68 comments

[ 5.5 ms ] story [ 25.3 ms ] thread
> I.am+’s first enterprise customer is Deutsche Telekom AG (DTEGn.DE), the German telecommunications giant and parent company of T-Mobile. Since July, the company has been using Omega to power an AI customer support chatbot and it plans to add a voice phone system soon, i.am+ said.

Wow. I had no idea ChatBots needed $117 MILLION in funding to ship.

A quick search of GitHub shows several 100s of chatbots, even 1 with just 200 lines of code. https://www.google.com/search?q=site%3Agithub.com+chatbot

This has Juicero written all over it.

> Wow. I had no idea ChatBots needed $117 MILLION in funding to ship.

You don't. But the startup game is often not so much about shipping, but more about name recognition and connections.

Agreed. Still, $117 MILLION seems really excessive. How many people do you need to build, ship, market and support an AI Chatbot?
This is after $89mm if I am reading it right
I have no clue what I.AM.PLUS is; this is a general comment:

Sure, creating a dead simple chatbot is easy! But you can't compare a bare-minimum open-source dumb chatbot with a enterprise AI company that happens to take the form of a chatbot.

Enterprise is a funny word here. It just seems like any other voice assistant except it connects to your inventory control system rather than your Nest at home. It seems like mostly marketing to me, I doubt the technology is particularly unique.
That's pretty much just what Enterprise is, though... a consumer product, sold at a much higher price, and including things companies need like support and customization.
Unique technology is absolutely not a requirement to build and ship useful and impactful product.
> A quick search of GitHub shows several 100s of chatbots, even 1 with just 200 lines of code.

Also why does Dropbox need funding? I can write a 2-line shell script that calls rsync.

A little tongue in cheek, but maybe not:

7M chatbot development

14M enterprise marketing materials

28M accounting and compliance response to enterprise requirements

56M enterprise sales team & expenses

12M reserve & voice mail development

They'll need another round in less than a year.

You sound like a solo programmer not considering any aspects on how to run a company that serves a global client base.
I guess that's what you get for being buddy-buddy with Benioff...

There's a joke in there somewhere about how most telco customer service is indistinguishable from AI chatbots.

Wow, I'm surprised that the voice quality is so atrocious compared to Siri, Alexa, Cortana, or Google Home. The latter sound welcoming, warm and natural, Omega sounds like an old TTS system from the late 90s.
Omega literally sounds like an unmodified MaryTTS (http://mary.dfki.de/) that you can get up running with a few api queries

I am continuously amused how people turn these things into 100 million dollar businesses.

Anyone have any experience with it? How accurate is the video?

For example, second use case - in the car, she queries the software the amount of some_specific_product in some_specific_warehouse, and the tool answers near instantly. Is that a currently functional scenario? Is this because the client has an API set up with `counts` already calculated for these queries, or is a part of the tool intelligently crafting these queries to answer these kinds of voice questions?

From the article, the launch product is simply a customer service chatbot (currently text, with voice coming soon)

The omega product video is supposed to be the future vision and not a real product.

No insight into Omega, but natural language interfaces to spreadsheets/databases exist.

Google Sheets has an implementation if you click the explore button in the bottom right hand corner.

I made a dummy spreadsheet with columns name & warehouse, then typed in "how many units of rose gold are there in la" and it got the query right. It even got "how many rose gold units are in los angeles" correct despite the spreadsheet never saying los angeles. But it's pretty easy to trip it up; if you make an item called "rose gold" and "rose gold ring" and ask it for rose gold, it assumes you want the exact match. If you ask it for items in la or sf, it will only give you a count for the first condition. If you ask it about California, it will get confused.

Still pretty impressive for a two column spreadsheet though IMO.

Assuming the system is designed to intelligently integrate with SAP and other ERP systems then it shouldn't be too hard. The trick (and what people might actually pay for) is making this integration work.
Unbelievable. Probably will be 10 times worse than Siri.
I'll gladly take half of the $117 million to license my, all-new, 2018, deep penetration, convoluted, networked neural network, AI chatbot (for the ENTERPRISE) "Alpha".

