Employment immunity: the quiet revolution of robotics and physical AI
As Startup Nation's traditional software categories face a wave of headwinds and layoffs, robotics and physical AI are emerging as one of the defining industries of the next decade, gearing up for an imminent boom of their own.
“If I were to bet on something in the future, it would be physical AI and robotics,” says Sapir Hadad, Executive Director of Fiverr’s AI Lab and Physical AI, and Co-Founder and CEO of Gen AI and automation solutions startup D-0. “This will be the most important thing in the next two decades at least.”
Amid a torrent of layoffs plaguing Startup Nation, one sector appears insulated from the headwinds hitting traditional software: robotics and physical AI, which has attracted $19 billion in capital this year alone.
The robotics revolution, which Hadad asserts will invariably “change the world,” is already materializing in the world around us, and well outside of expected industrial locations like factory floors and warehouses. In March, a free-roaming, animatronic Olaf debuted at Disneyland Paris, stunning social media users and tech enthusiasts alike with how realistically the beloved Frozen character was brought to life. More recently, in July, the US-based Sunday Robotics previewed the company’s ACT-2, a foundational AI model that achieved a 99% success rate autonomously folding laundry in unfamiliar home environments.
As Hadad explains, robotics is at an inflection point in which foundational AI models and an increasingly accessible hardware supply chain are making what was once considered “science fiction” viable.
“The ecosystem is being built as we speak,” she says. “There's a lot of things that are happening… It's not just the AI, it's also the hardware.”
Robotics, Hadad notes, "up until around five years ago, was like a research field. No one thought about getting it into deployment. Everyone thought it would take years, and then ChatGPT happened, and then we saw that we can create foundational models for the digital world.”
Hadad recently took to X to comment on the state of the industry. “It’s going through a similar transition to what we’ve seen with AI: from labs to companies, from research to products, and from pilots to commercial deployment.”
“We are still at the very beginning of the wave,” the post continued. “The investments are already here, but the vast majority of commercial deployments are still ahead of us. Those who enter now will be able to be part of the first generation that builds this industry at scale.”
As Hadid tells CTech, "now is the time that robotics is getting its software moment.” “And it's a massive market,” she adds. “It's trillions of dollars.”
Unsurprisingly, the local Israeli robotics and physical AI industry is cementing itself as part of this moment. In the last few months alone, former Israeli Finance Ministry official Shani Mandel-Laufer helped launch LTX, a Lightricks spin-off dedicated to building generative world models for physical AI. Enigma, a startup founded by 26-year-old Jonathan Jacobi, emerged from stealth with a $71 million seed round to build a foundational intelligence layer for robots. And a report by the venture capital firm 10D identified the local ecosystem as consisting of 125 Israeli physical AI startups that have collectively raised over $10 billion.
"I was surprised that we have around 80 companies that are doing physical AI or robotics in Israel,” said Hadad. However, she continues, “it's still relatively low. If we compare Israel to the rest of the world, we're still behind in the major robotics ecosystem.”
The US and China dominate the sector, whereby “the US is leading on AI and capital... and China has an enormous advantage in manufacturing.” Further, while China has historically held the upper hand in hardware and supply chains, Hadad points out that “the US and Germany have a lot of factories,” listing their automotive and heavy industrial manufacturing capacity, an infrastructure that Israel lacks in comparison.
Nonetheless, Hadad believes Israel’s high-tech engine has its own strengths that distinguish it in the robotics and physical AI space. In Israel, she says, “our advantage has always been to build very sophisticated technologies with relatively small teams. We're strong in AI, computer vision, sensing, and complex multidisciplinary systems.”
The country’s unique military environment further serves as an incubator and testing ground for the sector’s defense applications, ultimately expediting its development. "We have a knife on our throat, so we move fast,” Hadad explains. “We build things that need to be deployed and operational for protecting ourselves."
“We do things creatively. We know that our life depends on it; our existence depends on it. This is why I think we managed to bring out from ourselves more creative, more efficient solutions, more readily-deployed solutions.”
