may well change.

The person upthread wrote that they also don't enable voicemail.

Point.

I'm ... strongly antagonistic to voicemail myself. But even if you don't check/listen to messages, simply noting who's left a message is a pretty good screening method. For most medical comms, you're unlikely to have a meaningful message left other than "call back" in any regard, so enabling but not listening too hard is viable.

Many medical systems now have some sort of electronic patient record which includes medical staff (MD, RN, PA, etc.) messages, though that's no silver bulet either.

> I'd probably try to set up contact records for their caregivers and providers

Won't work. I have both elderly relatives and young children, and it's possible to get calls about them from just about anyone in their respective facilities. It's not possible to try to find and list the phone of everyone who ever watches over them. Including substitutes and temps and so on.

I agree.

It is frustrating, though, that they require you to provide an exhaustive list of people who can contact them but they are under no such obligation to do the same for you.

I find it easier to set up voicemail. Unknown callers that care will leave a message. That message is speech to text processed and I can decide if to play it or call back. Anything important enough gets through even from unknown numbers. Never have to answer a spam call.

well the telcos should have thought about that before agreeing to carry packets without validating headers

Doctors have no business calling, medicine isn't done through phone. Contact is possible: when I don't answer spam calls from banks, they send me sms stating their business.

Meanwhile, in the world where I live, doctor's offices call me on a not-frequent but not-rare occasion to reschedule appointments and the like. SMS is increasingly used for a lot of routine stuff like reminders but there's still a fair number of phone calls.

Per another comment, my suburban hospital system has merged with one of the two big city systems and as a patient the electronic health records system saves quite a few phone calls and faxes being sent around. But there are still some calls.

When a phone is receiving dozens of calls per day, answering each and every simply isn't viable, not matter how critical a given call might be.

Unfortunately, it's often older people with a greater dependence on healthcare providers who are also targeted by robocallers, fraudsters, and telemarketers, often to devastating effect.

I've been mulling over an idea for some time: set up an answering machine on freepbx with an IVR that requires the caller to push a button before it's forwarded to a (ringing) phone, otherwise sayonara.

That should cut off most robocallers, shouldn't it?

Only if so few people use it that spammers don't bother to target beaking the scheme.

Google Voice does this. It makes a human press a number to be connected.

I get almost 0 spam calls on my GV number, but my carrier number is garbage. I swear they sell my ph# to spam orgs on registering.

Android has a call screening feature that's amazing. I turn it on for any unknown number. I haven't missed any calls I wanted to receive, and everything else fails the screen.

Hopefully that doesn’t filter out important calls from unknown numbers too like job offers.

Is it difficult for the carriers or the phone makers to determine if a message is empty, and if so, delete it? This should be an option I can turn on.

> Current laws require telemarketers to have explicit consent from consumers before they can make automated calls using "an artificial or prerecorded" voice. Rosenworcel believes the same rules should apply for AI-generated robo-calls too.

I'm confused by this part. How is AI not already included the current law of "artificial or prerecorded"?

> FCC Chairwoman Jessica Rosenworcel today proposed that the FCC recognize calls made with AI-generated voices are “artificial” voices under the Telephone Consumer Protection Act (TCPA),

The FCC chair here is not saying "we should make a new law saying AI is artificial and illegal for robocalls" but saying "in our opinion AI calls are already classed as artificial under this well-tested law. Please go out and sue some people under the TCPA, here are tools to do that and we'll support your efforts in doing so."

Basically the FCC thinks the existing law covers it, but, it hasn't been tested in a court yet.

I never get robocalls on my private phone, but probably get 2-3 a day on my work phone.

Most are easy to detect, simply because they're foreign numbers.

What is worse, are the spoofed numbers. Because it could be related to work, I have to answer those.

I've lost count how many times I've called someone back, mere seconds after missing the call, only to get some frustrated person on the other end:

"I'm so sorry, people have been calling me all day, asking why I've tried to call them"

"I don't know why my phone number is being used for this"

"What, I didn't call you? You must have the wrong number"

etc.

This is specifically about AI-generated voices you'd recognize, like spoofing Joe Biden's or Nancy Pelosi's voice, for example.

