Google Executive Addresses Horrifying Reaction To Uncanny AI Tech (bloomberg.com) 205
The most talked-about product from Google's developer conference earlier this week -- Duplex -- has drawn concerns from many. At the conference Google previewed Duplex, an experimental service that lets its voice-based digital assistant make phone calls and write emails. In a demonstration on stage, the Google Assistant spoke with a hair salon receptionist, mimicking the "ums" and "hmms" pauses of human speech. In another demo, it chatted with a restaurant employee to book a table. But outside Google's circles, people are worried; and Google appears to be aware of the concerns. From a report: "Horrifying," Zeynep Tufekci, a professor and frequent tech company critic, wrote on Twitter about Duplex. "Silicon Valley is ethically lost, rudderless and has not learned a thing." As in previous years, the company unveiled a feature before it was ready. Google is still debating how to unleash it, and how human to make the technology, several employees said during the conference. That debate touches on a far bigger dilemma for Google: As the company races to build uncanny, human-like intelligence, it is wary of any missteps that cause people to lose trust in using its services.
Scott Huffman, an executive on Google's Assistant team, said the response to Duplex was mixed. Some people were blown away by the technical demos, while others were concerned about the implications. Huffman said he understands the concerns. Although he doesn't endorse one proposed solution to the creepy factor: Giving it an obviously robotic voice when it calls. "People will probably hang up," he said.
[...] Another Google employee working on the assistant seemed to disagree. "We don't want to pretend to be a human," designer Ryan Germick said when discussing the digital assistant at a developer session earlier on Wednesday. Germick did agree, however, that Google's aim was to make the assistant human enough to keep users engaged. The unspoken goal: Keep users asking questions and sharing information with the company -- which can use that to collect more data to improve its answers and services.
Scott Huffman, an executive on Google's Assistant team, said the response to Duplex was mixed. Some people were blown away by the technical demos, while others were concerned about the implications. Huffman said he understands the concerns. Although he doesn't endorse one proposed solution to the creepy factor: Giving it an obviously robotic voice when it calls. "People will probably hang up," he said.
[...] Another Google employee working on the assistant seemed to disagree. "We don't want to pretend to be a human," designer Ryan Germick said when discussing the digital assistant at a developer session earlier on Wednesday. Germick did agree, however, that Google's aim was to make the assistant human enough to keep users engaged. The unspoken goal: Keep users asking questions and sharing information with the company -- which can use that to collect more data to improve its answers and services.
Ian (Score:5, Insightful)
Re: (Score:3, Interesting)
I mean, packaged foods have to tell you about their ingredients, fish have to be labeled with country of origin....I think that a computer call should be forced to ANNOUNCE that it is a non-Human call and the beginning of the "conversation", and that your answers will be recorded and analyzed.
Re: (Score:2)
There is a distinct difference between YOU calling someone, and gettting voicemail...and SOMEONE ELSE initiating a call to you, and not identifying the source of that call as a computer call, that also is listening and recording your responses.
Here...let me introduce you to my friend, "Capt. Obvious".
Re: (Score:3)
Is that the extent of your argument? You need to realise that not all actors are well intentioned and Google is unleashing a technology that will be replicated in one or two years everywhere. Then your voice and your private personal information could be used to impersonate you. You will be put in a very uncomfortable situation if people can't trust that it is actually you speaking and not a Russian bot.
Re:Ian (Score:4, Insightful)
Google is unleashing a technology that will be replicated in one or two years everywhere.
Even if Google didn't do it, it's only a matter of time before you'll see it everywhere. Whether you like it or not, this kind of technology is just going to happen.
Re: (Score:3, Insightful)
"Scientist" is nonsensically vague here. I have a PhD in Computer Science. I've had jobs where my title was "Sr. Research Scientist". I worked with an atmospheric science research group in graduate school, so I feel like I'm probably above average in terms of climate knowledge for computer scientists, but you absolutely shouldn't care what I think about climate change. I don't have the expertise to provide reliably correct information there.
