Becoming Comfortable with AI: A Conversation with Andrew Ng

What is it that makes people so fearful about artificial intelligence?

You could say that we always have a learning curve when it comes to something new, and that that can trigger anxiety in people. But this technology wave seems different from anything that has come before. There’s really this enormous idea of giving over control to machines that a lot of people are just uneasy about.

But what are the specific fears that are most common?

I did some research and pulled up a list of top AI fears from the Pew Research Group. I’m just going to list some of them:

• Job displacement.
• Loss of human connection
• Deepfakes and imitations
• AI becoming too powerful to manage
• Smart AI taking over infrastructure
• Bias and discrimination

Some of the other things that researchers found are really different flavors of these broader concerns.

Alongside these, there’s just that intuitive feeling that we need to keep a lid on this type of powerful digital sentience until we better understand how it works and what it can do in the long run.

Tips For Getting Comfortable with AI

At the same time, there are certain things you can do to relieve many of your fears, and become more confident about interacting with the new technologies.

This resource from NM State Global Campus talks about a few overall tips.

One of these suggestions is to use free AI tools to start out looking at what they do firsthand. You might generate some images with Dall-E or create a little movie with Sora, or ask ChatGPT some things you’ve been dying to know.

What you’ll probably find as you interact with the technology is that you start to see not only its capabilities, but its limitations, and you can relax a little bit about the AI being superhumanly powerful. You’re also see how it can be assistive to you as a person.

That leads to the second point: automating boring tasks, or delegating unpleasant things to the AI itself.

Maybe you don’t want to go looking for a flight or hotel reservation – an AI can do that for you. You might not want to spend time coming up with detailed lists of items for a professional presentation, and you can just ask an AI model.

One of the best examples is search – even Sam Altman has told reporters that he uses ChatGPT to replace traditional search engine practices when looking for things. The AI goes out and scans the entire web for what you want, and brings it to you almost immediately.

Another tip is to take courses with AI to learn more about how it works. You might run into strange lingo like Boltzmann machines, or echo state networks, or anything else like that, but you will start to get a better handle on what people are doing as they engineer interactions with the technology.

Feedback from Davos

I was able to interview Andrew Ng at our Imagination in Action event in Davos.

“The tone here last few days has been very encouraging,” he said, comparing the summit to its counterpart in 2024. “I think last year that (we heard) a lot of hype and fears about AI: where is this going? Will we get to superintelligence? I think this year, at Davos, it’s a lot more tangible discussion of concrete business implementations of AI.”

In addition, he mentioned how businesses can use AI for experimentation:

“It turns out that the cost of experimentation has plummeted, because AI-assisted prototyping is actually very efficient,” he said. “So I’m seeing a lot of businesses reorganize themselves to run many experiments and then, in a systematic way, let the successful experiments get the resources to then grow to scale.”

AI for Business Success

Later in the interview, Ng brought up a main point that I think is very important when you’re talking about applying AI to business.

He said that the technology can either “double or halve” revenue and production.

We heard this kind of thing back in the age of the cloud, when people were talking about how implementations can either help or hinder a human workforce. The key is an intuitive interface and a technology that people can understand and use easily.

In other words, your application has to fit your workforce. If it’s something that people don’t like and can’t learn how to use, you might actually lower productivity in the long run and disrupt the business in a bad way. If on the other hand, it’s something that is intuitive, you’re likely to boost productivity dramatically.

So those are some helpful hints when it comes to integrating AI into your business and personal life. Being confident means you’re less likely to sit around worrying about the negative ramifications, and more likely to be using the positive applications to your advantage.