Understanding And Preparing For The Seven Levels Of AI Agents

As the calendar flips to the second quarter of the century, conversations about the transformative potential of Artificial Intelligence (AI) are reaching a fever pitch.

Jan 03, 2025

However, the buzz about AI is shifting from AI tools to creating and deploying AI agents. Many executives I speak with remain unsure about how to conceive, categorize, and capitalize upon the various agentic possibilities for their businesses. Understanding the evolution of AI agents—from simple reactive systems to hypothetical superintelligent entities—can provide a roadmap for organizations aiming to harness AI strategically.

The following framework I offer for defining, understanding, and preparing for agentic AI blends foundational work in computer science with insights from cognitive psychology and speculative philosophy. Each of the seven levels represents a step-change in technology, capability, and autonomy. The framework expresses increasing opportunities to innovate, thrive, and transform in a data-fueled and AI-driven digital economy.

Level 1: Reactive Agents | Responding to the Present

At the most basic level are reactive agents, which operate entirely in the moment. These agents do not retain memories or learn from past experiences. Instead, they follow predefined rules to respond to specific inputs. Reactive systems have their roots in early AI research and finite state machines, foundational concepts that emerged in the mid-20th century through the work of pioneers like John McCarthy and Marvin Minsky.

A quintessential example is a basic chatbot that answers questions based on keyword matching, or one that generates or translates content. These agents excel in environments where the scope of interaction is limited and predictable. For businesses, reactive agents can streamline repetitive tasks, such as handling customer queries or automating well-defined workflows.

More on Level 1 and the other six levels of AI Agents on Forbes

One Billion AI Agents Are Coming | ai16z

What is the role of AI in crypto, and how will it shape the future?

AI agents are taking the crypto world by storm, introducing unprecedented scale and efficiency to Web3. At the center of this revolution is Shaw, the creator of the Eliza framework and ai16z DAO, a groundbreaking initiative that’s reshaping decentralized investments.

The ai16z DAO has rapidly gained traction, becoming the #1 trending GitHub repository last month, with over 3,300 stars and 880+ forks. In this episode, we explore how Shaw and his team are leading a community-driven, open-source movement to integrate AI and crypto.

From managing DAOs to bridging Web3 with real-world applications, we dive into how the Eliza framework is setting the groundwork for a potential AGI future. This is more than just an AI experiment—it’s the beginning of a new paradigm for both industries.

Using Smolagents For Code Automation

If you’re not on the younger side, or you’re not familiar with a lot of the newer Internet tropes and memes, you might think you’re seeing a typo. But the use of something called smolagents to modularize artificial intelligence workflows is actually based on DoggoLingo – that new dialect developed to give voice to cute canines everywhere.

If you do know something about DoggoLingo, you may be familiar with sentences like “I am smol, do me a protec” or “Am smol potat.” If so, then the name of this new automated code design process will be more familiar to you. So let’s talk about what these are, and what they do.

A Rubric for AI Impact

First, you have the idea of the AI agent in general. The agent is an AI entity that does things actively instead of passively. There’s the idea that through AI agents, we’re going to give over power to our LLM brethren. But the Hugging Face survey of what smolagents are also gives you a neat agency “schedule” to show you how this works. There are five agency levels, with escalating LLM workflow impact.

At level one, the LLM output has no impact on the program flow. At level two, LLM outputs can determine basic elements of control flow. At level three, LLM outputs can determine function results. At level four, LLMs can “control iteration and program continuation.” Level five is the big one, where one agentic workflow can trigger another agentic workflow, or in other words, two agents can work together.

Hard Fork Podcast 2025 Tech Predictions

In this episode of the New York Times Hard Fork Podcast, hosts Kevin Roose, and Casey Newton discuss their tech predictions for 2025 — surprise, they include AI.

I’ve partnered with Moving On IT to help evangelize their IT products and services.