What matters is that an overwhelming majority of *climate* scientis
Phone CAPTCHAs (Score:4, Interesting)
I'm looking up to see if I have anything free on that date. While I'm looking can you please confirm the prime factorization of 28573782909827352?
Re:Phone CAPTCHAs (Score:4, Insightful)
Human: Knock-knock
Duplex: Um, who's there?
Human: Banana
Duplex: Um, banana who?
Human: Green
Duplex: Ha-ha, that's great. Now about that reservation...
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Too many drones out there that already don't understand jokes.
Re: (Score:2)
Yes.... the world needs more CAPTCHAs....
Re: (Score:2)
Wow, you much be human. Everyone knows 28573782909827352 is really 2 x 2 x 2 x 3 x 197 x 4391 x 1376343499
Re: (Score:2)
Wow, you much be human. Everyone knows 28573782909827352 is really 2 x 2 x 2 x 3 x 197 x 4391 x 1376343499
Hey everybody, I found the AI running on Intel!
Headline fail (Score:2)
There's a big difference between:
Google Executive Addresses Horrifying Reaction...
and
Google Executive Addresses "Horrifying" Reaction..
Re: (Score:2)
There's a big difference between:
Google Executive Addresses Horrifying Reaction...
and
Google Executive Addresses "Horrifying" Reaction..
Still not quite there. The proper word is "horrified."
Google Executive Addresses Horrified Reaction To Uncanny AI Tech
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
"Horrifying," Zeynep Tufekci, a professor and frequent tech company critic, wrote on Twitter about Duplex. "Silicon Valley is ethically lost, rudderless and has not learned a thing."
Both.
Re: (Score:2)
The article's headline quotes a professor who said the technology is horrifying. But yes, he was horrified.
Yeah, I know. What I meant is putting "horrifying" in quotes is still misleading. The reaction was not horrifying. The person reacting was horrified.
Re: (Score:2)
Maybe not a fail (Score:2)
I read that the same way you did, then thought - what is what was horrifying to the writer of the summary, was the fact that people did not really like the new assistant? They could very well find the reaction horrifying... :-)
Of course that is not what the link was about so you are probably right, but I thought it could be an interesting twist.
Re: (Score:2)
Scott Huffman, an executive on Google's Assistant team, said the response to Duplex was mixed. Some people were blown away by the technical demos,
I suppose it would be pretty horrifying to see people blown to smithereens by the demo.
Not sure what they saw (Score:2)
Re: (Score:2)
Not necessarily, it depends on how they implement semantic processing and discourse modelling. If the "um" is interjected in a natural way based on some abstract representation of the conversation and the goals of the assistant, then that could be pretty interesting technology.
My guess is that the system will fail in very weird ways in more complex conversations, because its mostly based on machine learning of certain dialogue types. But perhaps I'm wrong and it does some real processing.
Re: (Score:3)
Re: (Score:2)
Professional “critic” (Score:4, Insightful)
Silicon Valley is ethically lost, rudderless and has not learned a thing
Wow, exaggerate much? All that may be true, mind. But using such hyperbole when voicing concerns does nothing for the guy’s credibility. He comes across as someone who has already made up his mind about SV companies a long time ago, and sees every new issue only as something that confirms his fears, as something that’s part of a bigger plot to rape the planet and enslave humanity.
Instant hang-up (Score:5, Insightful)
Their only option is to make it mimic and impersonate human voice. If it sounds very good, and there is an announcement that it is robotic, no need, but people will hang up immediately just like any other robo-call. If it sounds obviously robotic, instant hang-up. The only way it works is if they can fake it long enough to get some information, and don't let anyone know about it.
Essentially, the only value to Google is if they trick people into using it.
Re: (Score:2)
They're not faking the voice I think, instead it's a set of pre-recorded voices. If it was AI, then I'd want to see it change accent according the location and service, maybe even switch language. "Jimmy-Bob, there's a yankee on the phone!"
Re: (Score:2)
They're not faking the voice I think, instead it's a set of pre-recorded voices
Neither, it's a neural net trained on real voices.
Re: (Score:2)
The only way it works is if they can fake it long enough to get some information, and don't let anyone know about it.