I use to prank the telemarketers but unable to with robo callers. One a woman called wanting me to switch long distance carriers. I halfway listened to her spill for twenty minutes while watching TV. Finally she wanted me to switch. Told her I couldn't for I was house sitting. She asked about my home number. Told her I only had a beeper. She asked me why I let her talk so long. Told her she had such a lovely voice and I enjoyed listening to it. She slammed the phone down on me.

For years I got a spam call every morning during my commute to work. I found Lenny the automated old person who would keep callers on the line from anywhere from 2-20 minutes.

It really was fun listening to callers freak out when they realized what was happening.

I'm on the federal do-not-call list. Isn't it $500 per call, and $1500 if they don't give a business address when asked? I get at least 6 per day on average. Occasionally I don't get any.

I stopped answering my phone a the 12 per day peak. I figure if it's really important they'll leave a message.

I'm not in any list and get like zero per day. One a week maybe. How am I so lucky?

> Isn't it $500 per call, and $1500 if they don't give a business address when asked?

Sure. How should they collect that from the overseas scammers who called you?

Internationally sourced calls should be opt-in, this would solve the non-political portion of the spam problem. Almost no one needs to receive calls from overseas.

I would applaud if FCC goes even further, and bans the robots from use in any business, which is required to accept calls

For example: government forces me to have health insurance, and health insurance is unable to satisfy my needs online - DON'T FORCE ME TALK TO YOUR ROBOTS. These don't typically recognize my voice, and frankly speaking "How can we help you? You can say anything" sounds like an insult to me. Many of the businesses don't even offer touch-tone navigation anymore. It used to be that "Say blablabla, blablabla, or blablabla" meant you could use "1", "2", or "3" respectively, but now even that is no longer working in many cases.

Hire f*ng human, or don't offer phone line at all.

As someone who lived in the US and now in Germany I just want to tell my American friends that this is not a problem in Germany as far as I know. My German number never gets any spam calls and my friends never talk about spam calls. It's the corporate greed that gets in the way of doing the right thing over there...

It's not like us Germans are free from scams... we have Turkish callcenters who scam dozens of millions of euros [1].

[1] https://www.tz.de/muenchen/stadt/hallo-muenchen/callcenter-b...

I believe you, scam is everywhere. What you experience in the US is different. In the news article you sent it's obviously something that they could track down and sentence the fraudsters to jail. In the US you get multiple calls from those robo-callers per day and you practically can't do nothing about it. This is why Apple added "mute unknown callers" to iPhones. This is the best strategy most people have at this point.

In the US I haven't had such a call in more than a year. The US already cracked down on spam calls (that is routing spam calls). You may remember complaints about how bad things were a year ago though, but things change.

What is new is now AI can fake voices and conversations. This makes the cost of such SPAM is lower, and also you can frame someone else. Both of these give incentive to find work around to the current systems (no system is perfect)

That could be selection bias. Not something that the telcos did.

Many of these scams are run out of India and Nigeria. They teach English as a second language fairly young. That lets them cheaply target English speaking countries. I am not saying it is because of that but if I was running a scam center (I am already looking for 'the easy way') I am not targeting something where the pool of people I draw from need to know another language. One scambaiting call I watched they had gave the person running the scam had a laptop from about 10 years ago with 4GB of RAM and a very low spec processor. In other words cheap and easy to get. Low monetary risk with high monetary reward.

The core issue is the system allows spoofing and poor verification of who is getting blocks of numbers. Also low enforcement of laws in originating countries. The laws actually force the telcos to sort of do this too. As they in many cases must route. But the reality is we have bad actors that we do not want to route. But that same tool to not route could be used for reasons like what many tech companies currently use to get rid of people they do not like for whatever reason.

> It's the corporate greed that gets in the way of doing the right thing over there...

That's absolutely bonkers. "Corporate greed" is a fuzzy, unprovable, politically-charged and highly unlikely to actually be true catchphrase that means nothing. The fault lies with the US government - there's nothing that prevents it from acting in the interests of their representatives and implementing anti-spam laws, like you did in Germany. There's no difference between German companies and US companies that somehow makes one "greedier" than the other, either. This comment is a nothingburger that, quite honestly, does not belong on HN.

