The AI Daily Brief: Artificial Intelligence News and Analysis - Predictions on How AI Agents Will Actually Be Used

Episode Date: December 2, 2024

A reading and discussion inspired by https://time.com/7178872/agents-unlimited-age/ With predictions from NLW on how we will and won't use AI agents. Brought to you by: Vanta - Simplify compliance - ⁠⁠⁠⁠⁠⁠⁠https://vanta.com/nlw Plumb - AI automation that just works - https://useplumb.com/ The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: https://pod.link/1680633614 Subscribe to the newsletter: https://aidailybrief.beehiiv.com/ Join our Discord: https://bit.ly/aibreakdown

Today on the AI Daily Brief, Mark Benioff on the era of agents. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. To join the conversation, follow the Discord link in our show notes. Hello, friends. We are back with another long reads episode of the AI Daily Brief. And as is, I expect, going to be a big topic as we come into end of your summary and next year predictions. We are discussing agents. Specifically, we're reading an essay from Salesforce CEO.

Mark Beniof about the agent era that is quickly emerging. Now, Beniof has a very particular stake in this. He has verbally assaulted Microsoft for selling the world a bill of goods with the assistant era of AI and has gone all in on their agent force platform as the future of Salesforce. In the essay we're about to read, or rather I'm going to turn over to the 11 Labs version of myself to read, Benioff discusses the disruption that's coming down the line and what it'll mean on the other side. Let's listen to what he has to say, and then I'm going to come back and play a quick game of agent predictions. Over the past two years, we've witnessed advances in AI that have captured our imaginations with unprecedented capabilities in language and ingenuity.

And yet, as impressive as these developments have been, they're only the opening act. We are now entering a new era of autonomous AI agents that take action on their own and augment the work of humans. This isn't just an evolution of technology. It's a revolution. It's a that will fundamentally redefine how humans work, live, and connect with one another from this point forward. Today, we're already used to predictive AI, which analyzes data to provide recommendations, forecasts and insights, and generative AI, which learns from data and uses patterns to seamlessly generate text, images, music, and code. Agents are software components that go far beyond this. They can perform tasks independently, make decisions, and even

negotiate with other agents on our behalf. And unlike the tradition, tech transformations of the past which required years of costly infrastructure buildout, these new AI agents are easy to build and deploy, unlocking massive capacity. This is a new horizon with radical implications. For the first time, technology isn't just offering tools for humans to do work. It's providing intelligent, scalable digital labor that performs tasks autonomously. Instead of waiting for human input, agents can analyze information, make decisions, and take action independently, adapting and learning as they go. Take, for example, a large retailer during the upcoming holiday season. Traditionally, human workers or pre-programmed software might handle customer

inquiries or inventory updates. But now, intelligent digital agents can respond to customer questions in real time, monitor stock levels, reorder inventory, and even coordinate with shipping providers, all without human intervention. These agents are enabling an entirely new scale of operations that was previously not possible. This shift to intelligent digital labor is our already unlocking capacity across industries. It's no longer constrained by human availability or physical limits, allowing businesses to scale their operations while driving down costs and improving responsiveness or by geographical limits. Like any change of this magnitude, the shift to agents comes with clear challenges and understandable fears. We need to make sure AI systems are

built with trust, accountability, fairness, and transparency as core values. We need to make sure as AI transforms how we work, that we invest in the training, creativity, and critical thinking skills that are uniquely human, and recognizing AI's impact on our carbon footprint, we need to make sure that we're investing in sustainability, ecopreneurs, and nature-based solutions. If we face and address these concerns, it's possible to envision new levels of abundance enabled by an expansive digital workforce that learns and grows more capable all the time. The potential of agents isn't limited to businesses. These technologies have the potential to profoundly enhance the lives of individuals as well. We'll all have access to specialized agents that can navigate

different parts of our lives. For example, every student will have an own, always-on, personalized tutor, an agent embedded in their everyday technology that acts as an intelligent companion guiding them throughout their learning journey at every stage. Our personal agents communicating with other agents will help manage our daily routines from ordering groceries for us to scheduling appointments. AI agents are already transforming how we deliver health care. We know that doctors and nurses are facing tremendous burnout, and there are provider shortages in many communities. As one of the world's largest medical systems has recently discovered using our platform, agents can alleviate administrative burdens, improving patient communication, while giving providers the space to focus on

