In their own words, the Facebook Artificial Intelligence team says,
“Facebook A.I. is on a quest to build the world’s largest shoppable social media platform, where billions of items can be bought and sold in one place.”
In this week’s Tech Tuesday, I want to look at the central components of their A.I. and talk about how important this is (or is not) to the world.
Shopping on Facebook
Let’s first look at what is required to build a shoppable social media platform. This involves two major pieces of technology:
- The ability to find and isolate products in social posts; and
- The ability to recommend those products to people who would want them
To handle product recognition and isolation at scale, Facebook has developed GrokNet, a universal computer vision system designed for shopping. GrokNet allows Facebook to “identify fine-grained product attributes across billions of photos — in different categories, such as fashion, auto, and home decor”.
This technology will soon power tools that allow Instagram users to tap a(ny) product in an image to purchase it or items similar to it.
In other words, Facebook A.I. is working towards making any image shoppable.
What to make of this?
This sounds wonderful! Wouldn’t it be great to find/buy that sweater you love in some random photo you found in a corner of the internet? Or to use visual search to find something similar but cheaper?
I remember sitting in an undergraduate class and a friend musing aloud about the possibility of TV shows where a user could use her remote control to select and buy products in any show, right from the TV. How in the world would we do such a thing? Facebook is now very close. Should we embrace this new technology with open arms?
I have a few questions first:
- How accessible will this tech be to smaller companies?
- Will this capability make it much harder for smaller companies to get started?
- What will Facebook do to democratize this sort of technology in any way?
- If shoppers want to use this technology will we be forced to shop from Instagram with all of its thorns?
I’m very far from being a contrarian. I enjoy, embrace, and develop new technology. And lately, there’s been a lot of talk about antitrust. Facebook, among its peers, Google, Amazon, and Apple, are all engaged in hearings across the globe. It got me thinking about how we can fairly create disruptive technology without creating an ever-widening divide between the big guys and everyone else.
As you read what Facebook A.I. is up to, I recommend an optimistic outlook while holding them to a high ethical standard. The courts of the world certainly will.
Have a great week!
-Zank
Of Interest
Judge Throws Out 2 Antitrust Cases Against Facebook
A federal judge has thrown out two antitrust lawsuits against Facebook, brought against the company by the Federal Trade Commission and more than 40 states. The judge eviscerated one of the federal government’s core arguments, that Facebook holds a monopoly over social networking, arguing prosecutors had failed to provide enough facts to back up that claim and that the states had waited too long to bring their case. The Federal Trade Commission now has 30 days to refile it. Read more here.
A.I. That Can Learn From Others’ Experiences
Humans progress faster when we learn from the experience of others, and scientists at the Defense Advanced Research Programs Agency, or DARPA, want to translate that to lifelong learning models for artificial intelligence. Read more here.
To Win With A.I., Focus On Our Humanity
For the first time in the history of our species, technology seems to take over the very domains where we thought humans excelled: innovation, strategic planning, and coordination. So troubling is the situation that some want to rely exclusively on data and algorithms to solve our myriad problems and make day-to-day decisions. But this is wrong. Here’s why.