The TechCrunch conference is an event that swoops in on the startup scenes of the US and Europe, gathering for appraisal the technology world’s rising movers, shakers and innovators.
It’s widely regarded as the place to get a feel for what’s happening at the industry’s grassroots – which is exactly what PwC Innovation Manager Marina Paronetto headed to London to do. As it turns out, it was the brains behind the machine – artificial intelligence – that was the star of 2016’s show.
It was said in sarcasm, but MC Jordan Crook couldn’t have been more accurate when she said that TechCrunch’s theme for its latest Disrupt conference was artificial intelligence. From what I saw, every solution on show here featured an element of AI.
Held in London’s Olympic Park, it’s not just at this technology and startup conference that artificial intelligence is going for gold. If the evangelising is to be believed, AI is poised to become the new normal for us all. “We truly believe that in five years’ time, every single website or app will have a layer of personalisation and machine learning built into it,” said Adam Spector, co-founder of LiftIgniter, a content personalisation platform for websites and apps.
In fact Mustafa Suleyman, co-founder of DeepMind, the AI research arm acquired by Google in 2014, raised the game further during the conference opener by stating that his goal is to “solve intelligence and make the world a better place.”
“Many of our most complex social problems are becoming increasingly stuck. We are struggling to make progress on climate change at the rate that we need to; we have enough food on the planet to feed everybody equally, but we are not distributing it in the right way.”
It’s the ability to to make sense of an overwhelming volume of data – both existing and yet-to-be-created, that’s powering the phenomenal rise of AI. What once may have presented an impossible problem is being brought into the realm of the solvable.
Calling itself ‘a Google Maps for science’, Iris AI, for example, is a startup tackling scientific research. Its AI-powered assistant can analyse research papers and visually map out the concepts. “Every day, more than 3,000 papers are published in science, technology and medicine,” explained Iris co-founder and CEO Anita Schjøll Brede. “Today, Iris is a context exploration tool. In three years from now, she will be a proactive science assistant and, in a decade from now, she will be a scientist herself, reading a body of content and coming up with a hypothesis, testing it in a robotic lab and publishing the results.”
Another example of AI being brought to life is Aiden. It’s a solution intended to act like a marketing co-worker, querying data from multiple dashboards and compiling it into charts, making the information easier to find and digest. Humanised in more than just name, marketers can message Aidan using text message, email, or even communications tools such as Slack and Skype.
Though the goal of AI is to develop an artificial solution, Schjøll Brede emphasised designing for humans. She touched on the importance of establishing a good user experience from day one, creating a tool that helps users navigate and enjoy their journey through complexity. If usability is absent, a tool simply won’t be accessed.
An example of this imperative in practice is Century, one of the startups chasing funding in the conference’s famous ‘Battlefield’. Century is an AI-powered platform that creates tailored learning experiences for school and college students. Rolling out such a project at scale would of course require designing for student needs and preferences – if they fail to engage, there is a huge education opportunity lost.
How can you and your organisation
step into this exciting future?
Start with fast and slow
If you want to understand AI you need to first understand theories of thinking and how knowledge is processed, before you can know what to do with it.
Dual process theory is a good place to begin your AI journey. Monica Anderson, CTO and co-founder of Sensai, is the originator of a theory of learning called artificial intuition. It challenges the view that thought can be based on logic alone, and that artificial ‘intuition’ could yield better results in areas such as speech recognition and natural language processing.
In this talk below at Stanford University, she discusses concepts such as Dual Process Theory, The Frame Problem, and some consequences of these for AI research.
We’ve learnt that experimentation, prototyping and casual play time is crucial for real immersion in a new concept. It allows the birth of new possibilities. To understand artificial intelligence and its potential for solving humanity’s great problems, you should roll up your sleeves and experiment.
Stop killing your data
We collect data and make sense of it. That concept is in the past. Every piece of information that’s disconnected from the world’s current flow of knowledge is dead. Which means that every spreadsheet you own is a corpse on your computer. AI shows us that data has value when it’s alive, happening right now, helping shape the present and the future. Stopping the world to make decisions is not a luxury we can afford any more.
Start small (and start now)
A couple of months ago, our colleagues reported back from Dreamforce in San Francisco. There too, AI was dominating conference discussions – even though it is early days for the technology. Artificial intelligence must be built from the ground up: it needs data to be able to learn and thrive. So right now, laying that data foundation is the focus.
Follow the sequence of events, however, and AI could eventually help form the core of your organisation. In ten years’ time, it may facilitate communication with your customers and employees, make financial decisions, protect against cyber risks, most likely help you shape your intellectual property, as well as assist you as a human to navigate the marvellous complex world around you. Think that’s going help solve some problems? Time to get on it.