Proception’s story reads like a cautionary tale—until you look closer and realize it’s also a blueprint for how early-stage robotics companies survive contact with the real world. The company, founded by CEO Jay Li, has reached a settlement with Tesla in a trade secret lawsuit, and it’s using the moment to pivot from legal turbulence to growth. Proception also announced an $11 million funding round, giving it fresh runway as it continues building robotic hands and the software that helps them operate reliably in messy, unstructured environments.
For founders, lawsuits are rarely just legal events. They’re operational events. They pull attention away from product iteration, force teams to spend time on discovery and documentation, and can complicate hiring, partnerships, and even customer conversations. Li doesn’t frame the experience as something he’d recommend to anyone trying to get a startup off the ground. But in an exclusive interview with TechCrunch, he suggests the company may be better positioned after enduring it—less because the outcome was inherently positive, and more because the process tested whether Proception could keep executing under pressure.
“I think it’s kind of like a resilience test, or pressure test,” Li told TechCrunch. “People say [ … ]” He didn’t present it as a marketing slogan. The point, as he described it, is that the company learned what it takes to keep moving when the environment stops being purely technical and becomes adversarial, unpredictable, and time-consuming.
That distinction matters, because robotics startups often sell a vision of engineering progress: better dexterity, better perception, better control. But the path from prototype to deployment is rarely linear. It’s a chain of dependencies—hardware reliability, sensor calibration, manufacturing consistency, safety considerations, and integration with downstream systems. When legal conflict enters the picture, it adds another dependency: the ability to protect your work while continuing to build.
What Proception is building—and why it draws scrutiny
Proception’s core focus is robotic manipulation, specifically robotic hands designed to handle objects with the kind of adaptability that industrial automation typically struggles with. In practice, that means dealing with variability: different shapes, textures, weights, friction coefficients, and failure modes. A robotic hand isn’t useful if it only works in a lab with carefully curated objects and perfectly repeatable conditions. The value comes when it can generalize—when it can pick up, reposition, and interact with items in ways that don’t require constant reprogramming.
That’s a hard problem, and it’s also a competitive one. Robotics is full of overlapping approaches: similar mechanical architectures, similar sensing strategies, similar control loops, and similar learning pipelines. Even when two companies arrive at comparable solutions independently, the resemblance can become legally meaningful if one party believes the other used protected information.
Trade secret disputes are especially sensitive because they sit in a gray zone that’s not always about who invented something first. They’re about what was known, what was shared, what was documented, and what can be proven. For a startup, that can be existential. You can’t simply “out-innovate” a legal claim in the short term. You have to defend your process, your records, and your intellectual property posture—while still shipping.
The settlement: what it signals, and what it doesn’t
Proception’s settlement with Tesla ends the trade secret lawsuit, but settlements rarely function as public verdicts. They’re often structured to reduce uncertainty and cost, allowing both sides to move forward without the risk of a prolonged fight. That doesn’t mean the underlying questions disappear; it means the parties chose not to litigate them to conclusion.
In the context of a robotics startup, the practical significance is immediate: time and attention are freed. Legal proceedings can stretch for months or years, and during that period, teams can lose momentum. Even if the company believes it’s in the right, the opportunity cost is real. Settlement changes the timeline.
It also changes the company’s narrative. Before the settlement, Proception had to operate under the shadow of allegations. After the settlement, it can speak more directly about its product direction and its fundraising plans. That’s not just a communications advantage—it’s a business advantage. Investors, partners, and customers often want clarity. While they may not require a legal win to invest, they do require reduced uncertainty.
Li’s framing of the experience as a “pressure test” suggests Proception sees the settlement as part of a broader maturation process. The company’s ability to continue operating through the dispute implies it had internal discipline: documentation practices, engineering continuity, and a capacity to keep building without letting the legal situation derail the roadmap.
The $11 million raise: why timing matters
Proception’s announcement of an $11 million funding round is notable not only for the amount, but for the timing. Startups often raise capital to accelerate product development, expand teams, or prepare for manufacturing scale. But raising after a high-profile legal dispute can be harder than raising before one. Some investors become cautious when they see litigation risk. Others worry about distraction and reputational fallout.
