Taking Physical AI to Market: What Founders Need to Know Before The Next Raise

If you're building physical AI, you already know the hard truth: it takes more R&D, more capital and more accurate AI models than software alone. On the Moonshots stage at NRTH 2026 (formerly Elevate Festival) in Toronto, five sessions covered critical minerals, construction, prosthetics, women's health, and space. Different industries, all with a unifying deep tech story. I found the talks useful and want to share more insights of what I know about marketing for deep tech companies.

That harder path in building physical AI is also your true advantage. Proprietary data, domain expertise, and hardware, are what make physical AI companies defensible and hard to replicate, as long as you can make investors and customers see it.

At Unicorn Rocket Marketing, we work with deep tech founders at the stage where they need to raise another round to continue working on research and development, to build out their more complex product roadmap, and find product-market-fit. The capital is flowing, and Canada has deep tech founders ready to lead.

What matters at this stage for deep tech is execution, and telling their story well. Investors, customers, and talent, all need to understand why a given founding team is the best one to back.

I've worked with some of the best teams in the deep tech industry in Canada. In 2024, I led marketing at Augmenta AI, one of the companies featured at NRTH, and in this blog post. Before that, from 2020 to 2023, I was VP of Marketing at DarwinAI, a Waterloo company building hardware and software for AI in manufacturing. DarwinAI used computer vision to detect defects in printed circuit boards. The team developed the world's largest dataset of circuit board defects and deep learning models for that very specific use case, and Apple acquired DarwinAI in 2024.

As an operator, leading marketing and communications, and being an integral part of the GTM engine, I've seen how VCs evaluate physical AI companies. I've also seen what go-to-market teams have to do to move them forward: proving commercial traction, showing a path to scale, and getting to the next round or an exit. That's why I wrote this post for founders typically raising at seed and Series A: to share what I've learned about building credibility with investors and customers in this space.

With that in mind, we start with the specifics explored at NRTH: the five Moonshots sessions and the companies behind them. From there, we look at how these companies turn impact, IP, and team, into a winning, fundable, story. Next, we look at today's funding trends, which validate much of what we heard on stage. We close with a practical playbook for deep tech founders raising from VC firms.

Who we heard from on the Moonshots stage

Clean tech: Destiny Copper and Carbonyx

The opening panel set the tone. Build a strong team, and stop treating sustainability as a liability. The speakers reminded the room how much material it takes to build the world we live in, from iPhones to air travel. Critical minerals are the foundation of the whole tech economy.

Founder takeaway: Don't pitch sustainability as a cost. Pitch it as supply chain resilience, a language both procurement teams and investors understand.

AI for the built world: Augmenta.ai

Augmenta.ai showed how spatial AI is changing the construction industry, with a focus on electrical design for large-scale commercial projects. Much building design is still drafted by hand in tools like AutoCAD, which slows everything down. The problem isn't the VDC (Virtual Design and Construction) managers in MEP (mechanical, engineering, plumbing), but rather it's the gap that exists between design, procurement, and manufacturing.

Their case studies scaled up in complexity. From Mount Hope Elementary School to a hospital, with piping, cooling, ventilation, medical vacuums, and constant last-minute changes, to a complex hyperscale data centre, where their system laid out 56 miles of electrical raceway in 80 hours. That meant fewer mistakes, less rework and major savings in time and labour. Next on their roadmap is the rest of mechanical, electrical and plumbing (MEP) design, and an agentic future for construction.

Founder takeaway: Sequence your case studies from simple to complex. Each one proves you can handle the next level, and together they form a ready-made narrative for your next round.

Physical AI for accessibility: SmartARM

SmartARM builds AI-powered prosthetic bionic hands for amputees. Their story was the most human of the day. A man in Florida needed a SmartARM when the company was out of funding. He ended up leading their next round, got two arms of his own, and helped them start selling in Canada.

Users can put the robotic arm on and have it working within minutes. It learns with use. A companion app lets people submit photos, and SmartARM pushes updates so every SmartARM in the field gets better. Affordability is built in, with government funding programs keeping out-of-pocket costs lower.

Founder takeaway: Your most committed customers can become your best investors and advocates. And a product that visibly improves with use is a data flywheel investors can see.

Femtech: Fibra, Science & Humans, Mystoria

Canada recently saw the announcement of Bill S-243: National Framework for Women's Health in Canada Act, in December 2025, to strengthen investment and deploy targeted solutions to improve access to healthcare for women. This NRTH panel featured founders who are taking charge today in women's health care innovation.