complex cases that demand their expertise. Over time, patients will have access to an AI agent that reaches out to check on you after a procedure, reminding you to follow up on test results, and asking if you have any unexpected reactions that need to be addressed, it will monitor patient progress and even reschedule labs or appointments as needed, all while maintaining a detailed understanding of the patient's medical history and ongoing treatment. Some companies will struggle to adapt. Nearly every job will change in some ways. And yes, some will go away. In the past, we've seen companies and sometimes entire industries rise and fall with new inventions, jets, satellites, the internet, the smartphone, renewable energy. Ultimately, however, these innovations create far more

new jobs that they displace. In 1950, for example, 43 million Americans had jobs. By 2020, over 152 million Americans were employed. Multiple factors played a role, but that is more than 100 million new jobs in a period of profound technological change, many in categories that did not exist before. The key, as always, will be to invest in the education and training that equips workers and young people with the skills to succeed in the new jobs and industries to come. The benefits AI agents bring both individuals and businesses will far outweigh the initial disruptions. After all, growth in a country's GDP is the product of growth in the labor force and in productivity. With the labor force stagnating or even shrinking in some regions in industries, countries will need to rely more than ever

on boosting productivity, especially in the services sector, which is now the bulk of modern economies. Today, with the human labor force growth stagnant in many places, exceptional productivity driven by a digital agent force is vital for GDP growth. agents amplify human labor driving innovation and efficiency. Productivity rose 2.2% in the third quarter of 2024 fueled in part by AI. Finally, AI agents will drive innovation. It will continue to jumpstart countless new companies, just as the birth of the microchip more than 65 years ago spurred the creation of iconic companies like Apple, Dell, and Microsoft. More than 5,000 new artificial intelligence companies have been funded in the U.S. alone over the last decade. This two will create significant

numbers of new jobs, both within tech and across the global economy. All of this is a reminder that technology itself is neither good nor bad. What matters is how we use it. Without proper oversight and training data, autonomous AI can make choices that conflict with our wishes or even with human values or ethics, such as prioritizing profit over safety or discriminating against certain groups. Harnessing the power of agentic AI effectively will require a multi-stakeholder approach businesses, governments, nonprofits, and academia working together to create guardrails and guidelines. We are already seeing some of this in efforts such as the framework by the G7 nations, emphasizing accountability, transparency, safety, and data privacy. Another example is

the Bletchley Declaration by 28 countries in the European Union that emerged from the UK AI Safety Summit that I and other tech CEOs attended last year, agreeing to collaborate on AI safety and development. AI itself can play a role in guiding us through the disruptions to come. As MIT economist David Otter argues, AI has the potential to act as a leveling force, lowering barriers to entering the workforce by giving people access to tools and knowledge that were once reserved for a privileged few. Look for example at how our agent force platform is transforming the college admissions process for the nonprofit college possible. In many areas of the country, high school college counselors are responsible for many hundreds of students, making individual

guidance almost impossible. In less than a week, College Possible used our platform to create a virtual college counselor for high school students. Now, any student can get college prep support to augment sessions with a human counselor. This virtual counselor tracks the conversations in those sessions, has deep knowledge of colleges and accesses student transcripts already housed within college possible to provide guidance. This is a powerful expansion of labor potential, unlocking new ways to support students where traditional resources have been limited. In fact, we're already seeing similar possibilities in recruiting and human resources. Billions of resumes are submitted each year, but finding a job shouldn't feel like submitting your resume into a void. The Adeko Group,

one of the world's largest recruiting companies, handles 300 million job applications a year, but historically can only respond to 10%. It's now using our platform to pre-qualify applicants, enabling it to engage with every applicant within 24 hours, while freeing human recruiters to work more closely with candidates on the way to job placements. I've always believed that business is the greatest platform for change. Today, as we stand at the brink of this new agentic era, I've never been more confident in the transformative change that's possible. AI has the potential to elevate every company, fuel economic growth, uplift communities around the globe, and lead to a future of abundance. If trust is our North Star as we navigate this

new landscape, agents will empower us to make a meaningful impact at an unprecedented scale. Today's episode is brought to you by Vanta. Whether you're starting or scaling your company's program demonstrating top-notch security practices and establishing trust is more important than ever. Vanta automates compliance for ISO-2701, SOC2, GDPR, and leading AI frameworks like ISO-42,001, and NIST AI Risk Management Framework, saving you time and money while helping you build customer trust. Plus, you can streamline security reviews by automating questionnaires and demonstrating your security posture with a customer-facing trust center, all powered by Vanta AI.