> Alpha, create a Mars civilization.

>> Okay, I've scheduled terraforming for 2 o'clock on Wednesday. Should I confirm this time?

> Yes.

>> Okay, terraforming Mars at 2 o'clock on Wednesday.

"Alpha™" is a registered trademark and isn't at all vaporware. Calling "Alpha™" vaporware is punishable by death in all 50 states.

Here's a video of the product if you're as curious as I was. Siri/Alexa for enterprise users -- https://iamplus.com/enterprise/

Seems like a smart market entry point. Enterprise employee virtual assistants are probably a wide open market today because Apple, Google, Amazon, Microsoft, and Samsung focus on consumer AIs. Easy to envision a startup getting solid, near-term customer traction with a keyword-driven voice product.

Genesys has been doing this for literally decades
Not going to knock it, but I don't really get it. How different can an enterprise version of this be? Probably a lot more to the plan here than this video.

Out of all the different voice things out there (Siri, Alexa, Google, Cortana, whatever), I've only ever done 3 commands really:

> Hey Google! What's the weather?

> Hey Google! Play ____ music on Spotify.

> Hey Google! What sound does a cow make?

Does anyone have stats on how the general population are using these things? So far, just for me personally, I can't really find a way to be more productive with any of this tech.

Conspiracy nut in me first reaction is, "Ok, a new way to spy on companies".

I use Siri to add things to OmniFocus multiple times a day, "Hey Siri, remind me to ...", a lot of people seem to use it for a lot of things[1]: https://www.reddit.com/r/apple/comments/549ytw/how_many_of_y...

[1] Well, a lot of people who use reddit and are interested enough in apple to be on an apple subreddit.

My most common pattern is "Siri, remind me to ____ when I get home". The geolocation-"awareness" really hits my soft spot.
These are Enterprise functions I would pay for, but are not currently an option.

> Hey Google! Clock out.

> Hey Google! Clock in.

> Hey Google! Is anyone on my team out sick today?

> Hey Google! Do I have any outstanding invoices?

> Hey Google! Is anyone on my team behind schedule on accomplishing X task?

>How different can an enterprise version of this be?

More control over the wake word and name space, such that you don't have to say "hey Google tell some-long-app-name to blah blah blah".

(comment deleted)
Useful for home automation. Useful in circumstances where you are working with both hands doing something else. I've been using voice commands daily now. Nobs and buttons are going to start disappearing and then you won't have a choice.

The bigger question to me is how a startup is going to compete with Amazon & Google's voice APIs/cloud services in the mid term much less the long run. Unless it is a play for acquihire.

Timer 15 minutes. I use that all the time. I would use it (siri) a LOT more if it didn't suck so much.
billy, you have to stop, you are wasting money.
there is something unexpectedly deeply depressing about knowing will.I.am is a top executive in your field
If you want depression I can give you a better one: Kim Kardashian's game made more than 200 million on revenue.
I'm surprised to see so much trivializing of the product and flaming of the entrepreneur.

Investors don't just give away $117M "because he's famous", and I seriously doubt their enterprise customers care that he's a singer. More likely he and the team have been making steady progress for years and are delivering real value. Kudos to them.

It's one thing when you read a modest "Show HN" where some guy like me (or you) builds something and we're sharing it with the world, hoping it brings us fame and fortune; it's something totally different when a very wealthy individual (will.i.am) raises a ridiculous amount of money ($117M) and builds a completely trivial product.

And for what it's worth, even Beats was completely trivial. I know 2 engineers and 1 product person that worked there and they all agree that their acquisition by Apple was solely based on the brand and the names behind the brand, not the product. (In audiophile circles, it was well known that Beats by Dr. Dre were awful bang for your buck.)

Do you genuinely believe that building a brand with that kind of power is "trivial"?
The brand is not the product.
Emotion shapes perception.

How a customer feels about a brand will have a very real impact on their experience with a product.

