At 2048 Ventures, we are obsessed with the future and invest in early-stage startups that build defensibility through data and technology. Lately, we’ve been thinking about the future of robotics.

Robotics Inflection Point

We are entering a new age of robotics, powered by commoditized hardware and advances in AI.

Supply chain disruptions and rising geopolitical uncertainty are pushing governments and enterprises to invest more heavily in automation and domestic manufacturing, accelerating demand for innovation in this space.

Over the past decade, the cost and availability of key robotics components have improved dramatically. Sensors have become commodity hardware, enabling robots to perceive their environments better and at a fraction of the cost. At the same time, compute has become significantly more powerful. Advances in GPUs and specialized chips now allow robots to run complex perception and control models directly onboard, rather than relying on external compute.

On the software side, breakthroughs in foundation models and generative AI are shifting robotics from deterministic programming toward learning-based systems. Companies like Skild AI and Physical Intelligence are developing models that enable robots to adapt to new environments and continuously improve through data and simulation.

Together, these shifts are fundamentally changing the economics and capabilities of robotics - now is the time to build.

Our Research

We’ve spent time speaking with experts across the robotics ecosystem and these are our takeaways:

1. Robots will require continuous “parenting” after deployment. Unlike traditional industrial automation, AI-powered robots are non-deterministic systems that will continue to encounter new edge cases once deployed. As a result, deployment is only the beginning of the robot lifecycle. Robots will require ongoing monitoring, retraining, and recalibration. Companies that control this post-deployment lifecycle layer may capture significant value in the robotics stack.
2. Simulation matters more than raw data. While proprietary datasets create moats, the real challenge is ensuring robots behave reliably in complex real-world environments. Winning platforms will rely heavily on simulation and synthetic environments to test millions of scenarios before and after deployment.
3. Specialized robots will precede generalized robots. For robots to gain widespread adoption, they must first prove they can perform specific valuable tasks that save labor or increase productivity in human processes. Early success will come from specialized systems designed for specific industries or workflows, rather than generalized robots.
4. Bipedal humanoids may not be the dominant near-term form factor. While humanoid robots have attracted significant attention, different robotic applications will likely favor different physical designs. Rather than a single dominant humanoid form factor, we expect a world where multiple robot embodiments coexist, each optimized for specific environments and tasks.

Our Thesis

We are excited about robots that drive efficiency and are purpose-built for industry. At 2048 Ventures, we are bullish on solutions that reinvent current workflows - especially in industries that remain tedious and heavily dependent on human labor.

We believe the most successful robotics companies will be vertically integrated and that specialized robots will win first. Systems designed to perform niche high-value tasks will drive early adoption. We see the most value in solutions for specific labor-intensive problems, whether they’re working alongside human workers or replacing manual processes entirely.

These models are already winning in warehouse settings. Systems like Kiva Robotics (now Amazon Robotics) transformed fulfillment operations by deploying fleets of autonomous mobile robots that navigate warehouse floors and move inventory more efficiently than humans.

While the long-term economics of Robots-as-a-Service (RaaS) models are still emerging, we believe some of the largest robotics companies may adopt this approach. Rather than selling hardware upfront, these companies will deliver outcomes like harvesting crops, inspecting solar panels, maintaining infrastructure, or automating construction workflows. By owning the full stack and delivering the service directly, robotics companies can capture more value while continuously improving their systems through operational data.

We are also excited about robots that enables entirely new capabilities - processes that were previously unsafe or impossible for humans to perform. Companies like Gecko are pioneering new levels of precision by operating in environments where human access is limited or dangerous. We see strong potential across applications such as deep-sea infrastructure maintenance, space robotics, surgical systems, and other high-risk environments.

We believe the next generation of robotics companies will not simply sell machines, they will reinvent entire service industries through autonomous systems.

If you are a founder working in vertical robotics, we'd love to connect: 2048.vc/pitch-us