Here's what the biggest names in tech and business think AI means for white-collar jobs
Anthropic CEO Dario Amodei kicked off the conversation by warning about how quickly large language models are advancing.
"We, as the producers of this technology, have a duty and an obligation to be honest about what is coming," Amodei told Axios. "I don't think this is on people's radar."
Amodei said it can seem weird that the AI companies would warn about their own technology.
Sam Altman
OpenAI CEO Sam Altman said some jobs will go away, but society will adapt.
"And the hard part about this is, I think it will happen faster than previous technological changes. But I think the new jobs will be better, and people will have better stuff," Altman said during a live episode of The New York Times' "Hard Fork" podcast in June.
Altman said that even if it were true that such a large number of jobs were about to be wiped out, "the inertia of society" wouldn't allow for it.
"And the take that half the jobs are going to be gone in a year or two years or five years or whatever — I think that's just — I think that's not how society really works," he said. "Even if the technology weren't ready for that, the inertia of society, which will be helpful in this case, is like — there's a lot of mass there."
Jensen Huang
Nvidia CEO Jensen Huang didn't mince words.
"I pretty much disagree with almost everything he says," Huang told reporters of Amoedi's views at VivaTech 2025 in Paris. "He thinks AI is so scary, but only they should do it."
Huang said that he's much more optimistic.
"If you want things to be done safely and responsibly, you should do it in the open," Huang said, likening AI development to medical research, where transparency and peer review are essential. "I believe AI is not that expensive. Do I think AI will change jobs? It will change everyone's — it's changed mine."
Marc Benioff
Salesforce CEO Marc Benioff said he's seeing no evidence of such a near-immediate upheaval.
"That isn't how I see AI," Benioff said during a recent onstage interview at the 2025 AI for Good Global Summit. "Maybe they have AI, I don't have. But in the AI I have, it's not going to be some huge mass layoff of white-collar workers, it is a radical augmentation of the workforce."
Benioff encouraged people to "shed their fear" about AI.
"When I'm talking to our customers, I'm not hearing them say, "Oh, now I'm laying off these people because this A,B,C technology increase because of AI.' So, I think we need to somehow shed the fear of what that all means."
Jim Farley
Ford CEO Jim Farley said he sees problems ahead.
"Artificial intelligence is going to replace literally half of all white-collar workers in the US," Farley said during an appearance at the Aspen Ideas Festival.
Farley said he's concerned that too much of the American education system is focused on four-year degrees instead of trades.
Mark Cuban
Mark Cuban said the situation will be the opposite of Amodei's warning.
"Someone needs to remind the CEO that at one point there were more than 2m secretaries. There were also separate employees to do in office dictation. They were the original white collar displacements," Cuban wrote on in a post on Bluesky.
"New companies with new jobs will come from AI and increase TOTAL employment," he continued.
Brad Lightcap
Like Altman, OpenAI's COO Brad Lightcap doesn't see the sky falling.
"We have no evidence of this," Lightcap said during the "Hard Fork" podcast taping. "And Dario is a scientist. And I would hope he takes an evidence-based approach to these types of things."
Lightcap said that every technology changes the job market.
"I think every time you get a platform shift, you get a change in the job market," he said." I mean, in 1900, 40 percent of people worked in agriculture. It's 2 percent today. Microsoft Excel has probably been the greatest job displacer of the 20th century."
Andy Jassy
Amazon CEO Andy Jassy said that AI is already changing workflows. He said it will soon lead to a reduction in some jobs.
"As we roll out more Generative AI and agents, it should change the way our work is done," Jassy said in a memo posted to the Amazon website. "We will need fewer people doing some of the jobs that are being done today, and more people doing other types of jobs."
Sebastian Siemiatkowski
Klarna CEO Sebastian Siemiatkowski said AI may cause a recession due to the sheer number of job cuts.
"I don't want to be one of them," Siemiatkowski said of CEOs who downplay the changes AI will bring. "I want to be honest, I want to be fair, and I want to tell what I see so that society can start taking preparations."