You should work on your branding.
Becoming a multi-platinum-selling musician/rapper is sure as hell not trivial, but I don't think that's what you're asking.
Was asking about beats more than Will.i.am.
Ah, I was referring Dr. Dre -- Beats' eponymous primary investor (but I think it probably applies to will.i.am too).
> Investors don't just give away $117M "because he's famous"

Investors gave Theranos several >$600M in funding. Today we know about the lack of due diligence. It was all about investing with that young female wunderkind because everyone else did. So that's no argument.

Regarding the founder: Looking at that cringe worthy Apple TV shark tank copy and how he forced one founder to list him as co-Founder I am pretty sure Will.I.Am just sits around talking BS and others doing the work for him.

Last but no least: Yes, $117M is actually a pretty damn lot for a company that does chatbots. This type of NLP isn't even resource heavy (like shitloads of GPUs heavy) so there is no need to put so much money into that. That said, of course the Germans are clients. They are always late to the party and most of the time get ripped off.

female

I used to feel similarly -- that being a girl sometimes gives you special privileges -- but then I concluded it's probably not true, and voicing those feelings is an old tradition, best left in the past. How would you feel if you got a job somewhere and all your coworkers suspected you didn't deserve it? Same thing.

It gets tricky when there's a grain of truth to the feeling, since it's confusing trying to figure out whether you're censoring your own thoughts to the exclusion of reality just to conform to social norms, or whether you're actually being regressive. But FWIW all evidence thus far in my own life seems to point to "being a girl may have mattered once, but if anything it tends to work against them rather than in their favor." E.g. you have to worry about shit like an investor coming onto you: if you say no, what will happen? Or if I dress a certain way, will it affect how I'm treated at work? We take it for granted that we don't even have to think about any of that, ever, but those are real concerns.

This is a sensitive topic, so we should try to keep the conversation neutral and substantive. But I did want to open the door for talking about some of this.

(Traditionally, the two options are to ignore the "female" putdown, or ask you to stop. Any time anyone even tries to ask "Why do you feel that way?" the conversation immediately devolves into a catastrophe. I'm trying to see if there's a productive third option, which is to frame the conversation in a certain way where people are inclined to learn and relate experiences, rather than to be combative or dig in their heels. I'm going out on a limb here, so please don't make me regret doing this.)

The truth is probably closer to "she was a rich kid with connections that we could only dream of."

This has been on my mind, so I may as well talk: When we see things like female-only conferences or events, it's really tempting to feel jealous, and to feel like girls do get special treatment. Or even if it's not really special treatment, it's still "special" because no one goes out of their way to try and recruit you to some event just because you happen to be a dude.

I think this is a symptom of feeling isolated in general. Church attendance is declining, and -- setting aside the question of religion -- church was the primary social activity for a long time. It was a weekly thing, and you'd feel special and included. When you're not religious, you lack this facet of your life, and it manifests itself even though we'd really like to believe otherwise.

You can go out and attend React meetups or whatever, but it's different than (a) you're expected to attend, because (b) you're [a girl, religious, a harvard grad, etc]. When you have nothing like that in your own life, it's extremely tempting to feel like the world is against you and you've had to work for every single thing you've ever had, so this special treatment is unfair.

But here's a concrete example. A female webdev friend of mine works at a startup, and they made it to YC's interview stage last week. She was employee #1. The moment they received that email, the founder started treating her differently. She and I were both mystified about why. Why would the prospect of transitioning to YC's pipeline possibly matter? She was an effective employee, and built their whole stack. Her work was identical to any male.

Then we had a very uncomfortable realization: one of the founders joked that if they got into YC, they'd all have to bunk together like it was college. You tend to live in the same apartment at first.

She asked me "do you think if I was a boy, I'd be invited to the boys club?" and it was heartbreaking to realize that it was very possible that this employee #1 may have had to worry about not being invited along to California solely because she was a girl. We stil...

I think you're reading too much into the word "female" -- although I do agree that its use in this context is kind of weird (why not say girl, or woman?).

I think you're reading to much into it because it's pretty well-known that in startup circles being a woman is, generally speaking, a net negative. So an argument that she got funded because she's a woman goes against matter-of-fact realities in Silicon Valley.

Eh, probably. But on the other hand, there's no point in mentioning that a girl is a girl. It'd be conspicuously out of place to say "young male wunderkind."

