Artificial intelligence (AI), machine learning, deep learning, neural networks

There has never been any other time in life when so many aspects of our world are focused on advancing artificial intelligence (AI) and machine learning as today.

Take a moment and think about this:

Even the Arms Race was mostly a competition among global superpowers. Man going to the Moon was primarily led by NASA. The invention of electricity involved a finite set of people moving it forward. The development and infusion of AI is more pervasive in all aspects of our culture than anything else I can find.

The military is racing to develop AI for military intelligence, for protection and to minimize troop casualties. All aspects of major marketing are racing to anticipate what customers want to buy next — even before they are aware of it. Companies are using it as a way to stay competitive, seeing these issues as a means to an end for saving money and reducing labor costs. Investors are racing to learn how to invest in the right AI companies and projects.

Yet another example: the self-driving car. I don’t know about you but self-driving cars have got to be safer than the driver on the road texting at 70mph. In addition, in the world of social media, we see Facebook using AI in many ways, equipped to pick up on what consumers want more of and delivering it via ads at the right place and time.

In business, cutting edge sales professionals are hungry to gain greater marketing insight using machine learning products like Crystal Knows or the Social CRM, Nimble. Self-learning bots are the first line of support and are helping human support staff. Amazon’s Alexa is the voice service that customers interact with to get information, products, shopping lists, timers, and control lights or thermostats in our home. All the major tech companies are competing to be our best “digital assistant,” including Amazon Alexa, Apple Siri, Google Assistant and Cortana (see my post “The Race to Be Our Best Digital Assistant” back in 2015).

Cities today are using machine learning systems to anticipate crime before it happens. Most of us welcome progress in AI in actually identifying terrorists before they act. Courts are using AI to sentence criminals. On the other side, in the criminal world, we will be seeing more AI resources used to steal information and money.

Let’s not forget Google’s focus on creating a general purpose AI system with their purchase of DeepMind. Google DeepMind grabbed the news with its success in beating the world champions of GO. IBM has bet their future on AI with solution-based Watson, leasing out access time a little like the mainframes of past. Watson gained its media fame by winning Jeopardy in 2011. Watson can even be found in the Research-Triangle NC area providing resources to help patients with cancer at the University of North Carolina Lineberger Comprehensive Cancer Center.

Even these references just scratch the surface of all the ways AI and machine learning are integrating into our lives.

I believe most of us are too busy to pause a moment and think about all of the ramifications. But it is important because it is affecting more and more of our culture each day, slowly infusing into our daily life.

What is Artificial Intelligence (AI)?

It is about creating machines and programs that can learn from mistakes, from their environment, other AI systems and from people, then take actions that affect an environment.

Machine Learning involves programs that can improve at a task with experience. Machine Learning has revitalized the field of Ai, considered a core aspect of AI today. David Amerland wrote a great article, “A quick introduction to machine learning”.

What caused the resurgence of AI?

We have been talking about the future of smarter computers and robots to make our lives easier for years. But suddenly it seems all you hear about are these issues today. What happened? What brought us out of the AI “winter”.  The breakthrough came via algorithms that could process large amounts of unstructured data and powerful computing that could make these massive calculations possible. During the AI winter we had algorithms that did not get smarter with more data over a certain threshold. So having big data did not get us better results. We did not have such massive data and we did not have such massive cloud-based computing power.

What is AI today good at?

AI today is good at modeling something that a human can do and do it as well or better. But today’s AI is not as good at doing something that we as humans don’t know how to do. Like world domination or world peace or sending a human to Mars and bringing them safely back. The problem is not us waking up tomorrow to find the Terminator outside our door but waking up tomorrow to find less of us are needed on a job because of AI.

The real risk at stake is the loss of jobs. Computers can share data and work together more efficiently than people with issues and cultural differences. For example, in New York City, legal offices have been asking their aides to feed data into machines — which will ultimately replace their jobs. This clearly will lead to fewer legal aide jobs being required. As AI replaces many of the tasks legal aids have done, current legal aids need to ask what are the next best uses of their skills in other areas.

What is the role of humans in the process today and throughout the world?

Clearly, if a company can get a less expensive AI system to do the job of a human and get equal or greater results from, most will choose this path. AI is creating new job opportunities for those that are paying attention. One of the areas where humans are needed is to aid in the development of artificial intelligence and to make it smarter and more effective. For example, if a chatbot can get all your basic questions answered, why would you want to wait on a phone for human support? The chatbot is not just text; it may be a visual avatar that understands natural language. If it is good and keeps learning from all mistakes, it will get better.

