"AI training should start in kindergarten and continue until retirement age"

"The State of Israel has been significantly lagging behind OECD countries since the digital age in laying infrastructure and preparing for the AI revolution. We are behind in building infrastructure, databases, data centers, regulation, laws, and training," said Dadi Perlmutter, chairman of the High-Tech Human Capital Committee, on a panel with Gali Shahar Efrat, GM of the Google and Reichman Tech School at Reichman University, moderated by Calcalist’s Hagai Gilboa, held as part of Google and Calcalist’s Startup Week.

According to him, the most important issue is that training must start from kindergarten and continue through to seniors. The greatest challenge in terms of infrastructure and resources, he added, lies in training older adults who will have to enter a completely new field.

"It's a three-fold shared responsibility. In Israel, people know not to wait for the government. You have to move forward on your own. But the government still has a crucial role. The education system, from early childhood through to university, is mostly under government responsibility. The institutions that define educational policy and the National Education Council are government bodies. Infrastructure, communications, digitization, and data, all of these are government-managed areas. Regulation, legislation, and oversight are all government responsibilities. And all of this must move fast.

"But companies also carry responsibility. They must decide to engage with AI, bring in experts, and—most importantly—define a strategy. It’s not just about learning ChatGPT or Gemini. Companies need to know what they want to achieve, how they plan to do it, and how to train their employees accordingly. For example, Amazon recently announced it would invest $700 million in training its U.S. workforce in AI-related skills.

"And finally, every citizen must take responsibility for themselves. When I worked at Intel, the principle was: the responsibility is yours, we’ll support you."

When should this training begin? At the point of entering the workforce? Or even earlier, in kindergarten?

"You have to start in kindergarten. One of the most important skills in using artificial intelligence is the ability to ask questions. And by the way, you don’t need to teach children how to ask questions, they’re naturally good at it. I have four children and eight grandchildren, and they’re expert question-askers. The key is not to turn them from question-askers into answer-providers, but rather to teach them how to ask structured questions, how to ask the next question based on the answer they receive, how to think creatively, and how to build knowledge foundations. These are abilities and skills that must be nurtured over time.

"Everything we talked about - creativity, critical thinking, task-based thinking, problem solving - these are now supported by a powerful tool. We no longer need ten years of rote learning to overcome complex challenges. In the age of AI, the most important thing is to know how to ask the right question and assess whether the answer you received is appropriate."

What will training such a broad population in AI look like? Will private companies take the lead? Universities?

"First of all, there will be private training programs, and many people will self-learn—there are already courses at Stanford, Coursera, and many other places. But the institutions best equipped to provide informed instruction are academic ones. They have the right tools, the teaching staff, and the ability to structure a curriculum. There must be foundational courses followed by professional training tailored to different fields, doctors, financiers, engineers, each requiring a unique approach.

"And training shouldn’t stop at the degree level. I believe academia should be the foundation, but there’s no reason why private institutions can’t participate. The government, as it has done in the past, should monitor quality, evaluate performance, and even provide funding to effective training providers."

Gali, you are involved in training hundreds of high-tech students at Reichman University. Tell us about your personal path.

"I studied economics for my bachelor’s degree, it was nice, but then the big tech boom happened, and I realized I needed to retrain for high-tech. I enrolled in a bootcamp that lasted several intense months and started working in tech as a programmer. One thing I didn’t do was continue studying the field further, as I moved into business roles. But there’s no question that academia is changing - and so are we. The big question is always: do I need a degree, or do I need professional training? The simplest answer is both. And you should never stop learning.

"The pace of change is dizzying. If you haven’t learned something new in three years, you’re already behind and might find it difficult to keep up. I lived this myself, which is why I believe deeply in this path. People can upskill and reskill at any stage of their lives. A degree can be a foundation, but you can always add more, especially now that AI has created new opportunities in the job market."

You, Gali, at Reichman University, have taken on a very ambitious goal: to train thousands of underrepresented individuals - ultra-Orthodox, Arabs, women. Tell us a bit about the program and what we can learn from it in terms of large-scale AI training.

"Two and a half years ago, we set a very ambitious target: to train 3,000 women and men from diverse, underrepresented communities for careers in high-tech. And of course, you know what’s happened over the past two and a half years,since the war began, it’s been an incredibly challenging time. Like any good startup, we adapted to the changing geopolitical reality in Israel.

"We’ve become a school that is heavily focused on artificial intelligence. Today, every course we teach is rooted in AI, whether it’s introductory training in generative AI or subject-specific programs. We’ve also developed advanced programs for bachelor’s and master’s graduates who want to specialize in AI. These are our foundations.

"These people are not disadvantaged. They are talented individuals, we simply give them a chance. And ultimately, we measure success through job placements. Today, 80% of our graduates are employed in high-tech positions."