Unless I'm reading it wrong, the TCPA does not define or allow the FCC to define a crime. It's purely a civil action. So the headline is misleading.

Meh feels like something without teeth that won’t really help. I’d rather see the FCC require telcos & celcos to require caller authentication so that spam calls can be traced to their source and cannot be anonymous, and so that spammers can no longer spoof numbers. I’m cynical about this and I’d love to be wrong, but get the feeling the FCC doesn’t dare to do things that will make a real difference, or worse, actively looks the other way, since the service providers make money on spam calling and will throw a fit if regulation gets in the way of their profits.

Why were Robocalls allowed to begin with, back when they were taped messages? From the beginning you couldn't screen calls to your home (no called ID) so it must have been infuriating to get up from the couch to answer a robocall?

If regulation had been sane from the beginning maybe this discussion wouldn't be necessary (I can't imagine robocalls are legal in many other places? I haven't gotten one in my entire life).

Before the big Ma Bell split up, due to them being a monopoly, even local calls were often charged.

And long distance rates were horrible, and there were none of these independents.

Back then, there wasn't a reason, because such call devices cost hard cash per call. And it's really only gotten super bad the last decade.

> Why were Robocalls allowed to begin with,

Because money. Duh.

I don’t follow. Companies don’t vote but voters have to answer robocalls.

Companies vote with their $$$, and are happy to donate to any and all political parties to do their bidding, so

a) Unless it's a core part of what differentiates party A from party B, voters won't pay that much attention (compared to other issues), so parties can have the same (lack of) policy that favors the companies, to keep those donations flowing.

b) If the parties all have the same policy, there's no-one the voters can vote for to work in their interests on that issue, so they'll vote for whoever they were going to vote for anyway on the other issues.

This sounds like a problem of having a) made money a core part of politics and b) having a two party system.

They were a loophole nobody expected so they didn't mitigate it. Robocalls would not have been affordable in the 1970s as long distance rates were so high - you would have needed to setup a call center in each city you target as long distance rates between cities was hundreds of dollars per hour. That means pay big city labor rates not cheap foreign labor.

Only with IP calling is calling from country to country cheap enough that robocalls can be worth doing.

Doubt it will do anything to curb the current level of spam, not to mention its transitioning from calls to text, now that people have learned to just not answer their phone.

A2P/10DLC registration seems to have curbed SMS spam for me significantly. As annoying as it was to comply with, I think it's useful (so far?)

What is that?

A2P = Application to phone, essentially any SMS/MMS sent to a phone by an app or automated process, and not by a human just texting.

10DLC = 10-digit longcodes; in the US (well, NANP), "regular" phone numbers are 10 digits long. This is as opposed to shortcodes, which are usually 5- or 6-digit numbers (though there are some that are shorter) that are sold to specific customers after an approval process where all US mobile carriers have to sign off on their use cases. If you spam, you get your shortcode revoked. They generally cost on the order of $1000+ per month, while you can usually get a longcode for $1/mon or less.

Over the past few years the US telcos have (due to regulatory action, not of their own choice) started requiring that anyone using 10DLCs for A2P use cases need to register: who they are, how responsible parties can be contacted, and what they plan to use the numbers for. Don't do this and your messages will likely be silently dropped.

I think it just comes in waves. Since the beginning of the year, I've received more spam texts and calls being silenced by my phone than I did all of last year.

All my "unknown sender" texts get filtered to a separate view, and do not create a notification. If I'm expecting a text from an unknown number (such as a 2FA code) I'll just open the "Unknown Senders" view. The rest of the unknown messages are ignored.

Unfortunately on iPhone you can't set some messages to unknown sender and some spam calls seem to be in known senders

Do you have “filter unknown senders” turned on?

Yes and that is how I know it does not work. There are senders I cannot get into known.

I keep hearing this same news story but nothing is happening. Maybe if this doesn’t get enough momentum, finding the phone numbers of the FCC chairmen and telecom company presidents and signing them up for spam would do the trick. It’s one of those things that would aggravate one person but would benefit the rest of the country.

