The AI Awakening and the Challenge for Society

Transcription

1 The AI Awakening and the Challenge for Society MIT, November 28, 2017 Erik Brynjolfsson

2 The Second Machine Age Changing the world requires two things: Power system: move or transform things Control system: decide where and how Industrial Revolution = Physical Power Steam engine (and Internal combustion engine, Electricity) Mostly a complement to humans Second Machine Age = Mental Power Computers, Software, Big Data, Machine Intelligence Complement or substitute? 2

3 The First Wave of the Second Machine Age Teaching machines what we know Coding = codifying knowledge Examples: Arithmetic and simple logic Well understood algorithms and processes Repetitive data processing

4 The First Wave of the Second Machine Age Teaching machines what we know Coding = codifying knowledge Examples: Arithmetic and simple logic Well understood algorithms and processes Repetitive data processing Mantra: Codify, Digitize, Replicate! Economics: High fixed costs, low marginal costs Big wins: routine information processing Insurance claims processing Tax preparation software ERP, CRM and SCM systems Mobile Apps

5 The Limits of the First Wave: Polanyi s Paradox We can know more than we can tell - Michael Polanyi

6 Take an example. We know a person's face, and can recognize it among a thousand, indeed among a million. Yet we usually cannot tell how we recognize a face we know. So most of this knowledge cannot be put into words. - Polanyi, The Tacit Dimension, 1964

7

8 The Second Wave: Machines that learn Artificial Intelligence A set of techniques used to try to imitate human intelligence Machine Learning Using large amounts of data, machines learn without being explicitly programmed Deep Learning A type of machine learning that uses deep neural networks

9 Machines Can Learn New Strategies

10 p 10

11 Applications of the Second Wave Vision and Language Recognizing faces or cancer cells Voice recognition Interacting with physical world Mobility Fine and gross motor control Problem Solving Answering unstructured questions Pattern recognition and classification

12 We ve Crossed the Inflection Point

13 Google Home Speech Recognition

14 Supervised Learning Systems

15 Supervised Learning applications Input X Output Y Application Voice recording Transcript Speech recognition Historical market data Future market data Trading bots Photograph Caption Image tagging Drug chemical properties Treatment efficacy Pharma R&D Store transaction details Is the transaction fraudulent? Fraud detection Accident history Expected future claims Insurance underwriting Purchase histories Future purchase behavior Customer Retention Car locations and speed Traffic flow Traffic lights Faces Names Face recognition

16 Supervised Learning applications Input X Output Y Application Voice recording Transcript Speech recognition Historical market data Future market data Trading bots Photograph Caption Image tagging Drug chemical properties Treatment efficacy Pharma R&D Store transaction details Is the transaction fraudulent? Fraud detection Accident history Expected future claims Insurance underwriting Purchase histories Future purchase behavior Customer Retention Car locations and speed Traffic flow Traffic lights Faces Names Face recognition

17 Applications of the Second Wave Vision and Language Recognizing faces or cancer cells Voice recognition Creating narratives Interacting with physical world Mobility Fine and gross motor control Problem Solving Answering unstructured questions Pattern recognition and classification

18

19

20 Machine Learning Can Assess Changes in Property values Decline Improvement Source: MaxGradient

21 Applications of the Second Wave Vision and Language Recognizing faces or cancer cells Voice recognition Creating narratives Interacting with physical world Mobility Fine and gross motor control Problem Solving Answering unstructured questions Pattern recognition and classification

22

23 Insurance Industry: Percentage of tasks potentially affected Job Title Percentage Insurance Sales Agents 60% Insurance underwriter 35% Claims adjusters 26% Actuaries 15% CEOs 25% Source: McKinsey

24 What does this mean for the business and the economy?

25 The Great Decoupling

26 Cause for Optimism, but Also Challenges Digital progress makes the economic pie bigger. But there is no economic law that everyone, or even most people, will benefit.

27 Coinvention Ain t Easy Copyright (c) Erik Brynjolfsson

28 The New Grand Challenge Digital technologies will continue to accelerate. Our skills, organizations, institutions and metrics are lagging. Business as usual won t solve this problem. We need to reinvent our skills, organization, institutions and metrics to keep up with accelerating technology Copyright (c) Erik Brynjolfsson

29 Legal notice 2017 Swiss Re. All rights reserved. You are not permitted to create any modifications or derivative works of this presentation or to use it for commercial or other public purposes without the prior written permission of Swiss Re. The information and opinions contained in the presentation are provided as at the date of the presentation and are subject to change without notice. Although the information used was taken from reliable sources, Swiss Re does not accept any responsibility for the accuracy or comprehensiveness of the details given. All liability for the accuracy and completeness thereof or for any damage or loss resulting from the use of the information contained in this presentation is expressly excluded. Under no circumstances shall Swiss Re or its Group companies be liable for any financial or consequential loss relating to this presentation. 29