With the possibility of the reduction of jobs, some are exploring the idea of “Guaranteed Income” where each person would receive a base income to live on. What will we do with lots of leisure time? Will we pursue our greatest dreams or our greatest addictions? I think most of us need a sense of purpose. Would this give our lives more purpose or less? So many of us, including me, get a real sense of purpose from our work. At the same time, if I did not have to focus on work for survival, I have a lot of projects I would like to spend more time in, including research.

I think we are underestimating how collectively as a species we will accept a virtual person that gives us a sense of compassion and connection, providing us with what we want efficiently without conflict. For people with challenging relationships in their own life, an AI system may be a good solution for some conversational companionship. Maybe AI will inspire us to be better humans by modeling higher ethics and good behavior. I believe we need to be open to possibilities but grounded in reality. The role of humans will be changing as AI evolves and improves. It is important that we are aware of how AI is affecting our lives, make the best of it and learn new AI tools that enhance our productivity. Ask the question of what will be the next opportunity in our field with AI that we need to start training for.  Don’t assume that AI could not replace a job in time. We all need to be part of the discussion of what is the role of humans as AI keeps evolving.

AI and Compassion

Most of us agree that as AI evolves we want it to demonstrate compassion, ethics and morality. Are humans the best model for creating an ethical AI system? Even the best of parenting does not guarantee their child will have compassion, the right level of assertiveness and good judgment. I want something much better and more ethical than the “average person”.  

What should WE do?

For investors: Get expert advice to evaluate emerging technology. Use trusted AI consultants to understand and verify the outcomes of the possible investment.

For employees: Pay attention to the advancement of AI in your own field. Instead of trying to stop it to secure your job, ask the question: How can I learn to use it? Also ask what will be the next skills needed by humans as AI advances in your field, and build the skills aligned with your own aptitude. Here is an article about AI helping with job search, that my friend John O’Connor contributed to: How Google’s AI-Powered Job Search Will Impact Companies And Job Seekers

For students: Make sure you are learning how to understand the concepts related to your field. Study them and write an extended post on LinkedIn showing your understanding to demonstrate how you can stay competitive for future employers.

For consumers: Start using new digital assistance and similar tools so you are aware of the changes. You may get an Amazon Alexa system, Google Home or just practice talking to Siri for things you used to type in and see how it is evolving. Unfortunately, criminals will be using more AI tools to break into things. At the minimum, keep your computer and smartphone updated, and learn about two-factor authentication to protect your passwords. Watch out for smarter systems trying to act like real people getting us to do something, not in our own best interest. I am already getting smart sales calls that interact and respond as though they were real people. Make sure your home Wi-fi does not have a password like your street address or dog’s name. Don’t just shut everything out, use good judgment!

For sales professionals: make sure you are keeping up with all of the new tools assisting with the complex sales process today. Listen to our “Social Selling Podcast” as well. We have interviewed some of the frontier experts and will interview more! ( ).

I don’t think we are going to become house pets of sentient AI anytime soon, but I do think it is healthy to explore how we set the foundation of future ethical and compassionate AI systems. What we need to do as humans is to ask today when our job is replaced by AI how are we going to stay competitive and relevant. We also need to explore how we can maximize the utilization of AI in our current field and keep up with the progress being made in the field.

Two of our Social Selling Podcast on this topic:

Security Considerations related to AI

We need to use AI to identify and prevent “social engineering” to gain a point of entry into a target system.  Social engineering involves tricking a user into doing something that gives malicious code access into a protected network making them believe it was the “right thing to do”. As AI evolves the capability to automate social interactions, it will enable malicious actors to automate the social engineering aspect of their attacks, which is currently one of the hardest parts. At the same time we need to keep improving AI to protect us from such attacks. “Think about it – if an AI can handle customer support calls and other interactions with humans, it can certainly be programmed to trick some of them into downloading malicious code or providing security credentials.” – Randy Earl

What is needed to take AI to the next level?

  • New breakthroughs in how AI can learn more efficiently and make future predictions better. The frontier of AI taking on tasks that no human has done before will be another important area.
  • More optimized hardware for AI. Google announced this year they are working to create more hardware optimized for AI with their Tensor Processing Units and Google Compute Engine. 
  • Another important frontier to watch is AI used to advance AI, Google also announced projects in this area this year.
  • Providing more fuller conversational digital interaction with a sense of emotional responsiveness, where we can carry on more of a connected back and forth dialogue, instead of just “Alexa, turn on the lights”. To quote David Amerland, “We need to crack the emotional level next (think of Commander Data’s emotion chip)”.
  • Of course, advancing AI to identify security threats and suggest solutions faster. Including security for the internet of things and self-driving cars so they can not be hacked.