Essentially, the only value to Google is if they trick people into using it.
I don't know. It might work both ways. I just need to convince the robot callers I'm human.
"Hello. This is Lenny."
Re: (Score:2)
Click!
Good work, Google! (Score:2)
This is impressive speech synthesis, though from the short demo it's hard to judge whether it's new and better than existing ones I've heard. The harder part is the domain-specific knowledge for understanding, it will be interesting to see how they deal with that.
Personally, I don't find it scary. The voice sounds dumb, but who cares. I'll probably make fun of these kinds of assistants once they become mainstream. I'm worried that similar technology will be used for robocalls by someone else in the future.
Avoiding the question is answering the question (Score:2)
Re:Avoiding the question is answering the question (Score:5, Insightful)
In this context, it's calling a business as a potential customer. Hanging up on what from all appearances is a potential customer is a good way not to have a job anymore. In fact, even if it is not a human but is calling to arrange the business of a real human customer, it's probably still a bad move to hang up.
Re: (Score:2)
If that's the case, then why are they so worried that people would just hang up if the voice was more obviously robotic sounding?
If they are reasonably expecting people to hang up if they know that the call is from a machine, then unless they have programmed the machine to deliberately lie (in which case you don't want to do business with them anyways, since they will deliberately engage in deceptive practices in order to acquire business), then why is it somehow different if you hang up on what seems to
Re: (Score:3)
Because presumably places of businesses get obviously recorded spam calls like everyone else, and for the moment no one is going to assume what sounds like a recording is going to be interactive and/or on behalf of an individual customer, and will try to keep the line open/save time by hanging up before they even hear enough words to recognize it wants to make an appointment/reservation.
All in all, impressive as it sounded, this is one area I can't understand why I'd need Google to take care of it for me.
Re: (Score:2)
Exactly.... so there's no difference. Even if it *was* a person on the other end of the phone, they are spamming you and not actually interested in any honest communication.
Re: (Score:3)
In this case, the business does not care about 'communication', they care that the call is a legitimate request to buy goods or services. Today a robocall is 0% chance, but this would represent a probable transaction.
Sure, it's creepy. Sure, I'd rather it be blatantly obvious what it is. However, they do have a legitimate concern that an obviously artificial behavior will get terminated by someone mistaking it for spam (no, even if it is by machine, a request for a reservation is not spam).
Now it can mor
Re: (Score:2)
Really give the AI a workout. As in "I have an opening 15 minutes after kickoff, is that frosty for you?"
Re: (Score:2)
Re: (Score:2)
Then they'll subtly modulate their opening words and recognize each other as bots and go modem sounds at each other for improved efficiency.
Re: (Score:3)
Like this https://www.youtube.com/watch?... [youtube.com]
Re: (Score:2)
Re: (Score:2)
salesman! (Score:2)
Most robocalls are garbage, "would you like new aluminum siding?" (I live in an apartment.)
Some robocalls are useful, "this is the town, we have declared a snow emergency, you have to take you car off the street."
Or, "this is Doctor Smith's office, you have an appointment Wendesday morning at 9."
It's obvious that these are all robocalls, and some of them are welcome or at least tolerable.
Having informative and valid caller-ID information will be helpful.
If the calls are for information that people want, the
Re: (Score:2)
This is an important point. Their use case is to call and make reservations or purchase services: basically do the stuff somebody might do online but with an analog interface.
I'll bet the companies in question won't really mind selling to somebody might not contact them via other means. Basically this lets them delay on building a web page or buying into some reservation service they don't care about/don't want and yet they can still interact with the digital world.
As long as the technology is *good enough*
Work in security, track serious threats... (Score:2)
get called a 'drama queen' and 'hyperbolic' and other ad-homonym attacks by the 'real experts' in social psychology. You know, the guys who have no technical background and aren't allowed to perform these experiments because they were deemed unethical... they keep saying the tech doesn't work.
Meanwhile the internet war is getting really insane. You guys have the tools to check (mostly) but here's some screenshots I uploaded to imgur: https://imgur.com/gallery/I3vE... [imgur.com]
On 3-2-1 (Score:2)
Bring in the telemarketers
Re: (Score:2)
But... "Don't Be Evil"...