I agree that it's pointless to go on like this. I didn't mean to put anyone on the spot or call anyone out. It's actually just been on my mind lately, and I've been looking for an excuse to talk about some of this stuff openly. It seems like an important conversation to have, but there aren't many opportunities or contexts to have it without attracting trolls.

> But on the other hand, there's no point in mentioning that a girl is a girl.

Agreed, that's why I think it was weird to bring it up. Plus, we all know who Elizabeth Holmes is; we all know she's a girl -- why mention it? I guess it was kind of salient that she was a woman.

No idea.

You touched on an interesting point:

it's pretty well-known that in startup circles being a woman is, generally speaking, a net negative.

The reason I spoke up originally is because where I'm located, it's very common to believe the exact opposite. It's become something of a trope: "Better respect women or the thought police will get you!" (said in a dismissive tone)

I don't know if I'm just living in a weird area, but that seems to be the default sentiment across the last N companies I've worked for. It's either treated as a joke or not discussed.

And then of course there are pockets of people with the exact opposite feelings, but they often go so far in the other direction that it's hard to take them seriously. There's some space between being completely dismissive and hyperventilating about women being oppressed, and I thought the latter was a myth until I saw it in practice.

So it's just really hard to figure out, y'know? It's not something that we generally get to talk about. It's socially weird to bring it up with strangers or publicly, so you're very likely to believe whatever the people around you happen to believe. So how do you determine whether your beliefs reflect reality?

> The reason I spoke up originally is because where I'm located, it's very common to believe the exact opposite. It's become something of a trope: "Better respect women or the thought police will get you!" (said in a dismissive tone)

I think this is a reactionary counter-attack by the hardline anti-SJW crowd. Don't get me wrong, I'm not in complete agreement with the ultra-(socially-)liberal SJW gamergate-style movements; however, if we're not lying to ourselves, it's fairly obvious being a woman in tech is not great. So you have nutty guys like Google's Damore who actually believe exactly what you're saying: that men are somehow being discriminated against because the powers that be are trying to slightly tip the scales in favor of women. Which is completely crazy to begin with, given that if you're an engineer at a tech company, you're already making like ~$100k+ which puts you in the top ~10% of US earners.

> So it's just really hard to figure out, y'know? It's not something that we generally get to talk about. It's socially weird to bring it up with strangers or publicly, so you're very likely to believe whatever the people around you happen to believe. So how do you determine whether your beliefs reflect reality?

This is a hard question and don't really have any answers. I studied philosophy in college and we talked about all kinds of controversial issues (abortion, affirmative action, gay rights, etc.) My theory is that most engineers just aren't mentally equipped with the tools required to make sound social judgments, e.g. being well-read in ethics or social theory, so debates or a friendly clash of ideas often end up being fruitless and awkward.

This is a hard question and don't really have any answers. I studied philosophy in college and we talked about all kinds of controversial issues (abortion, affirmative action, gay rights, etc.) My theory is that most engineers just aren't mentally equipped with the tools required to make sound social judgments, e.g. being well-read in ethics or social theory, so debates or a friendly clash of ideas often end up being fruitless and awkward.

This is an interesting observation. A lot of flamewars could be as simple as that.

If one were inclined to equip themselves, what would you do? It's very rare to have the opportunity to attend a philosophy class, and a lot of philosophy students tend to take it much too seriously. It's serious topics, but almost everyone who studies philosophy ends up feeling elevated over those who don't. This is remarkably ineffective when trying to persuade people to your way of thinking.

So I'm interested in what to study for my own purposes, and then try to integrate those concepts into my day to day life. What would you recommend?

I went into my Philosophy major wanting to study mostly logic and math. But I actually fell in love with ethics, a topic I had never been interested in before. I think most of these social issues find their roots in ethics, so if you're interested in that this would be my suggestion: start off with Plato's Republic[1], move on to Aristotle's Nicomachean Ethics[2] and then Christian philosophers (Aquinas[3], for example) and then (my favorite) John Stewart Mill. I often suggest his On Liberty[4] on HN. It's one of the absolute best primers to modern social ethics around. I don't think Kant is particularly useful (or interesting), but many would disagree, so your mileage may vary.