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
CNBC
an hour ago
- CNBC
VCs use this 'power law' to decide whether to invest in startups: We want 'outsized-outcome potential'
As pitches from founders come flooding in, one investor has revealed how VCs use a historic "law" to identify the best investments. "The power law in venture capital basically means that a small number of investments generate the vast majority of returns. So essentially, one or two massive winners outweigh all other investments combined," Peter Specht, general partner at European VC firm Creandum, told CNBC Make It in an interview. He described it as "significant in decision-making for VCs." The power law originates from the work of late-19th-century Italian economist Vilfredo Pareto. He noticed that 20% of the pea pods in his garden produced 80% of the peas, and applied this observation to economic conditions and wealth disparity. Specht explained that, as VCs often finance ideas that are risky and innovative, they know that some of these startups will fail. But the returns from the companies that succeed can become "massive winners," which "outweigh all other losses or smaller exits," he said. "From a return perspective, you want to really hit the outliers. We want the Spotifys, the Revoluts, the Klarnas," he said. "Outcomes at that size, or that scale, return more capital to investors and founders than five, billion-dollar exits. A $50 billion exit is way bigger than five $1 billion exits." Although the route to securing venture capital in Europe is improving, there's still some way to go, according to Atomico's State of European Tech Report 2024. In fact, the average seed round — the first official round of fundraising — in Europe was $1.4 million last year, compared to $3 million in the U.S., Atomico found. It's even harder at the growth stage, with American startups twice as likely to raise funding rounds of more than $15 million. Some 60% of respondents told Atomico in 2024 that it was harder to raise external finance compared to the previous year. Specht said one key way to convince VCs that you're building the next Spotify or Revolut is to show that your company has "outsized-outcome potential." "You need to pitch why it can become so big, why it can become a category-defining company," he said. This includes explaining how you're tackling the market and how the product can expand over time. At the seed stage, having an early product or pilot users with positive feedback can help. If you don't have the product yet, there needs to be evidence that the product is filling a gap in the market and is offering a genuine solution to customers. VCs also look for "exceptionally strong and ambitious founders," Specht said, adding that he wants to understand their motivation and drive. "One thing we obviously ... like to see is if founders have a spike in a particular area, or a skill, and shown excellence in that," he said. This could be work-related, such as having technical or leadership capabilities, or even excelling in sports or studies. Given the gap between fundraising in Europe and the U.S., Harry Stebbings, the founder of 20VC, identified a difference in the way founders pitch. American founders tend to be more bold and ambitious, Stebbings said, compared to their counterparts across the Atlantic. "I don't think Europeans are as good fundraisers [as Americans]. We need to tell better stories. We need to market ourselves better. We are too self-deprecating," he told CNBC Make It. Some founders have even been "badly" advised to include "exit slides" in their pitches, Stebbings added. "That makes me feel sick, like I'm planning my divorce when I get married." While Specht disagrees that European founders lack ambition, he thinks that startups in the region could tell better stories. "We have people and founders that dream really big and want to build massive companies," he said. "In the general culture between Europe and the U.S., I think U.S. founders and Americans are pretty good in storytelling and also pretty good salespeople." "There, we can focus on being even more bold, having an even more refined and visionary storytelling," Specht added.
CNBC
an hour ago
- CNBC
U.S. firms scramble to secure rare-earth magnets — imports from China surge 660%
China's exports of rare-earth magnets to the United States in June surged more than seven times from the prior month, as American firms clamor to get hold of the critical elements following a preliminary Sino-U.S. trade deal. In April, Beijing placed restrictions on several critical magnets, used in advanced tech such as electric vehicles, wind turbines and MRI machines, requiring firms to receive licenses for export. The move was seen as retaliation against U.S. President Donald Trump's steep tariffs on China. Beijing has a stranglehold on the production of rare-earth magnets, with an estimated 90% of the market, as well as a similar hold on the refining of rare-earth elements, which are used to make magnets. The U.S. received about 353 metric tons of rare-earth permanent magnets in June, up 660% from the previous month, data released by the General Administration of Customs showed, though the exports were about half that from June last year. The U.S. was the second-largest destination for China's rare-earth magnets, behind Germany, as it relies heavily on their imports for its large manufacturing sector, particularly in automotive, electronics, and renewable energy. In total, China exported 3,188 metric tons of rare earth permanent magnets globally last month, up nearly 160% from May, but 38% lower compared with the same period last year. The growth in exports came after Washington and Beijing agreed last month on a trade framework that included easing controls on Chinese rare-earth exports as well as a rollback of some American tech restrictions for shipments to China. AI behemoth Nvidia said last week it was planning to resume shipments of its H20 AI chips to China, after the exports were restricted in April. Last month, controls on American AI chip software companies' business in China had also been rolled back. Chinese rare-earth magnet producers started announcing the approval of export licenses last month. If exports continue to increase, it will be of great benefit to companies that have been suffering from shortages of magnets due to the lengthy time required to secure export licenses. For example, several European auto-parts suppliers were forced to halt production in recent months. The magnet shortages had also hit emerging industries such as humanoid robotics. In April, Elon Musk said production of Tesla's Optimus humanoid robots had been disrupted. China's controls on its rare-earths sector have prompted some global governments to reexamine their rare-earth supply chains and search for ways to support domestic mining of the minerals. However, experts say that setting up alternatives to China's rare-earth magnet supply chain could take years, as it requires an intricate process of rare-earth element refining and separation. "The separation process is quite complex, and China has a lot of advantages in this after putting in decades of research into the processes," Yue Wang, a senior consultant of rare earths at Wood Mackenzie, told CNBC last month. One way that the U.S. has been trying to compensate for lack of rare-earth magnets is through increased recycling. Apple and miner MP Materials announced a $500 million deal last week for the development of a recycling facility that will reinforce the iPhone maker's U.S. magnet supply chain. Peter Alexander from financial consultancy Z-ben Advisors said that Washington's latest concessions on tech restrictions were a reflection of just how much leverage China has in its trade relationship with the United States, speaking on CNBC's "China Connection" on Monday.