The state may need to train up to three million people, many of whom may not be eager to do so, in a very short time. Is that even a realistic goal?

"It has to be. It must be a goal the state is committed to. Every individual needs to ask themselves: how will artificial intelligence impact my profession, my organization, and my team? We must train more people across more fields. High-tech is no longer a niche sector. AI has broken through all boundaries and is affecting every sector and every area of activity.

"We know that by 2030, 20% of jobs and professions will change. We need to start preparing for that now."

anaged areas. Regulation, legislation, and oversight are all government responsibilities. And all of this must move fast.

"But companies also carry responsibility. They must decide to engage with AI, bring in experts, and—most importantly—define a strategy. It’s not just about learning ChatGPT or Gemini. Companies need to know what they want to achieve, how they plan to do it, and how to train their employees accordingly. For example, Amazon recently announced it would invest $700 million in training its U.S. workforce in AI-related skills.

"And finally, every citizen must take responsibility for themselves. When I worked at Intel, the principle was: the responsibility is yours, we’ll support you."

When should this training begin? At the point of entering the workforce? Or even earlier, in kindergarten?

"You have to start in kindergarten. One of the most important skills in using artificial intelligence is the ability to ask questions. And by the way, you don’t need to teach children how to ask questions, they’re naturally good at it. I have four children and eight grandchildren, and they’re expert question-askers. The key is not to turn them from question-askers into answer-providers, but rather to teach them how to ask structured questions, how to ask the next question based on the answer they receive, how to think creatively, and how to build knowledge foundations. These are abilities and skills that must be nurtured over time.

"Everything we talked about - creativity, critical thinking, task-based thinking, problem solving - these are now supported by a powerful tool. We no longer need ten years of rote learning to overcome complex challenges. In the age of AI, the most important thing is to know how to ask the right question and assess whether the answer you received is appropriate."

What will training such a broad population in AI look like? Will private companies take the lead? Universities?

"First of all, there will be private training programs, and many people will self-learn—there are already courses at Stanford, Coursera, and many other places. But the institutions best equipped to provide informed instruction are academic ones. They have the right tools, the teaching staff, and the ability to structure a curriculum. There must be foundational courses followed by professional training tailored to different fields, doctors, financiers, engineers, each requiring a unique approach.

"And training shouldn’t stop at the degree level. I believe academia should be the foundation, but there’s no reason why private institutions can’t participate. The government, as it has done in the past, should monitor quality, evaluate performance, and even provide funding to effective training providers."

Gali, you are involved in training hundreds of high-tech students at Reichman University. Tell us about your personal path.

"I studied economics for my bachelor’s degree, it was nice, but then the big tech boom happened, and I realized I needed to retrain for high-tech. I enrolled in a bootcamp that lasted several intense months and started working in tech as a programmer. One thing I didn’t do was continue studying the field further, as I moved into business roles. But there’s no question that academia is changing - and so are we. The big question is always: do I need a degree, or do I need professional training? The simplest answer is both. And you should never stop learning.

"The pace of change is dizzying. If you haven’t learned something new in three years, you’re already behind and might find it difficult to keep up. I lived this myself, which is why I believe deeply in this path. People can upskill and reskill at any stage of their lives. A degree can be a foundation, but you can always add more, especially now that AI has created new opportunities in the job market."

You, Gali, at Reichman University, have taken on a very ambitious goal: to train thousands of underrepresented individuals - ultra-Orthodox, Arabs, women. Tell us a bit about the program and what we can learn from it in terms of large-scale AI training.

"Two and a half years ago, we set a very ambitious target: to train 3,000 women and men from diverse, underrepresented communities for careers in high-tech. And of course, you know what’s happened over the past two and a half years,since the war began, it’s been an incredibly challenging time. Like any good startup, we adapted to the changing geopolitical reality in Israel.

"We’ve become a school that is heavily focused on artificial intelligence. Today, every course we teach is rooted in AI, whether it’s introductory training in generative AI or subject-specific programs. We’ve also developed advanced programs for bachelor’s and master’s graduates who want to specialize in AI. These are our foundations.

"These people are not disadvantaged. They are talented individuals, we simply give them a chance. And ultimately, we measure success through job placements. Today, 80% of our graduates are employed in high-tech positions."

The state may need to train up to three million people, many of whom may not be eager to do so, in a very short time. Is that even a realistic goal?

"It has to be. It must be a goal the state is committed to. Every individual needs to ask themselves: how will artificial intelligence impact my profession, my organization, and my team? We must train more people across more fields. High-tech is no longer a niche sector. AI has broken through all boundaries and is affecting every sector and every area of activity.

"We know that by 2030, 20% of jobs and professions will change. We need to start preparing for that now."