There is no "human like" intelligence in this (Score:2)
It is a clever fake, that is all. Basically, this is Eliza with a much larger database. The databases allowing this type of "conversation" have been build during the last 30 years in slow, tedious work. Still, the potential is endless, as somewhat interactive SPAM can now reach everybody that has a phone. It seems we will eventually have to go to a whitelist system for phones or to a micropayment scheme. (Deposit me a dollar and I will accept your call, then I will decide whether to give it back. What, only
Re: (Score:2)
Re: (Score:2)
And if you believe that, then you are lacking in that unique quality many (but not all) humans have called "general intelligence".
Re: (Score:2)
Re: (Score:2)
By a tiny amount, if even that. It is the same technology, no understanding, no insight, no modelling capabilities, just reading from a giant cue-sheet.
Re: (Score:2)
Fixed this for you (Score:2)
The unspoken goal: Keep users and those who don't know they're interacting with Google asking questions and sharing information with the company -- which can be complied and sold to governments, private companies, and other persons regardless of the desire to remain private.
Say what you want about Siri, but Apple doesn't sell tha
Re: (Score:2)
The unspoken goal: Keep users and those who don't know they're interacting with Google asking questions and sharing information with the company -- which can be complied and sold to governments, private companies, and other persons regardless of the desire to remain private.
If you are someone who gives out personal information to a random telephone caller just because they sound like they are a real human, you have worse problems than Google creating a system that sounds like a real human.
It should not matter if the "voice" on the other end of the phone says "please speak or enter your credit card number now, ending with the pound key" in a robotic voice or "hey, I need to process your order, ok. What are your digits?" in a realistic human voice.
Re: (Score:2)
That's all that's needed in a lot of cases to identify people in a small business environment. Between that and the various other methods of data gathering and identification Google could potentially get the person, the employer, and their voice. And this person had no real choice in the matter. You've got a very narrow vision to not see this.
> It should not matter if the "voice" on the other end of the phone says "please speak or enter
Re:Fixed this for you (Score:4, Interesting)
> Thank you for calling Bill's Hair Salon, this is Mike. That's all that's needed in a lot of cases to identify people in a small business environment.
Oh my. Someone calls a company and finds out Mike works there. I notice that the caller didn't need to say anything to get Mike to divulge that super-secret private information, so whether Google Assistant sounds human or not is irrelevant.
And, as I said, if this "Mike" is stupid enough to hand over personal information to a random caller, whether or not the voice he hears sounds human or not, then the problem is much deeper than Google Assistant having a human-sounding voice. If this "Mike" is not authorized to handle ordering and payments and I own the shop, if "Mike" hands out my credit card info in response to any request from a cold-caller for ANY reason he's fired, period, get your ass out of my shop. If "Mike" is the one who makes the payments and he hands out the card info just because someone asks for it, then once again, he's fired. Notice the common element here. Hand out private information to unknown callers, you're fired.
This has nothing to do with the type of voice - you missed the point entirely.
Of course it has to do with the type of voice. That's the HORRIFYING part of this whole process. Google's AI can insert "ums" and "uhhs" and stuff and make itself sound like it is a real human. THAT'S the horrifying "uncanny AI tech" being talked about.
It's that Google has gone one step further and is now contacting potentially non-Google users without their consent.
Google isn't contacting anyone. It's PEOPLE who use GOOGLE ASSISTANT to make the contacts. Google isn't cold calling anyone, and even if they were, then the point I made about stupid people handing out personal information to random callers still applies.
But it isn't GOOGLE that's doing this. It's Joe Smith telling his Google Assistant to make a call ON HIS BEHALF to someone that he does not need consent to call in the first place. I don't need "consent" from a barbershop to call them up to make an appointment. It's stupid to think you do.