• Fibra makes smart underwear with fabric sensors for non-invasive tracking. Fibra goes beyond cycle tracking, with AI-generated insights during major life changes. At pre-seed, Fibra spends real time educating investors.
• Science & Humans is one of Canada's large hormone health platforms, covering all genders and life stages. Their at-home test takes about six minutes, samples are picked up by Uber, and results link straight to telehealth and next-day specialists. Science & Humans bootstrapped before raising intentionally for scale.
• Mystoria grew out of its founder's own fertility journey. It puts your full health history in your hands and helps you share it with your care team, so patients no longer have to act as the general contractor of their own care. Mystoria's founder self-funded from a prior exit, then leaned on her community.

Founder takeaway: Lived experience builds founder-market fit. But in an underfunded category, expect to educate investors on the size of the market before they'll engage with the product. While women's health is a large market, it still has to fight harder for capital.

Space: Kepler Communications, NordSpace, Canada Rocket Company

The closing panel asked a big question: can Canada do the hard parts of space on its own? No Canadian-built rocket launcher has yet put a payload into orbit from Canada. Other nations have launched Canadian hardware for us, and made national decisions for us along the way.

That is changing. Space is now a defence domain as well as an economic one. Kepler is pioneering optical communications in orbit, building a mesh network that moves data around space and back to earth. Its uses range from detecting wildfires early, to monitoring Arctic shipping lanes, and illegal fishing. NordSpace and Canada Rocket Company are building sovereign launch capability, and both see export markets beyond Canada.

Founder takeaway: When the government is both funder and customer, align your roadmap with its priorities early. Sovereignty and defence are opening doors that were closed just a few years ago.

Your social good story helps you raise capital

On the Moonshots stage, impact wasn't a side note. It was part of the business case. The clean tech panel put it plainly: sustainability and growth go hand in hand.

For physical AI founders, a strong social good narrative does real work. It drives funding, builds market awareness and educates the people you need on your side.

It helps drive funding. Impact opens doors to grants, government programs and government contracts that extend runway before a priced round. For example, SmartARM keeps customer costs low through government funding. Meanwhile, as Betakit reported, Ottawa is supporting Canadian launch capacity with a $200M lease at Spaceport Nova Scotia and the $8.3M grants for NordSpace, Canada Rocket Company and Reaction Dynamics.

Impact-minded investors are looking for exactly these stories. Tie your story to these priorities and investors can see a market that's growing.

• It earns PR and good news coverage. A prosthetic arm that gets better every week, or at-home testing that catches disease earlier, is a story the media wants to tell. Good news stories build market awareness far faster and cheaper than paid channels. They also give investors third-party validation before you walk in the room.
• It educates your market. Physical AI is new, and most buyers don't know what's possible yet. Your impact story is how you teach three audiences at once:
  • Customers, who need to see the problem clearly before they'll buy the solution, framed in their own terms, like less rework or earlier diagnosis
  • The public, whose trust matters when your product touches their health, safety or national security
  • Investors, especially in underfunded categories like women's health, where founders like Fibra's are still making the case for the market itself
• It rides policy tailwinds. Defence, sovereignty and women's health all have new federal attention.
• It attracts talent. The space panel said sovereign ambition is helping reverse Canada's brain drain. A clear mission helps you recruit people who could work anywhere.

How deep tech companies build a moat

In general AI, a good idea can be copied in weeks. In physical AI, the moat is the hard part itself. Across the Moonshots stage, the winners are differentiating in five ways.

1. Proprietary IP. Fibra's fabric sensors, SmartARM's arm, Kepler's optical links in orbit. Hardware and science that took years to build are not easy to replicate.
2. Deep domain knowledge. Augmenta.ai knows hospital ventilation and medical vacuum systems as well as it knows machine learning. Science & Humans runs its own lab. You can't fake that expertise.
3. Error-free models for niche use cases. A chatbot can be wrong and you shrug. A building design that fails code compliance, or a prosthetic that misreads a grip, cannot. Rigour and accuracy in narrow, high-stakes settings is the product.
4. Data flywheels. Every SmartARM user makes every arm smarter. Every Augmenta.ai project adds to a dataset that doesn't exist anywhere else. As Eclipse Venture's Joe Fath told Crunchbase, the strongest moats will likely go to companies that vertically integrate and own multiple layers of the stack.
5. A clear path to scale. Augmenta.ai went from a school to a hospital to a hyperscale data centre, with the rest of MEP on their product roadmap. Science & Humans plugs into insurance and telehealth. NordSpace is already planning for export. Scalability is what turns a clever niche tool into a platform.