Yahoo
2 hours ago
- Yahoo
I'm an avid traveler and finally found a useful way to use ChatGPT to plan my trips that saves me hours
ChatGPT saved me hours of research when planning a trip to Door County, Wisconsin. Whereas I normally do a lot of background research before planning a trip, AI could do it for me. I found it was really good at giving a comprehensive overview of a destination. Before I plan a trip, I have a straightforward goal: Learn everything there is to know about the place I'm visiting. I know, I know. That sounds time-consuming. And truthfully, for me it is. But for ChatGPT? I stumbled upon this AI use case when starting to plan a relatively uncomplicated trip to Door County, Wisconsin, for later this summer. Rather than doing the hours of background reading that I usually do to get the lay of the land in a new destination, I let OpenAI's chatbot do it for me. For context, I travel frequently in my personal life and cover travel for Business Insider, but until now, I'd yet to find a use case for AI that I felt really made my trip planning process more efficient. I'd experimented with AI-powered trip planning tools but had never found them particularly useful. I'd also tried using ChatGPT as a glorified Google, describing to it vaguely what my interests were and asking it to recommend restaurants or attractions. The results were less than promising. From what I could tell, our individual tastes are still too personalized — and chatbot answers too universal — for specific recommendations to be helpful. But for replacing all my pre-planning background reading? It was great. AI does all my background reading for me Let me give you a sense of my usual process. Once I know I am visiting a place, I will Google the most generic things a tourist could think to ask: Top attractions. Must-do activities. Neighborhood guide. Best restaurants. One-day itinerary. Three-day itinerary. Weeklong itinerary. I will do all of these searches, open more tabs than any browser should be reasonably expected to host, and then, I read. I read the top 10 or so results for every search. Then I do more niche searches like best neighborhoods to live in or best vintage shopping, and do the process all over again, this time also rifling through countless Reddit threads where locals discuss the goings-on in their own neighborhoods. Next, I move to social media — often TikTok — to scroll all the videos I can find about the destination to get some visual context and, most importantly, to learn which restaurants or attractions are viral so that I can avoid them. It takes hours, and it's arguably more information than a tourist technically ever needs. But by the time I am done, I feel I have a shockingly full understanding of a place, as both a tourist destination and even as a place where real people live. I feel I could recommend to my friends which neighborhood would be uniquely right for them, which highly-rated restaurants are overrated, and which niche museum is actually a lot cooler than the one ranked first on TripAdvisor. The process is excessive, but it gives me confidence that when I get to the hard planning stage — selecting a neighborhood, narrowing down hotels or Airbnbs, booking restaurant reservations — that my hard-earned PTO is being put to its best possible use. Luckily for me, it turns out ChatGPT is pretty good at doing this. ChatGPT helped me plan my trip to Door County, Wisconsin I was recently planning a trip to Door County, which is a small peninsula in northeast Wisconsin situated between Lake Michigan and Green Bay that's known for being one of the prettier natural places in this part of the Midwest. I wanted to book a house on the water for a busy travel weekend, so I knew I needed to do it quickly. So, instead of embarking on my usual trip planning odyssey, I did something that I had previously been very skeptical about: I turned to AI. The kind of comprehensive overview that I get from reading all the top Google results, ChatGPT was able to give me with just a few prompts. I asked which popular attractions are frequently called overrated and which ones people say are worth weathering the crowds. I asked it to include any tips or tricks on the best times to visit certain places, and to provide several sample itineraries that were varied so I could get a complete picture of my options. I even described my vague travel preferences — good food, good drinks, nature, away from crowds, vintage shops, where locals actually go — and asked it what town I should stay in. It gave me a quick summary of what each of the towns were best known for and which were most likely up my alley. I also asked follow-up questions and played devil's advocate, as AI can tend to repeat marketing-speak or be overly optimistic. In about half an hour I felt like I understood visiting Door County almost as much as I would've if I had spent those hours consuming everything myself. It's able to summarize the 90% of recommendations that pop up on every list and then also include the more unique ones. The responses were not perfect. It recommended at least two restaurants that closed several years ago. And to be clear, I doubt that I know Door County as well as I would if I had done all that reading myself. But I felt like I knew it enough to be confident in my choices while planning — enough so to book a rental home that same night, a decision that would generally take me a lot more time. And yeah, I'm still going to do a bit of excessive reading for the hidden gems. What can I say? Old habits die hard. Read the original article on Business Insider
