SoylentNews is people
SoylentNews
from the I-can-help-if-you-give-me-your-PIN dept.
Predicting an eventual upturn in the sagging smartphone market, research director Ranjit Atwal told The Reg that while artificial intelligence has proven key to making phones more useful by removing friction from transactions, AI required more permissive use of data to deliver. An example he cited was Uber "knowing" from your calendar that you needed a lift from the airport.
"Today there an no good use cases for AI - it's just an enhancement of what we do on a phone. We're thinking ahead a few years, when AI can start to remove friction between us and the phone." This can be done by automating mundane tasks - such as ordering an Uber - but that will require users to share data with services they trust.
Another example Atwal cited was renewing house and car insurance. "If you haven't changed your car insurance there should be easier and more effective ways of doing that. But that only happens if you share your data."
That seems a tall order today. Since news broke that Cambridge Analytica used of [sic] Facebook data it should not have been able to access, Facebook has been on the end of the backlash for its permissive data sharing. And not just Facebook. Gay hookup service Grindr was found to be sharing medical information - including their HIV status - with third parties.
[...] "By 2020, AI capabilities on smartphones will offer a more intelligent digital persona on the device. Machine learning, biometrics and user behaviour will improve the ease of use, self-service and frictionless authentications. This will allow smartphones to be more trusted than other credentials, such as credit cards, passports, IDs or keys," Atwal concludes.
Putting the pieces together, then: if AI is to transform efficiency, and this transformation requires plenty of consumer data, and the data is valuable, then there are some interesting sums to be done. How much is your calendar worth? Will it be profitable for the likes of Uber to pay you for that data in order to get your business?
(Score: 0) by Anonymous Coward on Monday April 09 2018, @09:36AM
You don't need AI systems nor large data mining services to do those features. Each feature can be individually developed directly. It's not difficult to parse emails and the ical files airlines send you and then estimate travel and wait time to the airport from your configured home address (or most common GPS location or provide the user with a list of a few places to pick from). To do this through AI you'll have to data mine enough to find the pattern of people getting airline emails and then seeing an taxi charge on their credit card. But even that isn't good enough because it doesn't make the connection on how you used the taxi unless the charge includes all the taxi metadata on where you were picked up and where you went (which is what this guy is asking for).
But we don't want these middlemen apps anyway. They'd offer you Uber over a taxi, public transport, driving yourself, or having a friend drive you. These types of things offer based on whose willing to pay the most or scratch each other's back, they don't offer services based on quality. Look at the real estate industry, home inspectors are giving their customers contact info to advertisers for minor gift cards. You end up getting ads for home maintenance services. These guys aren't a 'real business'. When you call to make a claim they sell your claim to whomever pays the most for it or is a preferred subscriber. You pay for the subscription, the service companies pay for the subscription, and your repair job suddenly costs an extra $120 just to connect the two of you together. These are leeches on society. Basically an eBay for service jobs were you can't control who gets the winning bid nor how much they'll charge you.