Hear 50 technology leaders talk about AI. Here's what I learned.

This year, I've listened to and interviewed more than 50 technology leaders, from executives running trillion-dollar companies to young founders betting their future on AI.

Those building the future of AI continued to return to the same four themes across boardrooms, conferences, and podcast interviews.

1. Use AI. Because someone who understands AI better may be able to take your place.

This is the line I hear most often. Nvidia CEO Jensen Huang has said this many times this year.

“Every job will be affected immediately, there's no question about it. You won't lose your job to AI, but you will lose your job to someone using AI,” he said at the Milken Institute World Conference in May.

Other technology industry leaders echo his sentiments, with some saying that younger workers may actually have an advantage because they are already accustomed to using AI tools.

Sam Altman, CEO of OpenAI, told Cleo Abram's YouTube show Huge Conversations in August that it is inevitable that AI will eliminate some roles, but that college graduates are better able to adapt.

“If I was graduating college at 22 years old now, I would feel like the luckiest kid in history,” Altman said, adding that a bigger concern is how older workers will fare amidst the reinvention of AI.

Feifei LiThe Stanford University professor known as the “godmother of AI” said in an interview on “The Tim Ferriss Show” published earlier this month that resistance to AI is a deal breaker. She said her company, World Labs, won't hire engineers who refuse to use AI tools.

This change is already visible in day-to-day roles. Accountants and HR professionals said they are using AI tools, including vibecoding, to level up their skills and stay relevant.

2. Soft skills will become more important in the AI ​​era

Another consensus I've heard among technology leaders is that AI will increase the value of soft skills.

Peter Schwartz, Salesforce's chief futures officer, said in a May interview that rather than coding knowledge, “the most important skill is empathy and working with other people.”

“Parents ask me what their kids should study, can they become programmers? I say, 'Learn how to work with others,'” he said.

Chua Pei-Yin, lead economist for Asia Pacific at LinkedIn, also told me in July that soft skills such as communication and collaboration are becoming increasingly important for experienced employees and new graduates.

As AI automates parts of our jobs and slims down our teams, the human part of the job is starting to become more important.

3. AI is evolving rapidly and superintelligence is coming

As the years go by, the stakes around the future of AI feel bigger and more real. Technology leaders are increasingly talking about pursuing artificial general intelligence (AGI) and, ultimately, superintelligence.

AGI refers to AI systems that match human intelligence across a variety of tasks, while superintelligence refers to systems that exceed human capabilities.

Altman said in September that society needs to prepare for superintelligence, which could arrive by 2030. Mark Zuckerberg founded Meta's Superintelligence Research Institute in June and said the company is moving toward superintelligence.

These leaders don’t want to miss out on the AI ​​moment. Mr. Zuckerberg emphasized the urgency in September, saying he would rather risk “wasting hundreds of billions of dollars” than delay the superintelligence.

Some technology leaders, such as Databricks CEO Ali Ghodsi, have argued that the industry has already achieved AGI. Some people are more cautious. Google DeepMind co-founder Demis Hassabis said in April that AGI could arrive “within the next five to 10 years.”

Even if technology leaders differ on timelines, they tend to agree on one thing: advances in AI are becoming more complex.

I saw this acceleration from the outside as a user. New tools are being rolled out at a rapid pace, from ChatGPT adding shopping functionality and image generation to China's AGI cameras.

What felt like magic in January now feels normal.

4. Humans need to be at the center of AI

Many leaders have also returned to the need for human control amid the acceleration of AI.

Mustafa Suleiman, head of AI at Microsoft, said superintelligence must support human agency, not override it. He said this on an episode of “Silicon Valley Girl Podcast” released in November. that He says his team is “trying to build a humanistic superintelligence” and warns that systems smarter than humans will have difficulty containing or aligning with human interests.

Anthropic CEO Dario Amodei spoke frankly about the risks posed by AI if it is misused.

Advanced AI can lower the barriers to knowledge work, but the risks magnify along with the rewards, Amodei said in the New York Times' “Hard Fork” episode in February.

“If you look at our responsible scaling policy, it's AI, autonomy, CBRN – chemical, biological, radiological, nuclear,” Amodei said.

“This is about a serious abuse of AI autonomy that could threaten the lives of millions of people,” he added.

Geoffrey Hinton, also known as the “Godfather of AI,” said in August that as AI systems surpass human intelligence, protecting humanity will become a central challenge.

“We have to make sure they care about us even when they're stronger and smarter than us,” Hinton said at the Ai4 conference in Las Vegas.