Welcome back, Spectrum Visionaries.
CES 2026 offered a sneak preview of the AI roadmap for the next decade. AI has begun its transition beyond assistive chatbots toward operating in the real world. New processors are making AI-vision workloads feasible on edge devices, even at significantly lower power levels.
This issue explores AI’s transition from assistive chatbots to physical devices, why forecasts point to 2 million workplace humanoids by 2035, and how Physical AI is steadily shifting from cloud dependence to the edge over the coming decade.
THE AI UPDATE
AI Shifts From “Copilot” to “Autopilot”
CES 2026 signaled a shift in AI systems operating with minimal human oversight. Boston Dynamics’ Atlas (CNET’s Best Robot) is moving toward deployment for car parts handling and assembly coordination at Hyundai plants. LG’s CLOiD demonstrated a “Zero Labor Home,” managing laundry, cooking, and connected appliances without human input.
The shift represents what Bernard Marr calls moving from “Copilot to Autopilot”: systems that anticipate needs and act on their own, beginning to appear in factories and homes.
Timeline: Industrial humanoid deployments (e.g., Atlas at Hyundai) starting in 2028 for tasks such as parts sequencing, expanding to factory-scale integration by 2030, while home autonomy systems (e.g., CLOiD) demonstrated in 2026, targeting late-2020s commercial availability as the technology matures.
Deloitte Forecasts 2M Workplace Humanoids by 2035
Deloitte's Tech Trends 2026 report points to 2 million humanoid robots entering workplaces over the coming decade, with UBS estimates that the number could surge to 300 million by 2050. The consulting giant says Physical AI systems are transitioning “from niche to mainstream adoption” as manufacturing costs dropped 40% between 2023-2024, making humanoids economically viable for the first time.
Amazon has deployed its millionth robot, while BMW is testing humanoids for precision assembly tasks.
Timeline: Gradual scaling to reach 2 million workplace humanoids by 2035, with mainstream adoption accelerating from 2027 as costs fall and pilots (e.g., Hyundai, BMW) prove viability.
THE DEEP DIVE
Physical AI's 10-Year Shift from Cloud to Edge
Physical AI’s 10-Year Shift from Cloud to Edge
For years, AI vision relied on the cloud. Cameras captured footage, sent it to distant servers for processing, then waited for instructions to come back, but that’s changing.
The Cloud Bottleneck
Cloud-based AI works fine when connectivity is strong, and latency doesn’t matter. But autonomous vehicles can’t wait for a server response to detect pedestrians. Rescue robots in disaster sites can’t count on networks that went down with the structure.
Processing visual data remotely also means sending massive amounts of footage over networks, and scaling that across thousands of devices makes bandwidth costs explode.
Chips That Will See Locally
At CES 2026, chipmakers unveiled the processors built to move AI vision onto devices. Ambarella’s CV7 chip can process 8K video locally with 20% less power, entering production this year. Autonomous vehicles and industrial robots are being developed to make decisions in milliseconds using Arm platforms, no cloud, no delay, no dependency.
Battery-Free Breakthrough
ASYGN demonstrated the world’s first battery-free AI vision device at CES 2026. The prototype recognizes hand gestures using just one milliwatt of power, harvested entirely from ambient light. The device runs indefinitely on the same energy that illuminates a room.
Why it matters: Edge AI positions robots to operate in disaster zones without connectivity, drones to fly longer on smaller batteries, and wearables to run for months instead of days. As critical processing moves to the edge, AI can operate anywhere.
Timeline: Ambarella CV7 chips entering production late 2026-2027 for integration into cameras, robots, and vehicles. Also, ASYGN prototypes demonstrated in 2026, with commercial products likely emerging 2028-2030 as energy harvesting and low-power AI scale.
THE MARKET
Humanoid Robots
Growth
The humanoid robot market approached $5B in 2025 and is projected to reach $251B by 2035, driven by falling hardware costs and rising enterprise demand across automotive, logistics, and healthcare.Capital Flows
The sector raised over $5B in 2024-2025, with Figure AI securing $1B at a $39B valuation and Agility Robotics raising $400M. Strategic buyers like NVIDIA, Microsoft, and Amazon are moving early to acquire humanoid platforms and secure manufacturing partnerships ahead of deployment.Adoption
Between 2023-2024, manufacturing costs fell 40%, moving systems from pilot programs into early production deployments. North America holds 42% of the global market, with manufacturing and automotive leading adoption, while healthcare emerges as the fastest-growing segment.
Takeaway:
Market growth supports mass industrial deployment through 2027-2030, with automotive and logistics leading commercial adoption while foundation model development and cost reduction enable consumer applications by 2028-2032.
What did you think of today’s issue?
Hit reply and tell us why; we actually read every note.
See you in the next upload
