Episode transcript The transcript is generated automatically by Podscribe, Sonix, Otter and other electronic transcription services.
Hello, everyone. Here is Ricardo Vargas, and this is the 5 Minutes podcast. And today, I'm going to save you a lot of time by summarizing the 456 pages of the AI Index Report 2025 from Stanford University. Stanford publishes this every year, and it tries to make sense of what is happening in AI in the current year. And the best part of this summary is that this time, I'm not using artificial intelligence to do that, okay? I use the notebook LM to help me find the best pieces related to project management to share with you, but on this podcast, for sure, it's me here; it's not AI. Okay. And let's go straight to the point. What are the 11 key takeaways from the report, and how do they impact our projects and our profession? So the first insight is that AI performance continues to grow exponentially. Don't think that the ChatGPT you used two years ago is; it's even closer to the ChatGPT you are using today. Models are now outperforming humans in complex tasks like coding under time constraints. And for us, this means faster projects, learner teams, and a much higher pressure for top performers. It's to do more with less. More. More with less and less. The second insight is that AI is being applied just everywhere, from medicine to transportation. Just to give you an example, over 220 AI-powered medical devices were approved in 2024. This means that more and more projects will involve AI integrations requiring new technical skills from project managers. So it will be very easy to see project managers working with AI agents, working with AI on a day-to-day basis, like working with another team member. The third insight is real productivity gains and an astonishing record-breaking investment. Just to give you an idea, in 2024, AI investments in the US alone, I'm talking only the United States, reached a staggering US$ 109 billion. And just to put this in perspective, that is enough money to buy 100% of the Brazilian oil company Petrobras. That is the most valuable company in Brazil. And also with the remaining investments, you can buy 100% of Valley, one of the largest mining companies in the world. And just last year. And just the United States. I'm not counting China. I'm not counting Europe. I'm not counting any other country. So that shows how AI is not just a hype. It's where real money is going to drive productivity and efficiency. And the world never saw this amount of money being put into something like what we are seeing in 2023 and 2024, in AI. The fourth insight is about China. China is closing the gap with us in AI performance. The country is accelerating its scientific publications, model development, and AI applications. Just take the example of Manus and DeepSeek. If you manage international projects, these could impact partnership suppliers and even compliance because it's, it's a real technical war that is happening to see who will lead to the new frontier of AI performance. The fifth insight is about responsibility. Ai responsibility is still far behind where it should be, according to the report, few models undergo standardized, standardized audits and ethical incidents are on the rise. You know, fake news, fake images, fake videos. So this is happening pretty much everywhere. And that means that when we are working with AI, we must strengthen governance, risk management, and ethics, mainly when we are using and developing AI. The sixth insight is a global optimism about AI. But this optimism has not spread in the same way globally. For example, Asian countries like China and India are far, absolutely far more optimistic than Western nations. And it means that, for example, a lot of Western nations see an AI with much more skepticism about the future and the future of work, and why this matters to us. Because this affects how change is perceived and adopted across cultures and regions. So if we are developing a product using AI, we should expect a different perception of what will happen based on the different countries we are operating and we are developing. The seventh insight. AI is becoming cheaper, faster, and more accessible. For example, the inference cost, the cost of running a trained model in real time, has dropped more than 280 times in recent years. This is like taking technology that once belonged only to elite labs and big techs and making it available to startups and individuals. So it means it opens the door for us to experiment with AI in small projects, in small project office operations, in small project teams to explore advanced solutions. So it's not restricted anymore to just the, I would say, the big investment projects. The eighth Insight government is stepping up into the AI game. The US, France, India, Canada, and China are investing billions in shaping new regulations and creating an environment to foster AI innovation. So, public or regulated projects will need to adapt to a changing environment in terms of legal, ethical, and transparency in how AI behaves in this project. So this is something that is picking up. And it was a very clear difference between last year's report and this year's report. The ninth insight: AI education is growing, but readiness is still low. So many teachers and professionals don't yet feel prepared to teach or apply AI in practice. Many people still think that AI is just for improving their email or their social media posts. And this was, I would say, a trend in 2022. But today, this is pretty much close to nothing compared to the potential of AI and project managers, and they have a critical role in enabling this continuous learning. So we need to make sure that project managers know how to use AI and know how to apply AI in their work, because probably it will be a massive barrier for those who do not dominate AI to survive and thrive, I would say in few years from now and when I say few years, I'm saying probably 1 or 2 years from now. Okay. The 10th insight. Innovation in AI is now led by industry, not academia. Over 90% of notable models came from companies like OpenAI, Google, Anthropic, and DeepSeek. This shift in how we approach applied research, scale solutions, and think about innovation inside our projects. And last but not least, the 11th insight. AI is becoming a driver of scientific breakthroughs. It was behind major discoveries that awarded, for example, Demis Hassabis, the 2024 Nobel Prize in Chemistry for the AlphaFold turning awards. So, scientific and technical projects are increasingly depending on AI. So AI is driving these scientific breakthroughs in different areas. And we, as project leaders, need to navigate this intersection of research data and delivery. So, now what does all of this mean to our profession? It means that we must, of course, adapt and keep learning. I just wanted to give you an idea: In the past 45 days, I have taken three courses on AI. I'm spending about 10% of my time just studying what is happening, trends, and how this is reshaping the way I perceive my projects. And it's not just about managing scope, time, risks, and stakeholders. No, it's about evaluating the value of my project. So we saw, for example, PMI doing research on value and AI. It's shifting that, for example, maybe I'm producing a project where everything is perfectly fine. But the benefit that I'm producing, for example, the main characteristic of the product I'm producing, AI will do that. And then maybe the benefit I'm delivering with my project will be irrelevant in the future because, you know, AI will be doing that easily in the future. So this kind of landscape is what we need to be prepared, because it's a transition that requires from us ethics, agility and a massive strategic vision. So think about that. You can find this report online. It's free. Okay. If you go to a search engine and look for the AI Index Report 2025, you can download it. And it's nice for you to take a look at it and maybe find other insights on it. Okay. It's very important. This is probably the most relevant report we have in the market today. So I hope you enjoy this podcast. I'm sorry for it being too long, but I didn't want to split this into two episodes, so I hope you enjoy it. I'll see you next week with another 5 Minutes podcast.
