60% of US Jobs Could Be Impacted by AI

0:00Today on the AI Breakdown, new research from the IMFace suggests 60 % of US jobs could be impacted by genitive AI. Before that on the brief, Elon wants more control of Tesla before he builds more advanced AI and robotics. The AI Breakdown is a daily podcast and video about the most important news and discussions in AI. Go to Breakdown.net for more information about our YouTube, our Discord, and our newsletter. Welcome back to the AI Breakdown Brief, all the AI headline news you need in around five minutes. Well, Elon Musk is back in the news today, and it is surrounding both Tesla and artificial intelligence.

0:35The TLDR of the situation is that Elon Musk, who owns around 13 % right now of Tesla, wants his ownership stake to be around 25 % if he is to continue building advanced artificial intelligence and robotics applications at that company. So where is this coming from? Well, of course, it's coming from Twitter slash X. On January 14th, Tesla investor Jacob Hilton tweeted, When you have as much Tesla shares as Elon has, why would you need another huge compensation package to get motivated? I know everyone wants Elon to tweet 24-7 about Tesla, but how often do we hear creators complain about getting trapped in a niche?

1:08It's the same with Elon. Let him tweet about what he's interested in. Elon is and always has been hell-bent on accelerating the world's transition to sustainable energy. I don't think he needs a$1 trillion compensation package to stay motivated to do that. I'm not going to complain if he gets a new package for, let's say,$100 billion worth of Tesla shares if he gets Tesla to$5 trillion, but some of the numbers being thrown around are too extreme in my opinion. Yeshu Sharma quote tweeted and said, this is the right take. Elon owns around

1:38Now to that, Elon himself responded. He wrote,

2:07He continues,

2:15for a decision in my Delaware compensation case. The trial for that was held in 2022, but a verdict has yet to be made. I put compensation plan in quotes because from my standpoint, this is primarily about ensuring the right amount of voting control at Tesla. If I have 25%, it means I am influential, but can be overridden if twice as many shareholders vote against me versus for me. At 15 % or lower, the for or against ratio to override me makes a takeover by dubious interest too easy. I would be fine with a dual class voting structure to achieve this, but I'm told that it's impossible to achieve post-IPO in Delaware.

2:45So basically what Elon is saying is that he doesn't care about the money, he cares about the control, and that he doesn't want supreme control, he just wants it harder to be outvoted than it is currently. Now as you heard from that tweet, it is directly about AI and robotics that he's concerned. Many notable folks, including people like Kathy Woods, have long argued that Tesla is in fact the most important AI company in the world. Now whether one accepts that or not, the logic for it is that Tesla is where artificial intelligence meets the real world, where AI jumps off the screen and into machines that actually do things and interact in physical environments.

3:19As if to make the point, Elon posted a video of the Tesla Optimus robot folding a shirt. Now, the business community's reaction is pretty mixed. Tesla shares were down nearly 3 % in pre-market trading, and of course, Elon's mixed affiliations are already challenging for investors in his various companies. But given how high the stakes of AI and robotics really are, Is it possible that Elon will get what he wants? We will have to wait and see. Next up, one of the biggest themes of discussion in AI this year is, of course, how it is going to interact with the presidential election cycle in the U.S., as well as elections around the world.

3:53But obviously, the big dance is the big dance. Well, OpenAI has just released a new blog post about how they are approaching the challenges, announcing a number of new policies and initiatives. The company writes, Protecting the integrity of elections requires collaboration from every corner of the democratic process, and we want to make sure our technology is not used in a way that could undermine this process. So what are their actual strategies for doing so? The first pillar is around preventing misuse. So some updates to their usage policies in the context of elections. They write, We're still working to understand how effective our tools might be for personalized persuasion.

4:28Until we know more, we don't allow people to build applications for political campaigning and lobbying. So basically, you cannot use the OpenAI API if you're building tools that are around trying to convince people to do things or believe things. Similarly, they write, People want to know and trust they are interacting with a real person, business, or government. For that reason, we don't allow builders to create chatbots that pretend to be real people, e.g. candidates, or institutions, e.g. local government. A third one, which is interesting, because those first two are pretty common sense, is that they write, We don't allow applications that deter people from participation in democratic processes, for example, misrepresenting voting processes or qualifications.

5:02e.g. when, where, and who is eligible to vote, or that discourage voting, e.g. claiming a vote is meaningless. Now, obviously, the misinformation dimension of that is fairly straightforward. The one that's more interesting to me is the discouraging voting part. That's taking a little bit more of an editorial stance, but I think they're clearly erring on the side of caution, given the severity of the potential problem. The second pillar of their approach is transparency around AI-generated content. They say, Early this year, we will implement the Coalition for Content Provenance and Authenticity's Digital Credentials, an approach that encodes details about the content's provenance using cryptography, for images generated by DALI 3.