That Proception secured funding anyway suggests two things. First, investors likely believe the settlement reduces the most acute risks. Second, they likely believe the company’s technical trajectory remains compelling.
In robotics, capital is not just fuel—it’s infrastructure. Building robotic hands involves iterative hardware design, testing cycles, and often expensive prototyping. It also requires talent across mechanical engineering, controls, perception, and software. If Proception is aiming to move from early demonstrations toward more repeatable performance, it needs resources that go beyond what a small team can sustain.
An $11 million round can support multiple phases at once: refining mechanical designs, improving sensing and grasping reliability, strengthening data pipelines, and scaling production processes. It can also support the less visible work that determines whether a system survives real-world deployment: reliability engineering, failure analysis, and the kind of operational rigor that makes robotics products dependable rather than merely impressive.
A unique take: resilience as an engineering capability
It’s tempting to treat resilience as a personality trait—something founders either have or don’t. But Li’s comments point to something more concrete: resilience as an organizational capability. In robotics, where systems must handle uncertainty, resilience is already a technical requirement. The same mindset can apply to the company itself.
Consider what “pressure test” means in engineering terms. Under pressure, you find out which assumptions were fragile. You discover where processes were informal. You learn whether knowledge was concentrated in a few people or distributed across the team. You see whether documentation exists when you need it. You find out whether the company can keep making decisions quickly when external constraints tighten.
A legal dispute forces those questions. It can reveal whether a company’s engineering work is reproducible internally, whether it can explain its design choices, and whether it can maintain continuity when priorities shift. If Proception came out of the dispute able to raise capital and continue its roadmap, that suggests it didn’t just “survive”—it adapted.
There’s also a strategic dimension. Robotics companies often face a choice between speed and defensibility. Speed means iterating quickly, sometimes with less formal process. Defensibility means building a paper trail, protecting IP, and ensuring that the company’s claims about originality and ownership are supported by evidence. Those goals can conflict early on, when teams are small and time is scarce.
Li’s perspective implies Proception managed to balance them enough to reach a settlement and keep moving. That balance is difficult, and it’s one reason legal outcomes matter even when they don’t become public precedents. The settlement is a signal that the company’s position was credible enough to resolve the dispute without a drawn-out battle.
Why this matters for the robotics ecosystem
Proception’s case highlights a reality that many robotics founders learn the hard way: competition isn’t only about who can build the best hand. It’s also about who can build it in a way that withstands scrutiny—technical, commercial, and legal.
Robotics is increasingly software-defined. Many of the differentiators are not just mechanical but algorithmic: how the system perceives objects, how it plans grasps, how it learns from failures, and how it adapts to new scenarios. That means the “product” includes datasets, training pipelines, and internal methods. Those are precisely the kinds of assets that can become the subject of trade secret claims.
At the same time, robotics is a field where independent invention is common. Similar problems lead to similar solutions. Two teams might both converge on comparable approaches because the physics and constraints push them there. The challenge is proving independence and demonstrating that no protected information was misused.
Settlements, in that sense, are often the least-bad outcome for both sides. They avoid the uncertainty of court rulings and the expense of extended litigation. But they also leave the industry with fewer clear lessons about what exactly was at issue. That’s why founder narratives—like Li’s “pressure test” framing—are valuable. They offer insight into how companies experience these events internally, even if the legal details remain private.
What Proception can do next with this momentum
With the settlement behind it and new funding in hand, Proception’s next steps will likely focus on execution: turning engineering progress into repeatable performance and building a path toward scalable deployment.
In robotics, the hardest part is often not the first working demo—it’s the second, third, and hundredth. Systems that perform well in controlled settings can degrade when exposed to real-world variability. That degradation can show up as inconsistent grasp success rates, increased wear on components, sensor drift, or brittle behavior when objects are partially occluded or unexpectedly oriented.
An $11 million round can help address those issues by funding the unglamorous work: more extensive testing, improved calibration routines, better failure recovery, and tighter integration between hardware and software. It can also support customer-facing iterations, because real deployments generate the data that improves robustness.
There’s also the question of partnerships. Robotics companies often need integration with upstream and downstream systems—conveyors, warehouses, picking stations, quality inspection workflows, or robotic arms