Michael Lohmann

Jensen Huang is the founder and longtime CEO of Nvidia. He envisions Agentic AI as ubiquitous, spanning industries and use cases. Examples include:

• Knowledge Workers: AI research assistants simplify the analysis of complex documents like academic journals or financial reports, turning them into accessible formats like summaries or podcasts.

• Industrial Optimization: Agents monitor processes in manufacturing or logistics, identifying inefficiencies and recommending improvements.

• Software Security: AI agents continuously scan codebases for vulnerabilities, providing real-time feedback to developers.

• Healthcare and Drug Discovery: Virtual lab agents screen billions of compounds to identify promising drug candidates faster and more cost-effectively than ever before.

This indicates that AI Agents will be everywhere and performing any conceivable task.

AI Agents are poised to revolutionize how we work through three key transformations:

Autonomous Capabilities

AI Agents will manage complex tasks independently, learning from each interaction to improve performance. They'll coordinate across platforms, handle routine workflows, and adapt to new situations using contextual understanding.

Decision Support

These agents will process vast amounts of data to provide real-time analysis and recommendations, running simulations and offering data-driven insights that enhance human decision-making.

Workplace Evolution

Organizations will restructure around AI capabilities, creating hybrid teams where humans and AI Agents collaborate seamlessly. New roles will emerge focused on AI orchestration, while traditional hierarchies flatten through automated management.

Impact Timeline

- Near-term (1-2 years): Basic task automation and simple decision support

- Medium-term (3-5 years): Advanced operations and complex process automation

- Long-term (5+ years): Full AI Agent ecosystems and transformed organizational structures

Key Preparations

- Invest in AI literacy and training

- Develop human-AI collaboration frameworks

- Create ethical guidelines for deployment

- Build adaptable work processes

The future workplace won't just automate tasks – it will create new possibilities through human-AI partnerships, enabling achievements beyond what either could accomplish alone.

I wouldn't call myself an expert in running. I have never run a marathon or any other race.

Here's how I first got interested in running:

I was never athletic, but at 45 years old, I knew I had to do something for my health and fitness. So I thought about which sport would suit me best, and running came to mind. Why? Because it's so easy - you need next to nothing for equipment and no gym membership. You can run everywhere and at any time.

Over the years, I have run a whopping 12,000 kilometers. Some years more, some years less, but I eventually came back to running.

You want to know how I did this? Sure thing - consistency is the answer. I tried fancy training plans with three runs per week and rest days in between, but I never managed to stick to the plan. Some days I didn't feel like running - maybe I had a cold or slept poorly - so I skipped. The next day it was raining, and I thought, "Not today, tomorrow. Tomorrow I will run." But I didn't. Before long, I wasn't a runner anymore.

I knew I had to do something different, so I told myself I would run first thing in the morning, no matter what. Some days I run more, some days I run a little less, but no matter what, I step outside in my running shoes and do at least a walk. That way, I don't break the chain.

There are 3 topics I have been thinking about a lot lately:

Topic #1: Meditation

I am interested in meditation because sometimes I feel overwhelmed. My hope is that over the coming years, as I continue to learn, I'll be able to find some peace and make sense of it.

Topic #2: Keto

I first got interested in Keto a year ago. I read about it on the internet and found it fascinating. Fluctuating glucose levels were bothering me for a long time and keto promised a cure for that.

Topic #3: Literature

I've been a reader since my youth and recently decided to dedicate more time to reading. It's one of those fun rabbit holes I find myself exploring simply because I enjoy it.

It would be great to connect with other people who are interested in these same topics—so if any of the above resonates, feel free to reach out!

Here are some real-world examples for each type of AI agent:

Simple Reactive Agents

- Customer service chatbots that respond to predefined queries

- Trading bots that execute based on price triggers

- Smart home devices responding to voice commands

Goal-based Agents

- Calendar assistants that schedule meetings considering participants' availability

- Route optimization systems for delivery services

- Task automation tools that complete workflows in sequence

Utility-based Agents

- Energy management systems optimizing power usage vs. cost

- Investment portfolio managers balancing risk and returns

- Dynamic pricing engines adjusting prices based on demand

Learning Agents

- Product recommendation systems adapting to user preferences

- Network security tools learning to detect new threats

- Manufacturing robots improving task performance through experience

We've seen most of these agents for years, but goal-based agents are expected to take off this year. We will dive deeper into these agents and the different frameworks next time.

You've surely heard about them, AI Agents are all over the internet. But what exactly are they? What can they do? Can I use AI Agents? Is it difficult? Are they dangerous? How can we ensure they're used safely?

There are a lot of different types of agents. Today we're talking about software agents that operate on the internet. AI Agents are software programs designed to perform tasks or make decisions with some degree of independence. Let's start with the basics by defining their key characteristics:

• Agents act autonomously, meaning they operate without direct human supervision

• They take in information from their environment through various data sources

• They have a clear goal or objective to achieve, such as finding the best price or monitoring systems

• They make decisions on your behalf based on their programming and the information they gather