Over 8,000 global companies like Langchain, Lila AI, and Factory AI, use VATO Vanta to demonstrate AI trust and prove security in real time. Learn more at Vanta.com slash NLW. That's Vanta.com slash NLW. Today's episode is brought to you, as always, by Superintelligent. Have you ever wanted an AI daily brief but totally focused on how AI relates to your company? Is your company struggling with AI adoption, either because you're getting stalled figuring out what use cases will drive value or because the AI transformation that is happening is siloated individual teams. departments and employees and not able to change the company as a whole.

Super Intelligent has developed a new custom internal podcast product that inspires your teams by sharing the best AI use cases from inside and outside your company. Think of it as an AI Daily Brief, but just for your company's AI use cases. If you'd like to learn more, go to be super.aI slash partner and fill out the information request form. I am really excited about this product, so I will personally get right back to you. Again, that's besupor.a.i slash partner. All right, we are back now. I think a lot that Beniaf has to say here is dead on, even though,

of course, he is not a particularly unbiased source. What I thought would be fun to round out this episode is to give you a few of my quick hit predictions when it comes to agents. Specifically, a few common things that people think agents are going to do. I'm going to go through and say, first, the thing that people think agents will do in the future, and then whether I think it's actually going to come to pass. And let's start with one where I may be on a very different side than many personal assistant agents. It seems like every generalist agent uses ordering food or booking flights as its example of what it can do. So do I believe that that will become normal behavior? Absolutely not. I fairly fundamentally don't believe that we will turn over these

types of tasks that involve ultimately not a lot of effort and a ton of nuanced detail that's really hard to capture and guess at. Part of this is just that I don't think there's all that much pain in booking flights or ordering food. But the other part is that by nature of being humans, we're carrying a million different ways of looking at any different decision that could change on a moment's notice. I think programming the AI to be as good as us and making decisions for ourselves is going to be a lot harder than people think. Now, will there be agents to help with things like reviewing flight options and doing research? Sure. I just think the actual act of outsourcing the buy button, is mostly about showing off what agents can do, rather than really

solving a big problem that people actually have. But now let's move over into agents in the workplace. The big question I think that people are asking is, will agents replace roles? In other words, will agents replace entire job categories? The short answer is yes in some very specific situations, but more broadly, I think that the replacement and disruption is going to happen at the task and activity level. If one views a job as a collection of tasks and activities that by virtue of the person being who they are can add up to a whole that is more than the sum of its parts, there are a lot of tasks in anyone's given role that are going to be good fodder for disruption and automation. I think we will see entire categories of work that we do agentified and moved off of our plates.

I think the number of jobs that are completely made up of tasks like that is actually, fewer than we think. And so this gets to my next question, will agents transform roles? And that answer is absolutely yes. I believe that they will transform roles in the way that I just mentioned by automating certain categories of tasks and activities. But as to the question of whether in the future everyone is going to be a manager, where they have a group of agents at their disposal who do things for them, I actually kind of think that answer is yes as well. And this is one where a lot of very smart people aren't as sure as I am. I think that the modality of viewing an agent that's good at a particular thing as an employee or a consultant or a contractor to deploy against a particular problem that

you're trying to solve actually might end up being a pretty good way of looking at agents. I think it's a way that people will begin to be able to integrate them and view agents as fundamentally additive in allowing them to accomplish more. Flipping back to the consumer side, will we have to get used to interacting with agents? And the answer for here, is absolutely definitely yes. Already we interact pretty frequently with very unsophisticated agents, and the opportunity to up level from that into a customer service system that's actually sophisticated, good at solving your problems with agents, but also good at routing you to humans when they can't be solved easily with agents, I think it's going to be the way that most of us

have our first interactions. Lastly, will agents be released to generate businesses on their own? This is where we get a little bit more far out, but I think that the answer here is definitely going to be yes as well. The question will be what guardrails we set around them, but there was a while ago an idea that maybe the new Turing test should be whether an agent can start a business from scratch and make a million dollars. If starting a crypto meme coin and getting to a million dollar market cap counts for that, then it's already been breached. But I think in terms of making an actual productive business, it's kind of an interesting heuristic. Anyways, there is going to be a lot more agentic discussion coming up. Let me know how you're thinking about them, share in the comments

on YouTube or on Spotify. For now that that is going to do it for this Long Reads episode, appreciate you listening or watching as always. Until next time, peace.