Artificial intelligence gets physical

Technological revolution is a long-term megatrend that we believe will influence financial markets for decades. Within this area, artificial intelligence (AI) has been the focus for investors. However, as the technology matures, companies are increasingly looking for ways to embed AI into their operations. Robotics and automation are moving into the spotlight as AI becomes physical.

Historically, robots have been limited to repetitive tasks requiring accuracy, speed and reliability. As a result, they have mainly operated in structured and predictable environments such as assembly lines or operating theatres.

While robotics and automation have existed for many years, several factors are now coming together that makes the area particularly interesting:-

• Labour shortages: Ageing populations, urban migration, and a preference for safer, less repetitive jobs are reducing the supply of available labour.

• Advances in AI: AI allows robots to learn and adapt. This means they are now intelligent systems rather than rigid machines.

• Geopolitical tensions: Robotics is increasingly seen as a strategic technology, with countries competing to lead in advanced manufacturing, automation and defence.

As a result, robotics is no longer a niche industrial tool, but an economic necessity, playing a vital role in global supply chains and national security. Demand for robotics has increased in recent years, with industrial robot installations forecast to reach around 700,000 per annum up from 542,000 in 2024.

Robotics adoption is increasing across a growing number of industries and below we highlight some of the key areas with the strongest potential.

Industrial automation: This is the largest and most established market for robotics. The market is currently valued at around $24 billion and is expected to quadruple by 2034.¹ Growth is being driven by smart factories, automation and reshoring of supply chains.

Smart factories, for example, integrate Internet of Things (IoT) sensors, AI and machine learning, alongside flexible robots that can switch between tasks. This enables data driven, adaptive and self-optimising production, improving decision-making, creating more resilient supply chains and boosting profit margins.

Warehousing and logistics: The rapid growth of ecommerce has made warehousing and logistics one of the fastest growing areas for robotics and automation. Demand for robots is expected to grow around 17% per annum by 2033². Consumers increasingly expect faster delivery and real-time tracking, driving the need for speed and accuracy.

Automated systems can meet this demand at a scale that would not be possible with human labour alone. For example, automated pallet shuttles can move up to 11,000 pallets a week³. Despite this, most warehouses remain largely manual leaving scope for further growth.

Medical and agriculture: Both industries have long histories of automation, yet there is still plenty of opportunity for further growth. In healthcare, precision and accuracy make robotics particularly valuable as they improve operational success rates and recovery times. Medical robot sales rose 91% in 2024, driven by the need for diagnostic and laboratory robots⁴.

In agriculture, robotics has expanded from milking machines into harvesting, planting, weeding and crop management. Advanced systems can now scan crops and apply tailored doses of fertiliser to each individual plant. This reduces fertiliser use by up to 70%, lowering costs for the farmer while improving environmental outcomes⁵.

Humanoid robots have long captured the imagination, but the appeal is economic as well as aesthetic. Because the world is designed for humans, humanoids can collaborate with people and operate in existing environments, opening new opportunities for their use. There are three main aspects to a humanoid:

• Brains: Advances in AI means humanoids can increasingly process the large volumes of information at the speeds required for human environments. Nvidia’s Jetson Thor platform, for example, has been built with humanoid robots in mind. It can integrate vision, movement control and decision making into a single system.

• Battery: Battery life directly effects the operating time and therefore the return on investment. Currently, humanoids typically operate for one to four hours, with run time dropping significantly during strenuous activities, such as walking. Solid-state batteries offer higher energy density, improved safety, and faster charging, but are expensive and difficult to manufacture. As costs fall, this technology could be transformative.

• Brawn: This is the skeleton, muscles and sensors of a robot. Dexterity remains one of the biggest engineering challenges. The UK-based Shadow Robot Company produces one of the most advanced robotic hands, capable of near-human dexterity in controlled settings⁶. But it struggles with the soft and irregular objects. It is also expensive - between $60,000 and $100,000 per hand⁷. In comparison, a Jetson Thor AI module is forecast to cost around $3,000-$4,000 per robot⁸.

Companies are investing in this technology today. BMW have conducted what is widely seen as the most advanced humanoid trial to date. Its Figure robots completed an 11 month trial, working 10 hour shifts with 90,000 parts loaded at an accuracy of 99% - contributing to the building of 30,000 vehicles⁹.

Production has reached an inflection point. For the first time companies are simultaneously building out production lines: Optimus (Tesla), Figure (BMW/Figure AI), Apollo (Apptronik) and Digit (Amazon/Agility Robotics) have all moved from research and development to early stage production.

Robotics follows a classic thematic investment cycle. Valuations peaked in 2021 amid low interest rates, Covid related stimulus and enthusiasm for disruptive innovation. Investors priced in decades of growth at once. Having become over-owned and over-priced the theme broke down in 2022 as higher interest rates made future growth less appealing, while current economic growth slowed weighing on investment in the sector. Today, fundamentals and AI driven demand have improved, supporting a recovery in performance.

Robotics and automation are approaching an important inflection point. While humanoid robots remain at an early stage, broader adoption across industry, logistics and healthcare is already under way. For investors, the opportunity lies not only in headline robot manufacturers, but also across components, software and services that support this evolving ecosystem.

1. Robotics as a Service (Raas) Market Outlook 2025-2034: Market Share, and Growth Analysis

2.Robotics Market Size | Share | Growth Report [2033]

3. Better, Bigger, and Easier Automation in the Warehouse With Pallet Shuttle System | New Equipment Digest

4. World Robotics 2025 report – SERVICE ROBOTS – released by IFR - International Federation of Robotics

5. Fertilizer-Saving Robots : upside robotics precision dosing

6. Shadow Robot: Dexterous Hand & Teleoperation | Robots Canada

7. Dexterous Hand by Shadow Robot – Humanoid Robotic Hand - Humanoid.guide

8. Nvidia's 'robot brain' chip, Thor, goes on sale around the world

9. F.02 Contributed to the Production of 30,000 Cars at BMW

Past performance is not a guide to future performance.

Neither MSCI nor any other party involved in or related to compiling, computing or creating the MSCI data makes any express or implied warranties or representations with respect to such data (or the results to be obtained by the use thereof), and all such parties hereby expressly disclaim all warranties of originality, accuracy, completeness, merchantability or fitness for a particular purpose with respect to any of such data. Without limiting any of the foregoing, in no event shall MSCI, any of its affiliates or any third party involved in or related to compiling, computing or creating the data have any liability for any direct, indirect, special, punitive, consequential or any other damages (including lost profits) even if notified of the possibility of such damages. No further distribution or dissemination of the MSCI data is permitted without MSCI's express written consent.