Why the fuss about AI?
What is AI? It's not robotics. It's not the Terminator. Obviously, it's artificial intelligence, but it is an intelligence we can use to help us determine what's useful in all of the big data being collected through the Internet of Things (IoT). And we can teach it to tell us what's important to investigate there, too. AI won't replace humans, but it will give us a huge boost.
Paul Muller, vice president of strategic marketing at Hewlett Packard Enterprise, has a conversation with technology analyst Theo Priestley about AI, including what it is, how to use it, fears about it, and how it can help us long term. They talk about the myths around AI—such as it's a human replacement vehicle rather than a way to augment our work, or it's just software rather than hardware and software—and how artificial intelligence doesn't mean it's not intelligent and doesn't practice self-preservation.
Muller and Priestly discuss a number of examples of AI pushing the boundaries in music, big data, and predictions.
- Intelligence vs. augmentation: AI won't be a human replacement, but it will assist what humans work on and create today.
- Hardware has a role in AI, particularly around IoT and data center management.
- It's time to examine the human fears around AI and the reluctance to embrace it.
Shout-outs worth a look:
- On Intelligence: A book by Jeff Hawkins about how a new understanding of the brain will lead to the creation of truly intelligent machines.
- Emily Howell: A computer program created by David Cope, a professor of music at the University of California, Santa Cruz. Emily consists of an interactive interface that allows both musical and language communication.
- Singularity University: Co-founded by Ray Kurzweil, its mission is to educate, inspire, and empower leaders to apply exponential technologies to address humanity's grand challenges.
- Stephen Hawking's and Elon Musk's perspectives on AI.
Paul Muller: Hello, and welcome to another episode of Business Insider Studio. I'm your host, Paul Muller, and I'm joined by industry analyst Theo Priestley to talk about the topic of artificial intelligence. I guess both broadly, but also in the context of business.
Muller: So, Theo, let's start with what is artificial intelligence, because I think a lot of people confuse artificial intelligence with other forms of technology.
Theo Priestley: Yeah, sure. So you're right there, Paul. Artificial intelligence is kind of separate from robotics, and when people think of artificial intelligence they think of "Terminator" and walking machines and “the end of humanity” kind of thing. But you know, if you think of your everyday life, artificial intelligence is here to kind of help us improve our lives, make us more efficient, make our lives a little bit easier. Doing automating tasks, for example. So if you think of Siri, and Cortana and Google Now, they're a kind of a starting block for artificial intelligence.
They interpret our queries, and they feed us back information. And obviously they learn from that as well, so artificial intelligence is obviously a learning mechanism as well.
Muller: Now you're an industry watcher and have been for years. What have you noticed about the development in the space of artificial intelligence?
Priestley: I've seen that the research angle for artificial intelligence in trying to map the human brain, for example, and how that works is kind of dying off and people are now looking to the Facebooks, and Google, and Apple for the innovation around artificial intelligence because they are monitoring how people interact. And it's through how we interact ourselves, how we engage with our machines and everything, how we think and our intentions is where the breakthroughs are going to come from. It's also the information as well, so these companies are farming so much information that they have that in hand to be able to correlate certain actions and outcomes as well.
Muller: It's probably fair to say that I see the industry developing towards what I'd call augmentation rather than intelligence. So rather than replacing the humans, it tends to be extending what they're capable of doing. Would that be a fair assessment?
Priestley: Yeah. I mean they'll never replace us unless, like I say, "The Terminator" future kind of comes across, but again, it goes back to that “smart agent” kind of thing. So it's helping us achieve our goals. It's helping us be more productive in our lives. It's taking away some of the really mundane actions that we do and automating that as well, so it's improving and augmenting us.
Muller: Perhaps one way of looking at it is rather than replacing humans, artificial intelligence could be used to augment a human.
Priestley: Yes, certainly. I see artificial intelligence as a way of improving what we do, so like I said before, it's taking the information, making correlations, making decisions for us, improving our productivity, improving our lives, taking away the sort of mundane tasks that we do from day to day and automating that as well. So it gives us more free time, and it makes our lives a little bit better.
Muller: There are some practical applications to business that I think may start to replace humans. I'm not sure if you're familiar with, say, narrative sciences, which is software that now can take, for example, things like baseball scores or football scores and then combine it with the names of the players and turn it into a newspaper article. So we're seeing journalists being replaced or at least augmented by this sort of technology.
Similarly, there was a producer on iTunes, Emily Howell, and she produced I think three classical CDs, all of which are written by software. Emily Howell is not a person. It's a computer program. So I guess we're starting to see the line blur between augmentation and replacement.
The central point, though, is none of this involves a robot. This is actually more about the software than it is about the hardware. Do you see a role for hardware in the world of artificial intelligence?
Priestley: Oh, certainly. I think if we look at the Internet of Things, for example, the explosion of devices, I think artificial intelligence is going to be the central or one of the central pivots to actually control IoT. I think you'll need that sort of we can't do it as humans, it's just far too ... it's too much to manage, too much data, and we need artificial intelligence to be able to understand what's going on, control the world around us in terms of connected devices as well. So, yes.
Muller: Let's finish with the ultimate question, which is, what is the downside? We've talked about the upside and possibilities. What do we need to worry about?
Priestley: I think there's been a bit of doom mongering, for example, with Elon Musk and Stephen Hawking and they talk about the future where AI takes control and again, Skynet and "Terminator" sort of thing. I think the downside is perhaps our own fear of something that we don't fully understand. There was an interesting video on YouTube about a robot questioning orders from its human operator. He was asking it to walk forward from a table, and the robot perceived danger by walking off the table and so it questioned, “I'm not going to do that.” “I'm sorry, Dave, I can't do that,” sort of thing.
Muller: Yes. Yes.
Priestley: I think the downside is our knee-jerk reaction to “Hang on, why am I being questioned? You're a robot!” or “You're artificial intelligence.” So I think we need to sort of step back a little bit and understand. True, AI can actually help the world around us, give us so much more. Even, you know, crack codes, for example, like cancer and disease, and give us more options in making our lives better. And I think the downside is human fear.
Muller: Sounds fantastic, until we get locked in a luxury mansion in the woods. That's a movie reference for you all.
Muller: With that, if people want to know more we usually ask for a tip of the week. Have you got one?
Priestley: Yeah. Years ago I read a book by Jeff Hawkins called On Intelligence, and again, it harks back to building AI not by mimicking the human brain but by doing something different, and if anyone's interested in a really weird look at AI, then definitely go for that one.
Muller: With that, I want to thank you for joining us.
Priestley: Thank you.
Muller: If people want to find you, where can they go?
Muller: Fantastic stuff. Thanks so much for watching.