Related articles:
While there is undoubtedly a “real need for physical AI,” Hadad explains – a demand felt globally as robotics become urgently necessary in areas like factories, hospitals, and elder care – still, within the robotics field itself, “there is a huge labor shortage.”
Historically, the barrier to entering robotics was exceptionally high. “It was a research field,” says Hadad. “It requires understanding in many disciplines. You need to understand manufacturing and deployment, and you need to understand AI and control.”
Now, however, AI is bridging the technical knowledge gap and lowering the barriers to entry into this once highly elusive field. “I think that now we have the power of AI, we can learn… and become builders of it." says Hadad. “The digital AI transformation can help people that are AI engineers that are interested to move on to this field. It really explains everything.”
Currently, as the high-tech industry is incurring widespread layoffs, particularly across its software and SaaS sectors, some experts view the period as an opportunity for tech workers to recalibrate their career paths. This emerging field presents a unique opening in which, ironically, the same AI revolution driving leaner workforces in other areas of tech could be the very thing that staffs the physical AI sector and helps redistribute talent.
And it’s not just software engineers who are being sought after. As robotics and physical AI moves into commercial environments, the industry needs the same product, design, experiential, and infrastructure roles that built the modern software economy.
“All the rules and job descriptions that were true to digital AI are applicable to physical AI,” Hadad notes. “You just need to learn. It's not like transitioning from one SaaS company to another SaaS company, it's a bit more complex, but still the core skills are needed in both of them."
For the prospective tech professionals considering this path, “think of it like AI ten years ago,” she advises. “Talk with companies. All the robotic companies in Israel, the US, and Germany are eager to answer. You can send them a cold email and they’ll answer you. This field is growing and everyone needs people.”
Overall, “there is a huge opportunity,” Hadad emphasizes. “I really believe that we're at the beginning of everything, and whoever positions themselves in the next two years in this place, it's really an influential moment in this industry.”
One of the most consequential milestones the industry is working towards, says Hadad, is the race to build the “General Robot Brain.” The concept pertains to a hardware-agnostic foundational model that separates a machine's software intelligence from its physical body. By standardizing this intelligence, robots can theoretically share skills and adapt across different physical forms without requiring bespoke programming for every new task. Ideally, the Brain will also allow the sector to scale quickly.
"A lot of researchers are working on it,” she says. “Not only at DeepMind and OpenAI but also at companies like Physical Intelligence, like Enigma, which is an Israeli startup that just got out of stealth. So a lot of good minds are working on it now.”
In her X post, Hadad promised that robotics would be “the industry that will define the next decade.” But what does this transition look like in practice? Naturally, the technology has a number of hurdles to overcome throughout its teething process.
"There will be a lot of translation work between what happens in the lab to how it works in reality," she adds. When it comes to deploying these systems at scale, Hadad predicts, "we will see a lot of complexity."
For one, "there are a lot of manufacturing problems, a lot of bugs and fixes," which need to be handled on location. "It's not like a digital," she explains, describing how physical hardware inherently cannot be fixed remotely. "So the deployment is much harder than digital AI."
Beyond the obvious regulatory hurdles that stand to slow the sector's trajectory around its adoption curve, the industry is set to face significant psychological barriers, specifically concerning consumer trust. Hadad expects that the first round of dominos to fall, so to speak, will be a relatively controlled entry into a new normal: “I think in factories, warehouses, nursing homes, I think it will be normal; we will get used to it.”
However, "at home, I think it's gonna take a bit more time," she believes. "It’s discomfort, it's trust issues. Am I OK with a humanoid working at home when my five-year-old daughter is around? I'm not sure," she admits.
Hadad stresses that "there will be a lot of education" required to normalize "the feeling that it's OK that it lives in their house” while easing privacy and data concerns. Ultimately, integrating physical AI into the home will come down to "trust and feeling that you're willing to let go and to give it freedom... Does it record them? That it saves the data or doesn't save the data? Does it picture me because it has cameras? There are a lot of questions that will need to be answered," she says.
Still, while she acknowledges that "we're not there yet, we have a few years," she remains confident that "it will come together."