As far as the "modern" era, I suggest reading G.E. Moore[4], Elizabeth Anscombe[5], Phillipa Foot[6], etc.) Anscombe's Modern Moral Philosophy[7] is one of my favorite essays ever. Apart from that, it's important to read Supreme Court cases and decisions (especially important ones: Bakke v. UC Regents, Chaplinsky v. New Hampshire, etc, etc.) But most importantly (and what we were vehemently taught), is to always practice the Principle of Charity[8].

Some of these books and essays are very hard to parse, but there is a lot of material online that will help figure out their meaning.

[1] http://classics.mit.edu/Plato/republic.html

[2] http://classics.mit.edu/Aristotle/nicomachaen.html

[3] http://www.iep.utm.edu/aq-moral/

[4] https://www.gutenberg.org/files/34901/34901-h/34901-h.htm

[5] https://plato.stanford.edu/entries/moore/

[6] https://plato.stanford.edu/entries/anscombe/

[7] https://en.wikipedia.org/wiki/Philippa_Foot

[8] http://www.philosophy.uncc.edu/mleldrid/cmt/mmp.html

[9] https://en.wikipedia.org/wiki/Principle_of_charity

With respect, and not to belabor a silly topic, but 'female' is an adjective, and is the only correct choice between those three options in the original sentence. Although I would tend to agree that as collective nouns are concerned, 'women/men' is almost always correct when describing adult humans when one wants to specify a particular sex or gender. (Girls/boys/kids rustles my jimmies when used to describe adults).

'female wunderkind,' in an irrational, razzle-dazzle bullshit train like theranos was probably emphasized to create an aura of exoticism and magic (e.g. 'wonder woman'), not because of the sociocultural situation of women in tech broadly speaking. The symbolic significance of gender changes depending on where/how one invokes it.

'female wunderkind,' in an irrational, razzle-dazzle bullshit train like theranos was probably emphasized to create an aura of exoticism and magic (e.g. 'wonder woman'), not because of the sociocultural situation of women in tech broadly speaking.

I didn't even consider this. You're probably right.

Well, in that context, it was obviously absurd for me to start talking about any of this. Sorry about that. But HN is optimized for good conversation, so hopefully it will turn out interesting (if unexpected).

Looks like this thread got knocked off the front page anyway, so it doesn't really matter now. Drat; I was hoping to get people's thoughts on this.

There's no reason both can't be true. In the case of Theranos in particular, the fact that she was a female founder in a hypertechnical field gave her some level of cachet. It may have worked against her in other ways, and it definitely works against many women all over the place.

In order to have these discussions, we just have to be able to hold nuance in our minds. Yes, Elizabeth Holmes in particular due to accidents of fate may have benefited unfairly from being a woman at a time when VC companies wanted to be seen as progressive and friendly to women and minorities. However, that doesn't have much at all to say about the struggle of women generally in tech, or in society at large.

I'll be more direct than sillisaurus: You seem to be operating entirely within a world of stereotypes, some of which don't even make sense. Like, for example "the Germans getting ripped off": Even ignoring how unlikely it is to correctly infer one datapoint from data aggregated across something as heterogenous as "the Germans", they (or, more specifically, "Deutsche Telekom") are customers of this company (not investors), buying something rather non-essential (a customer support chatbot), making it seem rather unlikely that they are getting "ripped off".
I think you overestimate investors and how they think. Case in point: Juicero raised boat-loads of money because investors got hooked on their business model of selling over-priced juice packs to consumers. Too bad investors only learned too late that these packs could be squeezed by hand just as well as with a $400 machine.
He’s been making stuff successfully for years, well after music made him and the BEP famous.
> will.i.am has raised $117 million [...] as it announced its entry into the corporate computing market with a voice assistant for customer service.

Voice assisted AI startup from a pop star receives $100mm financing? Guys if this isn’t bubbly then I don’t know what is!

If you’re an IC, I hope you can take a little off the table while the gettin’ is good.

Celebrity worship is harmless when it consumes leisure time and disposable income. This is not going to end well.