No Limit to What We Can Accomplish with AI
By Sal Gerardo
“A student is not a container you have to fill, but a torch you have to light up.”
— Albert Einstein
Education has challenges, some of which were brought on by the pandemic, and some that were brought on because education has been painfully slow to change and move with the rest of the world and the rest of the world’s industries. These challenges, if we are not careful, will bite us in the backside so hard they will hear us screaming on Mars.
For those of you that do not know me, I am a rarity in the world of education. I served for many years on my school board in New Jersey, and professionally, I am in the computing business. Specifically, in artificial intelligence.
I am a tremendous fan of modern education, and especially, teachers. But increasingly, modern education is getting harder and harder to explain, and teachers are getting the short end of the stick.
Even before COVID, education was headed to an impasse of sorts, and then the pandemic hit. And here we are. Now, I think we can safely say the entire education establishment needs to change. And the only way we are going to get there, with the coming teacher shortage and all our other challenges, is through technology — the most significant of which is AI. In fact, I will go out on a limb and say that in five years, if you are not a practitioner of AI, you probably will not be an educator.
There is one thing teachers do better than anyone else, and better than any technology ever created.
But if we do not remove all the minutia from their professional lives, we number their days of teaching. And that is where AI comes in. Artificial intelligence, which I can tell you from personal knowledge, is getting more and more sophisticated. So sophisticated, in fact, that it can provide all the mechanicals, all the repetitiveness and all the assessments. AI can get the basics out of the way. It can empower teachers.
Teachers have instincts. Teachers have the unique ability to connect with learners. Teachers have the humanity that students need; that they crave. Teachers can provide the emotional warmth and connection.
Artificial Intelligence is getting very smart, very fast
Over the last few years, the artificial intelligence industry has retooled itself with new capabilities. For example, even specific straightforward tasks, like machine translation, require a machine read and write in both languages (NLP), follow the author’s argument (reason), know what is being talked about (knowledge), and faithfully reproduce the author’s original intent (social intelligence).
A problem, like machine translation, is “AI-complete” because all these problems need to be solved simultaneously in order to reach human-level machine performance and intelligence.
Just like a human brain at birth, which has very little knowledge or learning, the machine needs to gain knowledge and learn infrastructure to allow it to execute human-like brain functions and decision processes.
The human brain can store and learn the meaning of information at the same time. The storage process and the learning process are tightly connected, but limited to human brain capacity, and limited to a set number of sensory and visual inputs at limited input rates.
The computer brain, on the other hand, has one limitation. It cannot store and learn the meaning of information at the same time. That said, our tech uses a three-part process to implement information storage and the learning process to gain the meaning of information simultaneously:
- The Knowledge Library
- The Intelligent Grammar
- The Learning Process
This gives you unlimited knowledge and learning capability, and unlimited input and output capability on a global scale.
AI soon will use three distinct technologies:
- Learning and meaning engine (LME)
- Cognitive structures self-organize human knowledge into Knowledge Libraries.
- LME learns from conversations, documents, websites, and structured databases.
- LME learns concepts, sentences, stories, and expert behaviors with machine learning, structured induction, and metaphor.
- LME learns language discourse with selected restrictions and stories.
- Artificial Intelligence
- AI performs data engineering and programming.
- AI is goal-directed behavior.
- AI changes behaviors dynamically to achieve goals.
- AI uses continuous assessment of success against goals.
- AI has an extendable Knowledge Library of expert behaviors, including data engineering, data analytics, data analytics tutoring, and STEM tutoring.
- AI allows behaviors to generate software directly from the Knowledge Library.
- Natural Language Avatar
- Uses Natural Language dialog, including Socratic dialog for tutoring.
- Uses Natural Language Q&A.
- Learns a model of human interlocutors / students through discourse.
- Continuously assesses discourse, emotional sentiment, and adjusts to increase bonding.
Socratic tutoring is asking all the right questions
Socratic tutoring requires conversation and a procedural process (how-to) model of Socratic tutoring and reasoning. It uses three skills; each a complex intelligent behavior performed together to achieve a single goal: student mastery.
According to our research, there are 14 behavioral skills necessary for Socratic teaching, and they include:
- Speaking and listening to speech, including translation, as necessary: teacher and student utterances.
- Reading and writing text, including translation, as necessary: teacher and student produced materials.
- Question and answer linguistic forms, i.e., legal answers for given question types.
