Question

One might expect Evie, an AI personal assistant, to resemble "Hollywood AI", i.e. AI tools seen in movies. Discuss the drivers that may lead users to have such expectations.

Answer 1

1. Capital

Capital is “rushing” into the AI space to the tune of $2.4 billion in 2015. In the first half of 2016 alone, there were 200+ AI startups that raised over $1.5 billion. It’s an understatement to even say deal activity is fast and furious.

2. Algorithms

Jacobstein pointed to algorithms like deep learning and its hierarchical pattern recognition as a major force driving the adoption of AI. With software like RStudio and Sentient, companies who would never have thought about getting into AI suddenly can.

3. Hardware

Whether it’s Alphabet’s recently announced tensor processing unit (TPU), Qualcomm’s new neural processing unit (NPU), Nvidia’s deep learning chip, or IBM’s TrueNorth neuromorphic computing platform, more chips are being developed to enable faster and more powerful AI.

4. Data

As important as hardware is to AI, large data sets are where machine learning algorithms really learn by refining hypotheses iteratively. From real-time information discovery to the integration of algorithms and data with TensorFlow, more tools for working with data are enabling analysis of an increasing number of publicly available datasets.

5. Talent

For all the focus on software and hardware, Jacobstein’s thesis is that humans are as critical to the AI equation as machines. He points to the number of AI startups (Turi, Nervana, and DeepMind, to name a few) in recent years in which talent was the primary driver for acquisition.

6. Applications

Today, we experience AI through the applications we use. It delivers value by augmenting human skills and extends our capabilities. But this isn’t an overnight thing. When first released, Siri had plenty of issues, but over time, it’s become increasingly useful and more companies have released virtual assistants, such as Microsoft’s Cortana and Amazon Echo.

7. Responsibility

The final driver of AI adoption that Jacobstein highlighted was responsibility, especially noting that AI comes with tradeoffs that often boil down to trust. While applauding efforts like OpenAI to democratize access to AI, he emphasized the need for AI to demonstrate core human values. In the business world, this means that ultimately the winning formula is humans-plus-AI processes.