You DO know what Google Assistant is, don't you? I guess not. It's not a Google program to randomly call people and collect data. It's that stupid, annoying piece of crap that keeps popping up when you long-press the middle (or only) button on your Android device in more recent versions of Android, offering to help you do stuff that most people can do better themselves. Mine is turned off, so instead of Google Assistant popping up directly, I get a notification that says "Your Google Assistant is ready to help you get things done. TURN ON". THAT'S what the "horrifying uncanny tech is. This Google Assistant thing is now going to be able to make phone calls that imitate a real person's mannerisms (not a specific real person, by the way). It will now be able to send emails that look like they were written by a human. That's HORRIFYING.
Yawn. If someone asks their digital assistant to do something, then it's the person who asks for it to be done that is responsible, not GOOGLE, and it is no more horrifying to have to deal with that than with a real person.
Telemarketers already have this technology. I get a lot of crap calls that start out with a perfect human voice asking if they are speaking to me, by name. Unless you typically answer "rutabaga" instead of "yes" or "no", you don't know it's automated right away. Or maybe answer "who is calling?". Once they get the initial answer they continue in a perfect human voice. It's not new. It's not even particularly horrifying. It's annoying, but it's really no worse than a recognizable robotic voice. They don't get anything more than they would otherwise.
Re: (Score:2)
Fuck me. Are you that fucking retarded? GOOGLE is doing it, GOOGLE is the one making the goddamn call.
Do you jump up and down and spew profanity that "AT&T is CALLING ME WITHOUT MY PERMISSION!!!" whenever someone who has an AT&T cell phone calls you? There's a program on the cell phone that tells the AT&T hardware to process a call to you. Did you CONSENT to AT&T calling you? Of course not. They don't need your consent. They didn't call you. Even if your phone number is on the DNC list, when someone uses an AT&T cell phone to call you AT&T is not violating the law -- because they a
Re: (Score:2)
Thank you for calling Bill's Hair Salon, this is Mike.
That's all that's needed in a lot of cases to identify people in a small business environment
Why does Mike have an expectation of privacy when answering the business phone line and taking appointments?
And how does the fact that Duplex is calling materially differ from me calling with a tape recorder held up to the phone?
I want to use it... (Score:2)
Re: (Score:2)
I would think that if anything, this technology would make phone calls taken just as (non-)seriously as emails, because of the lower effort required to make them. If more people are robo-calling their representatives because it's easy to do so, then calls to representatives begin to carry less weight each.
The only real solution is to have a small enough population represented by each representative that they can realistically afford to care about each and every constituent.
Re: (Score:2)
Write a letter and mail it. Seriously, your representatives take snail mail VERY seriously. If a constituent took the time to write and mail a letter they will take action.
Re: (Score:2)
Even better when you write the letter in all caps.
Re: (Score:2)
Re: (Score:2)
My point was precisely that it's difficulty (having to, as you say, take the time) acting as a filter that makes some forms of contact weigh as more important than others (letters more than phone calls more than email), so if something like this Duplex makes calling your representative easier, that will make calls from constituents less important to your representatives.
Likewise if something made mailing letters to the representatives much easier, mailed letters would start to weigh less too.
Because represe
Re: (Score:2)
Because representatives realistically cannot be expected to attend to the huge numbers of people they are supposed to represent, so they have to let difficulty filter it down to the ones who care the most.
Might as well setup a pay-to-contact system. Contribute $5 to the re-election fund and your voice gets heard by your rep. It already exists for the wealthy folks, so why not the regular folks too?
Ignoring the actual problem (Score:5, Interesting)
Re: (Score:2)
Re: (Score:2)
This technology could easily be used for both parasitic purposes such as sales calls, and outright hostile purposes such as tying up phone lines with seemingly benign callers.
What an amazing planet you live on that doesn't already have telemarketers and other malefactors tying up phone lines with seemingly benign calls. This technology isn't new. Google isn't creating something new. No telemarketer is going to start using his Android phone with Google Assistant to start making his telemarketing calls, he's going to use his existing systems with his existing fat data pipe into the telephone system to keep making them.
Wrong number (Score:2)
You mean "horrified", not "horrifying" (Score:2)
I'm horrified by the lack of attention to grammar in Slashdot headlines. Some may find my reaction horrifying, but I'm fairly certain they are overreacting.