5:36They also say that they are continuing to work on a tool for detecting images that are generated by DALI. They write, Our internal testing has shown promising early results, even where images have been subject to common types of modifications. They also talk about how more attribution is coming to ChatGPT, writing, Transparency around the origin of information and balance in news sources can help voters better assess information and decide for themselves what they can trust. Finally, they are trying to have more definitive information when it comes to certain key voting-related questions. Their third pillar is improving access to authoritative voting information.

6:08They write, In the United States, we are working with the National Association of Secretaries of State, the nation's oldest nonpartisan professional organization for public officials. ChatGPT will direct users to CanIVote.org, the authoritative website on U.S. voting information, when asked certain procedural election-related questions, for example, where to vote. Lessons from this work will inform our approach in other countries and regions. For my part, I think this is going to be a massive test this year, not only of these AI defense systems, but how much the social response has evolved in terms of people's expectations about what's real and what's not.

6:40I think that's going to be hugely telling for the future of the development of generative AI. Moving over to the world of product releases, Microsoft is officially rolling out Copilot to all of its consumer office users. Now, Microsoft has been touting these tools for Word, Excel, PowerPoint, and other software for months now, but they've been mostly available just to business customers. Now, individuals can pay for that$20 subscription to unlock those tools inside their own suite of Microsoft applications. On the one hand, Microsoft Copilot may seem less sexy than some of the other new tools that we get access to on a daily basis these days.

7:14But when it comes to mass impact and how many people will have their very first experiences with generative AI, this is probably just about as expansionary as anything you can imagine when it comes to an enterprise or professional audience. Lastly, one slightly sadder story. Kevin Sistrom, the founder of Instagram, has announced that they are shutting down news curation app Artifact after just about a year. Artifact was basically attempting to use artificial intelligence to provide people personally curated news. But ultimately, they write, We have built something that a core group of users love, but we've concluded that the market opportunity isn't big enough to warrant continued investment in this way.

7:48Kevin goes on, It's easy for startups to ignore this reality, but often making the tough call earlier is better for everyone involved. The biggest opportunity costs is time, working on newer, bigger, and better things that have the ability to reach many millions of people. We live in an exciting time where artificial intelligence is changing just about everything we touch, and the opportunities for new ideas seem limitless. Now, while you might see some people try to use this as exemplary of a narrative of artificial intelligence startups facing a reckoning this year, which might be a possibility, funding dynamics are still really challenging in VC world, I take it more at face value.

8:21Kevin is someone who has learned that it really does in some cases when it comes to startups make sense to rip the bandaid off earlier. And he's right to identify that opportunity cost, exploring new areas and building other things that'll have a bigger impact as the biggest reason to not let something linger. So while this might be sad, especially for those people who loved this product, it undoubtedly means that more energy is now released into the universe to build even more exciting things. Good luck to all of the team moving on to new pursuits. And that's going to do it for today's AI Breakdown Brief.

8:51Coming up next, the main AI breakdown. Welcome back to the AI Breakdown. Right now in Davos, Switzerland, the World Economic Forum is hosting its annual gathering of the world's global leaders in intelligentsia and, depending on who you ask, Illuminati. And given the kickoff of that event, numerous companies and research firms are releasing their latest studies and surveys, and as might not surprise you, artificial intelligence is probably going to be the most discussed topic at this event. One of the reports that was released was from the IMF. Over the weekend, the IMF blog published their most recent survey, which found that almost 40 % of jobs around the world could be affected by artificial intelligence.

9:33Now, they make clear that some portion of those jobs will be augmented by AI, where people will find greater productivity. On the other hand, some meaningful portion could be replaced entirely, in other words, automated away. Fascinatingly, the study found that this 40 % was not distributed evenly across the world. We tend to be used to automation impacting the bottom of the socioeconomic pyramid first, but in this case, the IMF argues that it is in fact highly skilled and technologically enabled workers that are going to see the biggest impacts initially. So, for example, in low-income countries, AI exposure to jobs is around 26%.

10:09Emerging markets see that baseline 40 % exposure, but in places like the US and other advanced economies, the IMF argues that around 60 % of jobs might be impacted by AI. Now, in terms of what percentage of those 60 % of jobs that are impacted will be impacted positively or negatively, the IMF writes, roughly half of the exposed jobs may benefit from AI integration, enhancing productivity. For the other half, AI applications may execute key tasks currently performed by humans, which could lower labor demand, leading to lower wages and reduced hiring. In the most extreme cases, some of these jobs may disappear.

10:43So you're basically split down the middle of 30 % seeing a productivity improvement, with 30 % seeing their jobs fundamentally threatened or changed. Now these numbers are huge. We've never had any sort of force moving this fast that impacted this percentage of the workforce in such a dramatic way. Because of that, the IMF is particularly concerned about how this might impact inequality. They argue that both within and between countries, we could see some serious polarization. They basically say that workers who can harness AI might see an increase in their productivity and wages, while those who can't fall behind.

