1. 1. Better Understanding Of Context
Data is the lifeblood of AI—the more brands have, the better they can understand their customers and grow their businesses.
In 2018, there are myriad sources for this information. It can be gleaned from shared content, human interactions, and device sensors—and used to generate more accurate views of real-time context.
It’s this data that’s fuelling the current phase of machine learning, one that has resulted in a dramatic decrease in the cost of prediction, thereby making it accessible to many brands. This new capability enables companies to fill gaps in their data, pattern-match more effectively, and forecast what customers might do next.
As an example, YouTube has embraced computer vision to help spot extremist video content, and facial recognition specialists Real Eyes and We Are Human are able to measure expressions and their corresponding emotions.
2. More Tailored Experiences
The ability to personalise experiences helps brands differentiate. We already live in a world where the experience is more important than the price or the product features.
And we know that recommendations keep customers engaged longer. An Engine report about the state of the U.K.’s top 25 e-commerce fashion brands revealed that sites offering product recommendations saw 140% more page views and 10% higher task completion rates.
In the content sharing space, Twitter uses its Magic Pony technology to make pixelated images sharper and enhance the quality of video captured on mobile phones in poor lighting. Streaming is improved by lowering data volumes, and the entire customer experience is delivered via a PWA, which itself is a predictive mobile experience.
3. Complex Task Automation
Airbnb has a track-record of innovation and has recently turned its attention to developing tools that streamline the design process. The company is building an early prototype of an AI-powered product development tool that, in real time, turns design sketches into product source code.
The recent release of “routines” for Amazon’s Alexa voice AI means that customers can build their own complex tasks. These routines mirror those already available from IFTTT and enable the connection of multiple tasks to create a genuinely smart home.
Computer vision technology, a subset of AI, is being used to detect objects and people and create bespoke advertising units. GumGum is one of the companies that can display ads over appropriate images by understanding what’s in the image, such as an ad for cat food over a photo of kittens. This allows for perspective-aware manipulation of photos.
4. Logistics Support
Coca-Cola has embraced AI to optimize vending machine placement in Japan. The “Coke On” app has had 6 million downloads and enables customers to redeem rewards while standing right in front of vending machines.
Grocers including Amazon and Ocado have been using AI to automate warehouses, manned almost entirely by robots.
Unsurprisingly, academic-led companies have the greatest potential to solve challenges of this size. Informed Actions—run by Tristan Fletcher of Cambridge and UCL—helps organisations make effective decisions and avoid costly mistakes. The company is deploying proprietary technology to improve social housing through the integration of smart meters and remote monitoring. Its goal is to enhance living conditions and the provision of social care.
5. Computer Collaboration
Few brands have embraced working intimately with machines. One that has is Stitch Fix, the online personal styling platform that converts $1 billion in sales annually and listed on the Nasdaq in November. The monthly subscription service is powered by recommendations that are generated by humans and AI working in harmony.
Customers provide preferences through style surveys, measurements, and Pinterest boards. The AI algorithms digest this unstructured information and pass recommendations to the company’s 3,000 fashion stylists. These human experts send out five items from a variety of brands, and returned items help improve the system.
Additional value is extracted from the data through creation of own-label “frankenstyles,” which already account for 1% of products. This gives a glimpse of new tech businesses that are making use of AI-based machine learning to partner with employees for more-effective solutions.
6. The Changing Workforce
The acceleration in technological change means one thing for humans—we have to adapt, which is fine, because it’s all we’ve ever done, and we’re pretty good at it. Since the dawn of time (development of the first tool, mastery of fire, cross the oceans, etc.), we’ve courted what’s dangerous, tamed it, and turned it to our advantage.
And it’ll be the same with AI, as long as we recognise that machine intelligence is coded in a different way than traditional software—and we’re very careful. The new methods used by machine learning return open-ended results, meaning mistakes are likely. As this technology rolls out, organisations will need staff with strong people skills who can manage and communicate unexpected outcomes.
The reason that AlphaZero behaves “like an alien” (neither human nor computer program) is because it’s a self-taught AI. That might sound scary, but the larger implications are that we’re welcoming a new way of thinking into our world.
No one is in any doubt that 2018 will be a huge year for AI. Twelve months from now there’s only question that will matter: Did your company enter the fray in 2018 or not?