Artificial Intelligence made incredible advancement this year. From the first autonomous Cars to the AlphaGo victories — we’ve observed AI move into the spotlight. So, we’ve decided to decode some buzzwords, examine some big stats, and surface the trends that you deserve to know about.
“85% of customer interactions are projected to be managed without a human by 2020.” Gartner
“$1.5B Amount venture capital firms invested in AI as of May 2016.” CB Insights
Between Hollywood and your dusty stack of sci-fi novels, you are given many bizarre descriptions of the AI of the future. But AI is already here. It’s all around us — just in smaller forms.
Intelligent behaviour in an autonomous agent — This is AI. It is representing the brain, not the body, of intelligent machines (AI ≠ robots). The AI of today can do precise duties (driving a car, scheduling meetings, picking your next Netflix/Amazon show). AI research is leading toward something more advanced: artificial general intelligence, or AGI. This AI — when a machine can do things in a way that is indistinguishable from human behaviour — is what we’re all waiting for.
The Construction of AI
AGI « Artificial general intelligence », is the dedication at the end of a Construction. But before we get there, each construct must be built with high expertise. At Job Pal, we believe the cornerstone tools of AI include: Machine Learning, Deep Learning, Natural Language Understanding, Context Awareness, and Data Privacy.
Machine learning and AI aren't the same. Machine learning is a part of the 'AI Construction'. So what is Machine Learning — or ML? It is the ability for an algorithm to learn from previous data to produce behaviour. Machine Learning is teaching machines to make decisions in situations they have never recognised.
The most mainstream approach to ML is giving the algorithm a data set of situations and telling it what the right decision is — training a model, which supervises Machine Learning. Once the model is trained, we can inject new, more foreign data through the algorithm — and eventually, the machine offers intelligent decisions in these new, unfamiliar situations.
Deep learning is a branch of machine learning where artificial neural networks — algorithms inspired by the way neurones work in the brain — find patterns in raw data by combining multiple layers of artificial neurones. As the layers grow, so does the neural network’s ability to learn increasingly abstract concepts.
For example, neural networks can learn how to recognise human faces. How? The first layer of neurones takes pixels from sample images; the next layers learn the concept of how pixels form an edge, then that layer passes that knowledge to other layers, combining that knowledge of edges to learn the concept of a face. This process of layering knowledge continues until... — the neural network algorithms recognise specific features, and ultimately distinct faces.
Deep learning was the core technology that Google’s DeepMind used in their Alpha Go machine. Alpha Go beat a world champion, Lee Sedol, at the extraordinarily complex game, Go. What’s so exceptional about Go? The number of possible positions on the board is a number greater than the number of atoms in the universe.
Natural Language Processing
AI must interact with humans as well as humans communicate with each other. In AI, this is called Natural Language Understanding, or NLU. NLU is a huge priority and hurdle in AI research. Why? Because human communication is not simple. It’s a complex web — random, out-of-order, peppered with humour, emotion and conflict — and it depends hugely on context. Once AI succeeds the challenge of human communication, decoding complicated questions (natural language queries), making links, and giving answers that make sense, radical advancement is not far behind.
Like a human assistant, an AI companion can only be as intelligent as the data — the context — you give it access to. If your companion — human or artificial — only can see your calendar and reservations, but not your contact list and location data, it is not a very helpful assistant.
Context is king when it comes to complex tasks. It’s true of people, and it’s true of AI. Every section of data and context must be tuned correctly to play a different part in the construction of AI.
Some more Stats
“1,196: Number of AI startups listed on CrunchBase, an online listing of all VC funded startups in the world.” Crunchbase
“6% Of adults that say they are “very confident” that government agencies can keep their records private and Secure” Pew
“6x amount of increase in equity deals to startups in artificial intelligence from 2011-2015 — this includes companies simply *applying* AI: healthcare, advertising, and finance, as well as general AI tech.” CB Insights
“6 billion Number of connected devices that will be making service requests by 2018. Think: Your toaster calling a support number” Gartner
“16% of American jobs AI will replace by the end of the decade.” Forrester
“2 million: Number of employees that will be required to wear health and fitness tracking devices as a condition of employment (for their safety!).” Gartner
Advance of Voice Commands in Assistants
While the crusade for natural language communication continues, speech-to-text technology has improved immensely. A new-and-improved Siri and the launch of Amazon Echo and Google Home are prime examples of this science fiction storyline coming true.
AI is not the rise of the Machines; it’s the Machinification of Humans
While AI movies and TV feature robots with human bodies, many fail to explore AI (tiny little AI robots) IN human bodies. AI visionaries like Elon Musk are starting to talk AI-human symbioses, with AI nanotechnology effectively curing humans of … death! Sounds like a good storyline.
Content as a Testing Ground
To get smarter, AI requires lots of data, patterns and new conditions and enter content platforms. Users’ consumption patterns are already shaped by the machine learning behind Spotify’s “Discover Weekly”, Netflix’s “Recommended For You” and Facebook’s ability to keep you in a filter bubble of your making.
AI is learning to be less biased
When a group of scientists recognised that their AI was replicating the human bias (think: “man: businessman:: woman: homemaker”) — they broke down the origins of bias and fixed it.
Technology is both inescapable…and disappearing
The tomorrow of AI is reliant on data privacy because without data to learn from, AI cannot get smarter, and progress will decrease. Companies must act to creating private and secure software. Users need to know that their data will be preserved if they’re ever going to permit an AI full access.
Why sharing this?
Job Pal has put hard work and research together to build an AI-powered assistant SDK that enables any mobile app or connected device to integrate a chat assistant into their product. We want to create AI that is so good; it can eventually make technology disappear. But that’s not all.
Living in a world of AI like we do in Job Pal, we hear the same questions over and over. You deserve answers. So here it is — a quick guide to empower anyone to understand the basics of AI better because after all, it’s all around us. Sharing this knowledge matters.