11:14They think there's an age dynamic, given that research has shown that one of the impacts of AI is catching less experienced workers up more quickly. Overall, they write, the effect on labor income will largely depend on the extent to which AI will complement high-income workers. If AI significantly complements higher-income workers, it may lead to a disproportionate increase in their labor income. Moreover, gains in productivity from firms that adopt AI will likely boost capital returns, which may also favor high earners. Both of these phenomena could exacerbate inequality. Now one thing that this report isn't talking about is the potential that those higher income workers could themselves see some of their skills replicated by lower income or remote workers in other places, which would put downward pressure on wages.

11:53But still, ultimately, the IMF concludes in most scenarios AI will likely worsen overall inequality. Their argument? They write it is crucial for countries to establish comprehensive social safety nets and offer retraining programs for vulnerable workers. Now of course there are many sides of this debate. Speaking with Yahoo Finance at Davos, Bill Gates argued that AI is great for white-collar workers and coders. He said, I found it's a real productivity increase. Likewise, for coders, you're seeing 40 or 50 % productivity improvements, which means you can get programs done sooner. You can make them higher quality and make them better.

12:25So mostly what we'll see is the productivity of white-collar workers will go up. Now, of course, the pessimistic take on this is the replacement of a lot of coders because code assistants can do their jobs for them now. But the optimistic take that many folks in tech have is that basically humans have an insatiable appetite for new creation. And that to the extent that coders can be made more productive, we won't have some basal set rate of innovation and new things built that we'd like, where some number of people will all of a sudden sit idle. Instead, we'll just create more things, leveraging those new productivity gains to just build more.

12:58What about business leaders more broadly? One of the big surveys to come out in advance of Davos was from PwC. Hold aside everything else, what's clear is that they believe that AI is going to have a big impact. 68 % of U.S. CEOs that were surveyed think that generative AI will increase employee productivity in the next 12 months. Around 50 % of U.S. CEOs expect generative AI to help themselves become more productive. 44 % of those CEOs say they see generative AI providing a net increase in profits in the next 12 months. That's versus just 3 % who anticipate a net decrease. PwC wrote,

13:59invention. Here are some of the other statistics. When asked, in the next three years, generative AI will significantly change the way my company delivers, creates, and captures value, 68 % of US CEOs agree. When asked if generative AI would increase competitive intensity in their industry over the next three years, 65 % of US CEOs agree. When asked in the next 12 months if generative AI would improve the quality of their company's product or service, 61 % agreed. And here's a crazy one. In the next three years, generative AI will require most of my workforce to develop new skills. On that, 61 % of US CEOs and 69 % of global CEOs agreed.

14:34Now, there are some negatives as well. As part of the same report, around a quarter of CEOs expected that AI would lead to job cuts in the coming year. Specifically, 25 % said that generative AI would lead to job cuts of at least 5 % this year. The industries that were most likely to expect AI-related layoffs included media and entertainment, banking, insurance, and logistics. Now, the PwC report wasn't the only one that came out. Deloitte's AI Institute also surveyed 2 ,800 director-to-C-suite-level executives and found that only one in five said that their organization is highly or very highly prepared to address AI skills needs in their company.

15:09Just one in four say that their organizations are well-prepared to address AI governance and risks, and only 47 % said that they are sufficiently educating employees about AI. The majority said that right now when it comes to AI, they're focused on tactical benefits, specifically things like improving efficiency, reducing costs, rather than innovation reinvention, and as Axios put it, using it to create new types of growth. But it's really clear that there is a big, blaring talent question, and even more than talent, a skills question. Again, only one in five think that their organizations are highly or very highly prepared to address AI skills needs in their company, which seems to me like a massive opportunity for someone.

15:44Now that is part of an overall state of generative AI in the Enterprise report, which has just a ton of good information. Interestingly, despite all of the negative headlines, these leaders are definitely more optimistic than you might think. 30 % report uncertainty, but 62 % represent excitement. 46 % say that they're fascinated by AI. 79 % said that they expect generative AI to transform their organizations over the next three years. When it comes to addressing this talent gap, 17 % said that they're making changes right now, 24 % said that they're going to make changes within a year, and 31 % said that they're going to make changes to their talent strategy over the next one to two years.

16:20Lastly, one more statistic that I thought showed just how big the stakes are in these CEOs' minds. And this one, we go back to that PwC survey for 45 % of the 4 ,700 global CEOs surveyed do not believe that their businesses will survive barring significant changes in the next 10 years. Now, of course, that doesn't mean that they think that their companies are going to go out of business, just that they assume that in this AI-powered world, they're going to have to make major, major changes. Really interesting stuff from these surveys, and certainly not, I don't think, the last that we will hear from this Davos event.

16:52For now, however, that is going to do it for today's AI Breakdown. Thanks for listening or watching as always, and until next time, peace.