My vision of a virtual assistant is that you would be able to say: "Book dinner with Ross and David next week", and the only other interactions you have with the agent are "You're going to that Indian you and Ross like on Tuesday at 7" or "Can't be done, how about Monday the week after?", and possibly a message when you use your calendar "Trying to book David and Ross for dinner this evening" on some days the assistant has picked out. Meanwhile, Ross and David get e-mails that appear to be from you in address, tone and content, suggesting dates and places. If they have assistants too, maybe they don't see them; if they don't, then they can reply and the assistant understands them.
That's the sort of thing a PA actually does; they don't Google the flights for you, they know your diary, your preferences and your needs and they book the right damn flight. Your phone should be saying "Walk out of your office at 2pm and get into the cab that'll be waiting. At the airport, walk to check-in desk 301 and hand me over. I'll take it from there."
The predictions, especially with regard to leveraging data in Facebook sound closer, so I'm waiting eagerly.
"I don't think you should meet Ross any more."
"Why do you hate him so much?"
"I don't hate anyone. I don't have feelings. I'm just predicting another fistfight based on your heartbeat charts from previous meetings. I've booked you in to see the psychiatrist for Wednesday."
"I DIDN'T ask you to do that!"
"Shouting is not good for you. Findings from a study published this morning suggest..."
"Find me a PA on Craigslist. Thanks."
I'm sorry Dave, I'm afraid I cant do that
I wonder how long before Asimov's three laws become a necessity...
I'd rather be playing music.
Given the time and resources Apple has undoubtedly provided the team since then, I think we should all be expecting much greater than that demo.
I'd pay for a tool that helped me organise social events without the irritating e-mails. Take Doodle: all it does is polls for dates, and it doesn't handle diary integration, contact integration, negotiation of possibilities, or choosing the place as well as the time. Despite those shortcomings, its a successful company. That's where I see an agent actually working.
I don't recall any awesome new AI-related technologies in the last 3 years (slightly improved google translate/voice? bit better wolfram alpha?) while I can think of half a dozen new databases.
Example:
Me: "Siri, what's up today?"
Siri: "You should have lunch with Sarah at 5 p.m"
Me: "Make a reservation at that Chinese restaurant which we were in last time"
Siri: "Ok."
Source: http://www.appleinsider.com/articles/11/09/30/mockups_demons...
It'll be interesting to see what's actually announced, but I'm having a really hard time swallowing the idea that Apple's putting a Star Trek AI on a phone.
Norm: I’m not familiar with Apple’s roadmap and any delays but I can say that AI takes a lot of computing power. The Siri software needs to cache data, needs to access a big dataset at wide bandwidth and needs a big processor to crunch all of the numbers. When we originally released Siri for the iPhone 3GS, we had to perform all kinds of optimizations and shortcuts to get it to work efficiently. All I can say is that it will likely run much better on a faster phone. (Source: http://9to5mac.com/2011/10/03/co-founder-of-siri-assistant-i...)
There is a big jump in performance from 3GS to even iPhone 4, let alone iPhone 5 with possible A5 processor (dual core), 1GB RAM, iOS optimization, cloud performance, and network performance--and that's not even considering if the Sprint iPhone 5 exclusive is true which would mean the device runs on a 4G network that would make this type of interaction even more interesting if interacting with the cloud in some way.
> with possible A5 processor (dual core), 1GB RAM, iOS
> optimization, cloud performance, and network performance
with the possibility of the dual-core A5 processor, 1GB
RAM, iOS integration, cloud performance enhancements, and
network improvements
This could be why it would be available only on the newest, 4G-enabled iPhone. If this was all handled in the cloud and delivered at 4G speeds, it could be compelling.
* Will Assistant be world-changing?*
No. But it will probably be a damn good feature.
Will it address foreign languages/cultures
Probably, since Nuance can do a fairly impressive number of languages and all signs is Apple is partnering with them.
relationship questions
No. God no.
open task API
Maybe, but I would be a little surprised.
payments
Now we are just throwing buzz words together. Want to know why the "speech to text world" is "untapped"? Cause it is made up. Plus it would essentially be either a search ad or affiliate marketing situation, neither of which seem very Apple-like.
Facebook social knowledge
Again, doubtful. I mean, they've gone out of their way to integrate Twitter in iOS 5, which reeks of Facebook and Apple not coming to terms. But maybe they banged out a last minute deal.
I am sure Assistant be like most Apple features. Fairly impressive, a bit more detail oriented than the competition, but ultimately not that much better. If anything, I think it will be the new Facetime. Good demo, good for ads, some people will use it a bunch, but mostly a showy feature that will get used once and then ignored by the vast majority of users.