- Learning and remembering knowledge to be taught.
- Learning and remembering knowledge about individual students.
- Understanding the meaning of student utterances (including inference).
- Planning teacher utterances.
- Natural Language Dialog.
- Learning and remembering (and self-organizing) SME (Subject Matter Expertise) knowledge about the Socratic teaching process.
- Goal-directed behavior, in this case the Socratic teaching process.
- Reasoning about student responses to questions.
- Assessing student mastery.
- Self-referential modification of SME Socratic teaching process knowledge, changing the process to achieve student mastery.
Extensive research shows students achieve significantly better learning outcomes in a one-to-one tutoring environment, where they can learn at their own pace, in their own style, and where their questions guide what they are taught.
The soon to come Artificial Intelligence, will engage students in a natural-language Socratic conversation, continuously monitoring student comprehension. With infinite patience, students are taught at their own pace and level of understanding. If a concept is not understood, the AI tutor ensures that real comprehension is achieved before moving on to a new topic. Real comprehension is gauged through dialog and open-ended questions and answers. This approach is much more effective than multiple-choice tests that can be repeatedly taken until the student chooses the right answer through the process of elimination, potentially learning very little, but passing the test.
Currently under development, there is an AI solution that contains a natural language meaning engine. It simulates the way people think and understand the meanings and intent of words, sentences, and stories — unlike other computerized educational solutions, search engines, or Q&A systems such as Chatbots, with scripted answers to set questions. This developing AI technology will form the foundation for the first truly intelligent, conversational, human dialogue interface.
It can generate its own dialogs, without scripts, by using a Knowledge Library, meaning that as the Knowledge Library grows, it can engage in different conversations in different domains. Its differentiator is its natural language student interaction, which other computer-based training does not yet offer.
Soon to come, Artificial Intelligence will allow teachers to spend more time on students’ personal and human relational issues, and will have a dual function:
- The ability to provide a Socratic tutor for every child, under the direction of the teacher.
- Using the Socratic Dialog Tutorial.
- Delivering personalized content.
- Inventories Student Ability: Learning Style and Learning Level.
- Perceives and responds to a student’s emotions.
- Answers questions in context.
- Provides remedial and/or enrichment tutoring.
- Will provide an assist for every teacher to use as needed.
- Collaborates with teachers.
- Learns teacher preferences.
- Automatically generates tutorial course materials from the Web, teacher materials, and textbooks.
- Assesses students in real-time.
- Alerts teacher to learning barriers.
Key differentiators in present “AI” tutors and the “what’s next” tutors
- More STEM (Science, Technology, Engineering, Math) information has been created in the last five years than in all human history.
- Increasing un-met STEM educational needs because of information acceleration.
- STEM teaching for students and STEM training for teachers.
- Creation of course materials for STEM.
- AI Tutor Program Key Differentiators.
- Natural Language Socratic conversations tutor students and teachers.
- Real-time teacher monitoring/assessment of individual student learning.
- Automatic creation of course materials tutorials from the web, teacher materials, and textbooks.
- Highly scalable Artificial Intelligence technology.
It will not be long before AI tutors will have unscripted conversations. The AI and the student will solve problems together. The tutor will work at the student’s pace, using Socratic dialog to stimulate critical thinking. It will be a tremendous help for teachers, as the AI tutor can collaborate with them, identifying learning problems and continuously assessing student mastery without tests.
Just imagine, going from a world where teachers are asked to do everything, INCLUDING spending 10–20 hours per week just searching and putting together lessons, plus the additional 40 hours they are asked to work, plus now being asked to make up for the estimated one million teacher shortage beginning this year. It is no wonder we are losing teachers by the bucket load. But help is coming.
Teachers are talented, amazingly dedicated individuals. And it is important they understand technology is not coming for their jobs. Technology is coming to set them free. Proper logistics can take us a long way, but AI is coming to do the heavy lifting. This may be the most difficult time ever to be an educator. Artificial Intelligence will remove the minutia, the repetitiveness, and provide the assessment. As AI does the heavy lifting, teachers, the professionals in the room, will do the heavy thinking. With the help of AI, teachers will have the tools they need to connect with students on a very personal level. They will be the torch that fires up a generation. And when they do, look out world. There is no limit to what we can accomplish.
About the Author
Sal Gerardo is chairman of Global Education Media.
This article was first published on (ET) Magazine.