October 14, 2016      

President Barack Obama recently talked about artificial intelligence with Wired. While Obama’s interview is insightful by itself — it’s not every day the most powerful man in the world talks about such technologies in depth — there are several AI tips behind his insights that are valuable for companies and countries looking to take advantage of artificial intelligence and robotics.

Here are five takeaways from President Obama’s interview:

1. Government: The missing piece?

What will it take for AI to advance faster? Right now, most investments are coming from the private sector. But what about government investment and leadership?

Currently, AI investments from the U.S. federal government stand at less than $1 billion. Obama said he was reminded that the space program cost half a percent of U.S. GDP, which is equivalent to $80 billion today. Obama went on to say that he believes government spending in AI will “accelerate.”

What does this AI tip signal? Two things. First, the government wants to play a bigger role in the way AI is developing.

Second, it can do this by increasing its investments to higher levels. This would affect which applications get priority.

2. Forget cybersecurity; focus on AI security

What keeps President Obama up at night? Pandemics. Why? Because there is no real way to stop an airborne disease from spreading. Instead, there need to be systems in place all over the world to track such diseases as they emerge. Obama said that a similar approach is required to deal with AI threats.

President Obama offers AI tips to Wired.

“Part of what makes us human are the kinks. They’re the mutations, the outliers, the flaws that create art or the new invention, right?” -President Obama to Wired.

And that’s because cybersecurity threats can now emerge from any part of the world, targeting any industry, government, or company with any number of objectives.

Yes, current computer viruses operate along similar lines, but the difference is that, in the future, AI would enable cyberattacks to react and adjust in real time. This means that future digital defenses also need to be “thinking” to keep up.

Is this something you can offer — a new form of security that is smart enough to protect a network or database from AI attackers?

3. Build human understanding

Wired mentioned the term “extended intelligence,” which is the idea of building human understanding into artificial intelligence.

Among his AI tips, Obama refers to an ethical challenge around self-driving cars.

“If the car is driving, you can swerve to avoid hitting a pedestrian, but then you might hit a wall and kill yourself,” he said. “It’s a moral decision, and who’s setting up those rules?”

AI tip: Robots need to understand us and our values.

Robots need to understand human values, noted the president.

The next wave of AI needs to go beyond simply replicating tasks. It needs to understand the cause and effect of those tasks. AI that can keep track and update inventory is one thing. AI that can also understand how increasing inventory levels will affect the overall budget and influence future spending decisions is something else.

The president pointed out new requirement for AI: It won’t work unless it has “human understanding.”

4. The military needs these AI tips

A key section of the interview had to do with the age-old question of whether artificial intelligence will take over the world. Obama provided his own insights, which suggest where the U.S. military is focusing its efforts.

The president said that he tells his national security team not to worry about a doomsday-type scenario regarding AI, but instead look into “specialized AI” — AI with a specific goal, like acquiring nuclear codes and launching missiles.

If this AI tip is a mandate for the U.S. military, it reduces the ambiguity about what a company can sell to the U.S. government.

Can you provide an AI bot that can plug gaps in real time? Or, do you have a tool that senses intruders and takes immediate action without human intervention?

More on AI and Robotics:

5. Automation changes everything

A large chunk of the Wired interview centered around the potential job losses, economic inequality, and changes in the social contract that could occur once AI goes mainstream. Obama touched on ideas such as a universal basic income, saying they will be debated in the coming decades.

He also noted that those at the top of the skills and pay scale tend to do well when new technologies surface, unlike those at the lower end of the ladder. However, AI will affect both ends, and this will eventually lead people to question the true value of certain skills such as teaching and nursing — based on what people actually need.

This AI tip is an important business point for two reasons.

First, Obama’s comments demonstrate that the U.S. government is paying attention to potential for job losses from automation. Will it introduce regulations to address any rise in unemployment stemming from technology?

Second, if consumers will begin questioning the value of certain skill sets in the age of automation, then your product/service can fall along a pay scale that doesn’t make any sense (compared with existing markets).

Will AI-based tutors cost $30/month while AI-based doctors cost $1/year simply because the value attachment to such professions shifts when artificial intelligence comes into the picture? How will your business or current profession compete in such a model?

President Obama is one of the first world leaders to speak publicly and candidly about AI threats and policies. This demonstrates just how important AI is to the U.S. government as it looks ahead.

The question is, will this importance translate into concrete action, or will it fade as politics and economics take center stage?