To me, this is exactly what it will be. Even if it does work a lot better than Android and Windows Phone 7, I can't imagine too many times where I'd be using it. I haven't used the voice features on my Android/WP7 devices very often, not because it doesn't work well enough, but because 99% of the time I use my phone I'm somewhere that I don't want to be saying commands out loud into my phone. They may solve the technical problem, but the social one will hold it back.
I can definitely see why you'd think this could go the way of Facetime, but its completely different. Why? Because while Facetime seemed to solve a problem we all wanted solved, in reality its simply not practical to video call someone with a mobile phone a lot of the time. If I'm on the go, walking a city or driving in the suburbs, Facetime is a terrible idea.
Assistant is exactly the opposite. It's almost essential in those situations and extremely practical. I can pretty much see Oprah endorsing this feature alone in her anti-texting-while-driving campaign now.
And YOU get an iPhone Assistant, and YOU get an iPhone Assistant...
What's Apple going to do beyond that, though? They can't build a keynote on "Hey, now we have this feature that our competitor has". It has to be something bigger and better. Android exposes speech input backed by Google's unfathomable processing power and recognition dataset to every application that wants to use it; what can Apple put on a phone that will be significant enough to not be a "me too"?
I'm obviously not doubting the potential, since I haven't seen the app yet, but this article is terribly low on substance.
Honestly, after watching the 45 minute presentation, there were a ton of interesting tid bits I transcribed word for word (that's what you see in all the quotes and the video is embedded if you prefer that).
For skimming purposes, I tried to bold the most important bits in the quotes, so it was straight from Siri's Co-Founder, CTO and VP of Design, Tom Gruber's mouth.
Honestly, skimming through the article quickly myself (it took me 3 hours total to write including watching the 45 minute keynote and transcribing the important bits) I can't see how you think it is low on substance, because there is a ton of it in there.
I'm honestly fascinated to see if Apple open's up the Assistant API to 3rd party developers as the Siri team clearly planned to pre-acquisition. In my mind it makes a lot of sense given the new "verb" world Facebook introduced, and I think developers would have a ball with it.
The technology came out of a government funded project on which hundreds of people worked.
It's iterations ahead of the market and so it's real AI with real use.
It will be useful to most people most of the time.
It can't understand languages other than English.
It can't answer relationship questions.
It may have an open API, in the future.
It may allow you to pay for things, in the future.
It may use social data, in the future.
I guess I'll just wait for the actual release to see why and how this is world-changing. If the product is truly world-changing, I don't think anyone can do justice to it with a blog post. Imagine if people tried to explain Google the search engine like this.
In my opinion it all stems from the trust on Apple. Apple has the reputation that they don't ship half-products and with the rumours going strong that Siri would be part of iPhone 5, everybody is simply jumping on the bandwagon and _assuming_ that Apple has solved the voice recognition problem.
As much as awesome Apple is, I still find it hard to believe that they have figured out voice-recognition.
As I quoted from 9to5mac in my post:
"In 2003, the US Government began the most ambitious Artificial Intelligence program in its history called the “Cognitive Assistant that Learns and Organizes” or CALO program. The name was inspired by the Latin word “calonis”, which means “soldier’s servant”. Funded by DARPA as part of its Personal Assistant that Learns project, the program ran for five years and brought together more than 300 researchers from 25 of the top university and commercial research institutions, with the goal of “building a new generation of cognitive assistants that can reason, learn from experience, be told what to do, explain what they are doing, reflect on their experience, and respond robustly to surprise.
The program was coordinated through SRI International in Menlo Park, CA. As the program ended in 2007, SRI took the knowledge gained by the CALO and some of its key players and formed Siri."
Not trying to sound like a fanboy, but simply saying I think Apple's recent track record has proven they're pretty good at releasing products that work flawlessly.
Heck, this could have been the simple reason Siri (now Assistant) was acquired way back in 2008 and is only now coming to light in an iPhone. Jobs may have simply held it back to meet your expectations of him.
I thought that Siri was acquired by Apple in 2010.
The ideal AI assistant would be entangled with copious amounts of machine learning algorithms. And it'd have to be plugged into my email, sms's, phone convos, facebook, twitter etc. The AI would have to evolve not just from my experiences, but the collective experiences of the ecosystem (privacy alert!).
There's way too many dangling variables here -- privacy implications, govt. regulation, software error (30%+ for state of the art natural language processing) etc.
Yet, I've gotta say I'm stupidly giddy about today's keynote.
Can anyone elaborate on this?
Well, if your boyfriend/girlfriend also has an iphone, this service might well be privy to some information that would be relevant to this question...