Re: (Score:2)
Well, given that /. really no longer has user submitted stories per se, one really doesn't have to look far for the source of horror that is mix of unique grammar, bias, and messaging of facts and truth that comprise the headlines and summaries these days.
Oops, won't be able to mod for another lifetime for posting this remark, either, I suppose.
Not sure I see the issue? (Score:2)
Re: (Score:2)
Re: (Score:2)
Your phone calls (Score:2)
Another data point hidden in the TFA:
"Google is taking advantage of its primary asset: data. It trained Duplex on a massive body of “anonymized phone conversations,” according to a release. Every scheduling task will have its own problems to solve when arranging a specific type of appointment, but all will be underpinned by Google’s massive volume of data from searches and recordings that will help the AI hold a conversation."
Yeah, that's your data and your phone calls they're talking abou
Said it already once today: I'd hang up (Score:2)
Re: (Score:2)
it's either a live human being making the appointment or reservation or you hang up on it. Could be a prank
If I wanted to make a prank call to set up an appointment, I could just call your business myself.
Re: (Score:2)
Yeah, and that's childish and takes up your resources. Much the same way that you could handwrite a prank letter/ad and send it. However, once you could mass send "prank" emails (chain mail and ads) things got very different very fast.
I see this as making mass phone calls (more of a) thing. Yay, a need for a better spam filter on your phone.
Re: (Score:2)
I'm pretty sure that Google isn't going to let you make mass prank calls.
Re: (Score:2)
Technology famously never expands beyond it's inventor's initial goal.
Re: (Score:2)
Because someone with the excess spending money to blow on a Duplex making appointments for them is definitely not what a restaurant or other business wants as a customer!!
That's what you worry about? (Score:2)
If you're in panic mode because of Duplex instead of Alexa and all the other "A.I. assistants", your priorities are not in the right order.
When these technologies become self-aware (Score:2)
I only hope these technologies treat us benignly when they combine and become self aware and don't destroy us like we've destroyed many species, and continue to, knowing that we're doing it.
You'll get over it (Score:2)
Press 1 if you are a human. (Score:2)
Re: (Score:2)
I'm surprised that you haven't made a phone call recently. You may have forgotten a tiny bit of info; so your prompt should come much later,as in:
For English, press 1. Para Espanol, oprima dos If you are human,
YOU: Hasta la vista, Baby!
How long until all Google Voice calls recorded? (Score:2)
They record your voice, over and over and over, mannerisms, the lot and can replicate you speaking perfectly?
Given their skills, their resources and if they had such recordings, I imagine they could emulate a person fairly easily, at least to a stranger.
Re: (Score:2)
Usually some pretty stupid shit.
https://www.twitch.tv/videos/1... [twitch.tv]
Re: (Score:2)
What I want to see is when robot call another... what do they say?
It would be a series of bleeps and bloops.
What is a recording really (Score:2)
That is an interesting point, but I have to think a transient record of a persons voice would possibly not be considered a "recording", especially if it were broken out into abstract components as soon as received.
Re: (Score:2)
It is illegal all over the EU as well. So if "somebody" calls me with a prefix "this call may be recorded...", I just hang up. Incidentally, I also hang up if it is a dialer robot, i.e. the remote person does not identify itself, but the robot waits for me to say something. Creepy.
Re: (Score:2)
Re: (Score:2)
The person being called must be recorded to be analyzed. That is illegal without consent in most states
Most states are one-party-consent states. Google knows they're recording, so they have their consent.
Also, IANAL, but I don't believe ephemeral recordings (like an Echo sending the voice to AWS for analysis, which discards it as soon as it's processed) have been litigated yet. There's also devices currently used by deaf people which do speech-to-text and text-to-speech, so there may be something already in the recording laws to cover this kind of use. (It used to be a human relaying the conversation, but
Re: (Score:2)
I would like to pit Jolly Roger against Google...
http://jollyrogertelephone.com... [jollyrogertelephone.com]
Re: (Score:2)
uR sTeP pArEnTs ArE dEaD