Moving On IT is your authorized partner for IT, AI and Cybersecurity hardware and software solutions. Moving On IT has been awarded an Ingram Micro partnership, which provides access to all the leading manufacturers. From Aisera to WatchGuard, and everything between — as well as managed, and on-site professional IT services.

Contact Moving On IT today for a complimentary quotation from their knowledgable team of industry experts. Call MOIT at (727) 490-9418, or email info@movingonit.com.

Dictatorships Will Be Vulnerable To Artificial Intelligence Algorithms

It may seem like a perfect fit: dispassionate software that streamlines the agendas of dastardly regimes. But they’ll find that the tech cuts both ways.

AI is often considered a threat to democracies and a boon to dictators. In 2025 it is likely that algorithms will continue to undermine the democratic conversation by spreading outrage, fake news, and conspiracy theories. In 2025 algorithms will also continue to expedite the creation of total surveillance regimes, in which the entire population is watched 24 hours a day.

Most importantly, AI facilitates the concentration of all information and power in one hub. In the 20th century, distributed information networks like the USA functioned better than centralized information networks like the USSR, because the human apparatchiks at the center just couldn’t analyze all the information efficiently. Replacing apparatchiks with AIs might make Soviet-style centralized networks superior.

Nevertheless, AI is not all good news for dictators. First, there is the notorious problem of control. Dictatorial control is founded on terror, but algorithms cannot be terrorized. In Russia, the invasion of Ukraine is defined officially as a “special military operation,” and referring to it as a “war” is a crime punishable by up to three years imprisonment.

More on how AI would cut both ways on dictatorships

AI Can Do Standup: Here Are Some Actual Jokes From Anthropic’s Claude

Anthropic’s Amanda Askell prompted their AI model Claude to create a stand-up comedy routine, revealing its flexible intelligence and ability to mimic human-like humor. Through jokes about JavaScript debugging, hallucinations, and AI alignment, Claude highlighted its capacity for clever observations and nuanced humor.

However, one joke about smaller models raised questions about “punching down.”

This experiment demonstrated the importance of prompting in shaping AI creativity, with Claude’s performance showcasing the potential of language models to simulate human-like personas and reasoning. As AI continues to evolve, this playful exploration hints at its broader capabilities in creativity and storytelling.

Samsung’s Smart Fridges Will Use AI To Suggest Groceries To Buy On Instacart

Samsung Bespoke fridges equipped with AI-powered cameras will soon be able to add items you need to your shopping cart.

Samsung’s smart fridges will soon be able to identify when you’re running low on something and add items to your Instacart app so you can order what you need from the grocery delivery service right from your fridge. Today, Samsung announced a multiyear partnership with Instacart that will let you shop for groceries from the screen on your Samsung Bespoke fridge — the 32-inch one or the one with the new 9-inch screen the company is debuting at CES this month.

According to the press release, the tech uses Samsung Vision AI food recognition technology to identify what you have in your fridge and determine what you’re running low on. Then, using Instacart’s product matching API, it suggests items from the service you might need and lets you order them from the fridge.

The service uses “AI Vision Inside” on Samsung’s fridges, which leverages a camera above the fridge door to see when you put items into your fridge as well as when you take them out. It also has cameras inside to keep an eye on what you have in there, although it can’t see items in the refrigerator door bins or freezer.

More on Samsung’s AI Fridge and Instacart partnership on The Verge

What’s Our Relationship to AI? It’s Complicated | AC Coppens, Kasley Killam, Apolinário Passos | TEDNext

In a lively conversation from TED's brand-new Next Stage, social scientist Kasley Killam, technologist Apolinário Passos and futurist AC Coppens explore the intricate dynamics of human-AI relationships — and show how AI is already changing the ways we live, work and connect with each other. Recorded at TEDNext 2024.

Thats all for today, however new advancements, investments, and partnerships are happening as you read this. AI is moving fast, subscribe today to stay informed.