Previous link: https://news.ycombinator.com/item?id=39218669

I still get invalid number calls sometimes and I use Google's PixelAI screening. It doesn't even show me the option to screen the call manually. Also, if a robocaller calls twice then they bypass my Pixel's screening feature.

I still wonder why this isn’t a problem in Germany, i get at most one spam call per year.

Here's some clues:

https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(PPP)...

https://en.wikipedia.org/wiki/List_of_languages_by_total_num...

[dupe]

More discussion: https://news.ycombinator.com/item?id=39220462

They barely scraped thru human spam, they better ban this sht

See "Computer Fraud and Abuse Act."

Thank you! I welcome this.

They need to ban the software and equipment that let's them make the calls.

I don't see how that would work? Isn't this all general purpose stuff at this point?

The ingredients to make an explosive are too, but we've made that illegal, I think we can figure out something

It would be nice if we could hold the carriers responsible. If they have to start paying penalties from allowing spammers on their network, things might change.

Software which allows robocalls is ~10 lines of code these days. You can't ban that idea. You can create your own in under an hour if you've dealt with VoIP systems before.

Making an explosive device or 3d printing a gun is similarly easy, but illegal

Lock picks are just some metal wire, but are illegal unless your a locksmith to carry around.

Quite a few issues with this comparison: That's heavily dependent on location and usually requires the intent as well. I'm not a locksmith and have a set. They're also quite a bit more involved than a metal wire - you're not making your own quickly.

Then there's multiple use. There's no/minimal difference between a script doing automated support callbacks and spam calls. You take a list of numbers, dial and connect the other end, handle hangups, do some reporting on connection success/failure. The completely isolated "other end" is either a person or an automated message or a combination of both.

There's really no reason to deal with any code here specifically, rather than just making the result illegal and enforcing that.

Which is what they do with lock picks, no?

He he he, because it's AI and new and shiny and they want to be seen actually doing something by the public eye.

Where have they been all this time that regular spam calls have been a thing lmao

The solution to this problem doesn't have to be complex at all.

Conceptually (meaning, all details are not covered here):

  - The caller/message-sender pays a per-call fee
      - For example, $1.00 flat fee
  - If the caller is in your address book, the charge is $0
  - If you call them back, the charge is $0
  - If you approve the caller, the charge is $0
  - The person receiving the call earns a portion of the fee (30% ?)
  - Etc.

If implemented correctly, on average, no legitimate caller pays this extra per-call or per-message fee. Machine learning could be used to augment effectiveness and reduce or eliminate fees when two people or entities know each other.

For example, a company receives tons of customer service calls. Most callers might be unknown to the company (think Amazon customer service). There has to be a mechanism through which callers don't get stuck with a $1 fee just for calling to get support. This is where a detailed design of this type of a solution would be necessary.

What this does do, I think, is create a situation where it would (should) cost spammers millions of dollars to reach millions of people with unwanted calls or messages.

Another element of this is that consumers would make money from these stupid calls or texts. They should. If someone wants to consume somebody else's time, they should pay for it.

Which brings-up another twist to this idea:

Maybe people should be able to set a rate for unsolicited calls or messages. I could, for example, set my rate to $10 per minute and a 15 minute minimum. If you call me and I don't know you, the cost to talk to me or leave a message is a minimum of $150. If I call you back, the charge is negated. I can also tell the system not to charge you. Etc.

Upon making a call, the caller hears:

  "This number charges a $150 for anonymous calls.  Press 1 to accept."

If it is a message, they receive a reply:

  "This number charges a $150 for anonymous messages.  Reply YES to accept."

The receiver gets a pop-up that might say:

  "Unknown caller: <insert name or number>."
  "Accept and charge $150 fee"
  "Accept and charge $0"
  "Cancel"

Details would have to be worked out, of course. There are lots of ways to implement something like this such that legitimate callers are not charged any fees. Legislation and a technical infrastructure would be required to drive such a system.

The reason this nonsense persists is because the cost is low and potential return on investment is high. If that equation is flipped, the nonsense stops.

Reply on news.ycombinator.com