This is an exciting time to have a front row seat to the development of AI. It’s like watching the wild west as it was happening all around us. If today AI is no smarter than an amoeba, just think what will be possible as AI evolves!

I want to hear your comments below of other areas related to AI to consider, as well as what you think we need to do as individuals in the face of the rapid integration of AI in our society.

A few terms you need to understand 

AI – Artificial Intelligence – “is intelligence exhibited by machines. In computer science, the field of AI research defines itself as the study of “intelligent agents”: any device that perceives its environment and takes actions that maximize its chance of success at some goal. Colloquially, the term “artificial intelligence” is applied when a machine mimics “cognitive” functions that humans associate with other human minds, such as “learning” and “problem-solving”.” – Source

Neural Nets / Artificial neural networks (ANNs” “ are computing systems inspired by the biological neural networks that constitute animal brains. Such systems learn (progressively improve performance) to do tasks by considering examples, generally without task-specific programming. For example, in image recognition, they might learn to identify images that contain cats by analyzing example images that have been manually labeled as “cat” or “no cat” and using the analytic results to identify cats in other images. They have found most useful in applications difficult to express in a traditional computer algorithm using rule-based programming.” Source:

Machine Learning – “is a method of data analysis that automates analytical model building. Using algorithms that iteratively learn from data, machine learning allows computers to find hidden insights without being explicitly programmed where to look.” Source

Under Machine Learning we have:

  • Unsupervised Learning – The system is finding patterns without assistance by humans. “Unsupervised Learning is a type of machine learning algorithm used to draw inferences from datasets consisting of input data without labeled responses. The most common unsupervised learning method is cluster analysis, which is used for exploratory data analysis to find hidden patterns or grouping in data.” – Source
  • Supervised Learning – Humans are helping with the training and adjusting of the machine learning. An example is a SPAM filter where you help “educate” it on what is spam then it makes inferences from there. “Supervised learning is the machine learning task of inferring a function from labeled training data.[1] The training data consist of a set of training examples. In supervised learning, each example is a pair consisting of an input object (typically a vector) and a desired output value (also called the supervisory signal). A supervised learning algorithm analyzes the training data and produces an inferred function, which can be used for mapping new examples.” Source
  • Re-enforced Learning – The machine learning is working with other machine learning algorithms to enhance itself. The best example of this was how Google’s Deepmind learn GO by having the machine learning program play against itself over and over. “Reinforcement Learning is a type of Machine Learning that allows machines and software agents to automatically determine the ideal behavior within a specific context, in order to maximize its performance.” Source: 

Deep learning / deep structured learning: “is the application to learning tasks of artificial neural networks (ANNs) that contain more than one hidden layer. Deep learning is part of a broader family of machine learning methods based on learning data representations, as opposed to task specific algorithms. Learning can be supervised, partially supervised or unsupervised.” Source:

Cognitive computing “is the simulation of human thought processes in a computerized model. Cognitive computing involves self-learning systems that use data mining, pattern recognition, and natural language processing to mimic the way the human brain works.” -Source:

IBM Watson – “Watson is a question-answering computer system capable of answering questions posed in natural language, developed in IBM’s DeepQA project by a research team led by principal investigator David Ferrucci. Watson was named after IBM’s first CEO, industrialist Thomas J. Watson.” – Source

The key thing to understand it IBM Watson is working on solving specific task while Google’s Deepmind team is working toward creating a more “general AI” system.

Google’s TensorFlow – An open source (free) machine learning library for research and production. It is being offered to encourage developers all over the world to advance the field of AI and Machine Learning. “TensorFlow is a Python library (Phyton is a programing language) that allows users to express arbitrary computation as a graph of data flows.” Source: ,

Chatbots. ChatBots and talkbot: What is a chatbot or talkbot? Wikipedia defines it as: “A chatbot (also known as a talkbot, chatterbot, Bot, chatterbox, Artificial Conversational Entity) is a computer which conducts a conversation via auditory or textual methods.”

We are going to see more of them in a lot of “customer service” roles due to the progress of machine learning. Often how they work is the bring in information from the company like a Q&A section or database. Then they have humans oversee it or respond to responses that are not correct and the “bot” “learns” from this feedback or just mistakes.  It will help some businesses and professionals but also may take away jobs so I feel it is important to understand them. Then ask both how can we use them to enhance our field and what can we do to stay competitive in the face of AI and “bots”. Here is an interview with a Chatbot developer.