CAT | Business
While researching the anecdotal history of some local property, I did what I’ve done previously: ask Siri. In this case, asking about actors dates of birth and death. In the past, these type of questions would have pulled up the relevant IMDB or Wikipedia page with Siri saying “I’ve found some links for you on the web” or similar.
It took several rounds before I realized that, while the pages were still being pulled up as before, Siri was parsing out the answer to the question I’d asked, and gave that to me directly. I never had to glance down or open my phone.
Similarly, in Mail, there is now a predictive mailbox making suggestions (usually accurate) into which email box I might want to move the selected email.
In Calendar, I find addresses being suggested for my events, based on whether I’ve been there or not, address book entries, or other information.
It’s clear to me that these are all improvements related directly the Apple’s increased use of Machine Learning across it’s software products.
The Wall Street Journal (firewalled but details available here) reports that Apple is planning to join its competitors in original programming.
I would say the move was inevitable, and predicted it seven years ago in Dec 2009. More from the WSJ”
Nonetheless, the entry of the world’s most valuable company into original television and films could be a transformative moment for Hollywood and mark a significant turn in strategy for Apple as it starts to become more of a media company, rather than just a distributor of other companies’ media.
Robert Cringely has never backed away from controversial ideas and in among a rant about Apple losing its ‘mojo’ he proposes that Apple buy up all the ‘Hollywood’ writers as an end run around Studios. And that’s an idea I proposed about seven years ago in my piece What if Apple or Google simply bypassed Networks and Studios?
One of the powerful way Artificial Intelligence ‘learns’ is by using neural networks. Neural Networks are trained with a large number of examples where the result is known. The Neural Network adjusts until it gives the same result as the human ‘teacher’.
However, there’s a trap. If that source material contains biases – such as modeling Police ‘stop and frisk’ – then whatever biases are in the learning material will be contained in the subsequent AI modeling. This is the subject of an article in Nature: There is a blind spot in AI research and also the praise of Cathy O’Neil’s book Weapons of Math Destruction that not only brings up that issue, but the problem of “proxies”.
Proxies, in this context, are data sources that are used in AI programs that are not the actual data, but rather something that approximates the data: like using zip code as a proxy for income or ethnicity.
Based on O’Neil’s book, I’d say the authors of the Nature article are too late. There are already institutionalized biases in very commonly used algorithms in finance, housing, policing and criminal policy.
In this episode we discuss the future of Avid and how AI will affect post production. Only one subject has a positive looking future!
Over recent years, I’ve read a lot on Apple* but only during the flight back did I start reading anything on Google: In the Plex by Steven Levy. While I’m not yet finished it struck me the fundamental difference between Google and Apple is “who’s in control”.
With Google, engineers rule. Data rules. Everyone else is in the service of the engineers.
At Apple, designers rule. (Design in the full sense of how something operates and feels, not just how it looks).
And right there is the difference between the two companies. All else leads from that fundamental focus.
*Becoming Steve Jobs Brent Schlender & Rick Tetzeli
Design Crazy Max Chafkin
Insanely Simple Ken Segall
Inside Apple Adam Lashinsky
Steve Jobs Walter Isaacson
Maybe I’m pushing this subject a bit hard, but I really believe we are on the cusp of a wide range of human activities being taken over by smart algorithms, also known as Machine Learning. As well as the examples I’ve already mentioned, I found an article on how an “AI” saved a woman’s life, and how it’s being used to create legal documents for homeless (or about to be homeless) in the UK.
I recently commented on the importance of metadata for rights management during distribution. While cleaning my email inbox I revisited a story from late last year, on how over-the-top content providers (generally niche) can use metadata from social media and other sources to help grow their audiences.
Comments off · Posted by Philip in Apple
Apple has reportedly purchased machine learning company Turi. The expectation is that they will use it to improve other Apple products, most likely Siri. But what is machine learning and how does it fit with the algorithms I’ve been talking about lately.
Stephen Galloway writes at The Hollywood Reporter:
The majors are still the first port of call for any significant project; they still have an unparalleled ability to get that project developed, cast, shot, marketed and into theaters; and despite extraordinary technological and economic change, they haven’t allowed any upstarts to challenge their hegemony.
He then goes on to document the changes that Amazon and Netflix have wrought on the “old bastions of Hollywood Power.”
All of which is good news: more outlets lead to more production.