At the end of the day, it's the founding team

Technology, IP and market size all matter. But investors keep coming back to one question: why is this team the one to execute best?

NRTH gave clear answers. The Fibra founder lived through a frightening PCOS diagnosis as a teenager. Mystoria's founder turned her own fertility journey, and the exhausting work of managing her care, into a company. SmartARM's most committed customer believed so strongly that he led their funding round. NordSpace's founder chose to build in Canada rather than head to the US. The clean tech panel said simply: build a strong team.

That is founder-market fit. When a founder has lived the problem, has the technical depth to solve it, and has the grit to push through long R&D cycles, investors take notice. In a sector this capital-intensive, the team is the bet.

Further validation: what today's investment trends say

Physical AI is being treated as the next leg of the AI boom.

• Funding has nearly quadrupled. Per Crunchbase's article in August 2026, physical AI startups raised $47.4 billion across 521 deals in the first half of 2026. That compares with $12 billion in the second half of 2025, and $26.4 billion in the first half of 2025. The first six months of 2026 alone beat the $41.9 billion raised across 2022 to 2024 combined.
• Rounds are getting much bigger. For pure-play robotics and embodied AI companies, New Market Pitch found the median round jumped from $61 million to $165 million year over year, with 72% of 2026 deals above $100 million.
• Capital is backing proven teams. Follow-on rounds made up 92% of those 2026 deals. Investors are doubling down on companies that have already cleared a credibility bar.
• Strategic investors are leaning in. NVIDIA, Bezos Expeditions, and industrial corporates, show up again and again, because they benefit directly if physical AI scales. These strategic investors own the infrastructure, data centers, factories, or logistics networks that will power the physical AI revolution.
• Defence and space are hot. Crunchbase notes that 2026 exits have clustered in aerospace, defence and drones, led by SpaceX's record IPO.

The Canadian picture from NRTH fits this pattern at an earlier stage. The Government is stepping in with money and regulation and investors are increasingly taking calculated risks with deep tech AI portfolio companies. The caveat is that investors increasingly want to see real deployments, uptime, repeat customers and safe operation.

The NRTH space panel summed it up: the opportunity is here now, and it's all about execution. The funding data says the same thing.

What this means if you're raising at seed or Series A

The headline numbers are megarounds, but the signal for early-stage founders is different and investors are increasingly backing teams that have already cleared a credibility bar. That bar is higher than it used to be, and you need to show proof earlier.

Here's how to build your proof

The tactics below are ones I've used with VC-funded deep tech companies:

1. Borrow credibility and drive marketing and sales narratives through channel partners. Enterprise buyers are cautious about unproven AI in their operations. At DarwinAI, partnerships with known infrastructure players like AWS and Intel gave enterprise pilot customers, including Honeywell, Cisco and Lockheed Martin, the confidence to say yes. A trusted partner shortens the path to your first logos.
2. Turn deployments into case studies, in your customer's language. Investors want evidence the model works in the real world. Show efficacy, time saved and cost saved, using the metrics your customers already care about. A plant manager speaks in yield and downtime, for example. When your proof points resonate with buyers, they resonate with investors too.
3. Make your team's caliber visible. Investors always bet on the team. At DarwinAI, we highlighted our chief scientist's academic conference awards and our CEO got on stage at many speaking engagements. Awards, visible article publications, academic papers, and coveted keynotes are third-party proof that your team is the best one to execute.
4. Tell your moat story as a flywheel. Don't just list your IP. Explain how every deployment adds data, sharpens accuracy and makes the product harder to copy. That's the story SmartARM and Augmenta.ai told on stage, and it's what investors pay premiums for.
5. Match your proof to your stage. At early seed, show technical proof and credible pilots with named partners. At Series A, show pilots converting to paid deployments, repeat customers and a clear path to scale. Your marketing and go-to-market should be built to hit the milestones your next round depends on.

Building in physical AI? Let's talk

If you're a founder preparing to raise, I'd love to help. At Unicorn Rocket Marketing, I help deep tech AI teams turn their technology, data, and early customer wins, into a story that investors and enterprise buyers believe. That means positioning your moat, building case studies in your customers' language, and making your team's expertise impossible to miss.

Let's Talk