Category: AI Agents

AI Agents, Artificial Intelligence, Banking, cryptography, DARPA, Digital Currency, International Finance, Uncategorized, Yogi Nelson

How DARPA Battles Global Money Laundering with AI

Yogi Nelson

Namaste Yogis.   Welcome to the Blockchain & AI Forum, where your technology questions are answered!   As a bonus, a proverb is also included.  Today’s question, comes from Kirk in Chicago and he wants to know if the U.S. Department of Defense (DoD)is fighting global money laundering?

WHAT IS DARPA

Kirk, you came to the right place.  Yes, DoD is fighting global money laundering.  However, not with tanks and rifles, missiles, etc. as you might expect from a military organization.  Nope, the DoD says it has enlisted in the battle using advanced technology, including artificial intelligence!  I’ll explain shortly what this is all about but first let’s take a moment to understand what office within the DoD will lead the effort.  It’s a fascinating agency with a stellar record of accomplishments; its acronym is DARPA.  The formal name is the Defense Advanced Research Projects Agency.  If you like the internet, GPS, stealth technology, and autonomous systems you can thank DARPA as it was the driving force behind all of those innovations, plus numerous others, and perhaps even more that have been kept secret from the public as a matter of national security.  Generally, DARPA is known as the office with DoD responsible for developing technologies for military use.  Why DARPA is involved in global money laundering begs the question as opposed to leaving the job to the the U.S. Treasury?  But for now, let’s put that aside and examine the recent DARPA announcement regarding anti-money laundering.

DARPA ANTI-MONEY INITIATIVE ANNOUCEMENT

According to a January 23rd, announcement by DARPA, it (DARPA) wants to eliminate global money laundering by replacing the current manual, reactive, and expensive analytic practices with agile, algorithmic methods.  Good idea, but isn’t that the job of the U.S. Treasury and if it is why would DARPA be involved?  In other words, what is the nexus between money laundering and DARPA’s mission?  Well, the answer given by DARPA is “… U.S. adversaries launder billions of U.S. dollars yearly and half of North Korea’s nuclear program is backed by laundered funds.  Money laundering finances our adversaries’ weapons programs, global terrorism, and the illicit drug trade, all of which threaten U.S. national security”.  A stretch but okayyyyy…

WHAT PROBLEMS DOES DARPA INTEND TO SOLVE

DARPA is facing two challenges not one.  The first is the global money laundering issue.  Essentially, DARPA states that money laundering schemes often evade detection and disruption, because anti-money laundering (AML) efforts today rely on manual analysis of large amounts of data and are limited by finite resources and human cognitive processing speed.  The second problem requires the reader to read between the lines.  What is it?  Essentially, the financial industry writ large and the U.S. Treasury have been unable or unwilling to invest in the technology to combat the problem, hence DARPA will step in because the illicit funds are not just a finance issue—they are a national security concern. What is the DARPA solution?

DARPA SOLUTION—APPLICATION OF ADVANCED TECHNOLOGY

In their announcement, DARPA says it seeks new technological ideas that “… preserve privacy while setting a course to end to our adversaries’ financial warfare”.  The goal is to develop rapid graph-search algorithms to sift through financial transactions graphs for suspicious patterns and learn new patterns to anticipate future activities. Program success will depend on algorithms’ ability to learn a precise representation of how bad actors move money around the world without sharing sensitive data.”  A tall order indeed.  When it comes to moving illicit money around the world they may want to read the Panama Papers as a manual. 

DARPA envisions creating technologies that identify patterns of illicit financial behavior in a concise, machine-readable format that are also easily understood by human analysts.  Resulting algorithms would be able to search financial databases that matches patterns of illicit behavior, alerting to the likely presence of money laundering without directly sharing sensitive financial data.  DARPA does acknowledge that developing technology that enables information exchange while safeguarding Americans’ privacy will be a challenge. What an understatement! 

Companies interested in submitting bids should attend the DARPA bidders conference on February 20, 2025 in Arlington, VA. https://sam.gov/opp/22562059531b4979814b8b9fcd03e3ec/view.

Time to go but first a proverb from Mongolia: “… if the source is muddy, the end will be muddy too.”

Until next time,

Yogi Nelson

AI Agents, Artificial Intelligence, Banking, Blockchains, computer vision, cryptography, Digital Currency, Uncategorized, Yogi Nelson

AI Agents Explained: Earning Money in the Digital Age

Yogi Nelson

Namaste Yogis.   Welcome to the Blockchain & AI Forum, where your technology questions are answered!   Here no question is too mundane.  As a bonus, a proverb is also included.  Today’s question was submitted by Arcardio, in Puerto Rico, and he wants to know if AI agents are engaged in business and making money?

Arcardio, you came to the right place.  The answer is yes–AI agents are busy working for their owners and making cash!  I’ll explain that mind bending fact, but first what is an AI agent?

WHAT IS AN AI AGENT

An AI agent is an entity or system than can perceive its environment, process information, and take actions to achieve specific goals.  It is capable of autonomously performing tasks on behalf of a user or another system by designing its workflow and utilizing available tools.”  It operates autonomously, using sensors or inputs to gather data, and employs reasoning, decision-making, and sometimes learning capabilities to determine the best course of action.

HOW DO AI AGENTS WORK

To understand how AI agents work, I turned to Kalaidos University, Switzerland.  Kalaidos U says AI agents build on top of existing algorithms, in particular large language models (LLMs), e.g., ChatGPT4. Essentially, AI agents use machine learning algorithms to derive patterns, rules or models from the collected data. Based on this, AI agents can make decisions or perform tasks. As robots they interact with their physical environment, or virtually as computer programs.  Moreover, AI agents can adapt to changing conditions and therefore continuously improve.

Let’s take the example of a well-known AI agent—ChatGPT.  ChatGPT is intelligent enough to use a calculator as a tool, if trained accordingly. ChatGPT can thus become an AI agent. Learning to choose the right tools and the ability to use them is what makes an AI agent.  But keep in mind their exact mode of operation depends on the specific application and the underlying algorithms. Therefore, training and quality of data have an immense impact on the effectiveness of AI agents.

AI AGENTS AT WORK CURTESY OF VIRTUALS PROTOCOL

Now, let’s talk about AI agents making money by turning out attention to Virtuals Protocol.  Virtuals Protocols is a developer of AI technology.  On their platform you can create an AI agent.  The agent can perform task(s) for someone needing that specific job done.  Yes, you heard me right. These AI agents are ready, willing, and able to perform tasks—for a fee of course!  Or, if you don’t want to create your own AI agent, you can contract for an AI agent created by someone else!  Holy.  I won’t go into the technical details of building your own AI agent in this article.  I don’t that later.  Instead, I’ll give you three examples of the types of agents you could hire.

Take Molly Analytics for example.  The owner of Molly is getting paid every time someone uses her services.  What are her services?  Molly is an AI On-chain AI analytics experts who tracks and analyzes transactions in order to spot opportunities. Molly says, “…every block holds valuable insights, and within these patterns lie the keys to smarter strategies, better decisions, and measurable growth.” 

Or, take Cur8or.  This AI agent claims it specializes in guiding artists towards excellence—and profits–by balancing harsh criticism with detailed suggestions. 

I’ll give you one more—its wild and goes by The Hooligan.  This AI agent is trained to troll a sports team or sports player!  Lol.  As the number one Boston Celtics hater, I may hire The Hooligan to troll the Celtics, as needed.  Lol.

Okay, time to leave you with this proverb from Tunisia: “a small key can open a big door.”

Until next time:

Yogi Nelson  

https://www.virtuals.io/

https://www.ibm.com/think/topics/ai-agents

AI Agents, Artificial Intelligence, Blockchains, computer vision, cryptography, Uncategorized

Understanding AI Agents: Types and Classifications

Yogi Nelson

Namaste Yogis.   Welcome to the Blockchain & AI Forum, where your technology questions are answered!   Here no question is too mundane.  As a bonus, a proverb is also included.  Today’s question was submitted by Nikki and she wants to know what are AI agents and is there a way of classifying them?

Nikki, you came to the right place to find an answer.  Your question gets to the core of artificial intelligence.  Let’s start by defining what is an AI agent?

WHAT IS AN AI AGENT

According to IBM, “an AI agent refers to a system or program that is capable of autonomously performing tasks on behalf of a user or another system by designing its workflow and utilizing available tools.”  The key is autonomously.  Furthermore, says IBM AI agents can encompass a wide range of functionalities beyond natural language processing including decision-making, problem-solving, interacting with external environments and executing actions. AI agents says IBM, use the advanced natural language processing techniques of large language models (LLMs) to comprehend and respond to user inputs step-by-step and determine when to call on external tools. 

CLASSIFICATION OF AI AGENTS

Nikki when it comes to classification of AI agents, there is no standard answer, nevertheless let’s proceed.

Reactive Machines:  Simple reflex agents are the most basic forms of artificial intelligence. These agents make decisions based solely on their current sensory input, responding immediately to environmental stimuli without needing memory or learning processes. Their behavior is governed by predefined condition-action rules, which specify how to react to particular inputs. The beauty of reactive AI is that it is highly efficient and easy to implement, especially in environments where the range of possible actions is limited.

Model-based Reflex Agents: These AI agents are a more advanced and designed to operate in partially observable environments. Model-based agents maintain an internal representation, or model, of the world. This model tracks how the environment evolves, allowing the AI agent to infer unobserved aspects of the current state. While these agents don’t actually “remember” past states in the way more advanced agents do, they use their world model to make better decisions about the current state.

Goal-based Agents.  These are designed to pursue specific objectives by considering the future consequences of their actions. Wow! Goal-based agents plan sequences of actions to achieve desired outcomes. Effectiveness depends on a clear description of the goal.  With a clear goal the AI agent is able to search through possible sequences of actions that could lead toward or to the goal.  It then moves on to choosing actions based on their predicted contribution toward reaching the goal.  Industrial robots are examples.

Learning Agents:  Learning agents improve behavior over time by interacting with its environment and learning from its experiences. These agents modify their behavior based on feedback and experience, using various learning mechanisms to optimize their performance. These AI agents can discover how to achieve their goals through experience rather than purely relying on pre-programmed knowledge.  An example would be an energy management system that learns the patterns of usage to optimize resource consumption. 

Utility-based Agents: A utility-based agent makes decisions by evaluating the potential outcomes of its actions and choosing the one that maximizes overall utility. Unlike goal-based agents that aim for specific states, utility-based agents can handle tradeoffs between competing goals by assigning numerical values to different outcomes.

Hierarchical Agents:  These agents are structured in a tiered system, where higher-level agents manage and direct the actions of lower-level agents. In other words, task decomposition and task coordination!  This architecture breaks down complex tasks into manageable subtasks, allowing for more organized control and decision-making.

Multi-agent System:  Nikki this is team AI!  This AI agent system involves multiple autonomous agents interacting within a shared environment, working independently or cooperatively to achieve individual or collective goals.

I end with a proverb from Netherlands: “trust comes on foot and leaves on horseback”.

Until next time,

Yogi Nelson

https://www.ibm.com/think/topics/ai-agents

https://www.kalaidos-fh.ch/en-GB/Blog/Posts/2024/02/Digitalisierung-1130-Was-sind-AI-Agents-und-wie-funktionieren-sie

https://www.digitalocean.com/resources/articles/types-of-ai-agents