AI Beyond the Chatbot: The New Value Chain

This piece takes a look at some of the current applications of AI outside of the LLMs we are all familiar with by now. We’ll dive into technologies like world models, agentic orchestration, and AI drug discovery, as well as the pressing challenges of energy consumption and how the industry is addressing them.

A key objective of the current U.S. administration has been a very strong push for reshoring. While the specific methods used to achieve this may change between administrations, the overall push for independence from competing world powers is unlikely to go away.

The biggest challenges to overcome when bringing production back to wealthier countries are labor costs and demographics (shortages). Automation is the singular tool that solves for both, and AI is playing a critical role in making it viable. It is accelerating traditional development and, more importantly, enabling brand-new possibilities. World models are at the heart of this. These internal simulations allow robots to “see” the world around them while understanding the underlying physics. This allows a robot to practice an action internally before executing it physically, drastically reducing the likelihood of damaging inventory, the people around it, or the machine itself.

These models are instrumental to the development of embodied AI, a vital building block of reshoring. It allows for faster automation of new factories, but more importantly, the rise of humanoid robotics opens the doors of automation to existing factories. This could prove to be a pivotal shift for domestic manufacturing.

Where world models bridge the gap between AI and the physical world, agentic orchestration seeks to bring AI into our day-to-day lives. Agents transform AI from a tool that we ask questions or give orders to into one that proactively handles complex, multi-environment tasks.

Consider the following application in healthcare:

• Current state: AI enhances the precision of cardiac monitors.
• Agentic future: Upon detecting an anomaly, an agentic system can autonomously alert a cardiologist, schedule necessary imaging and tests, and coordinate insurance authorizations across multiple stakeholders.

This freedom to execute and coordinate at high scale would allow a substantial number of workers to be deployed to other areas of need within the economy. In other areas like e-commerce, agents are already beginning to browse, compare, and purchase on a user’s behalf, potentially changing how we interact with online consumption. As Dr. Henrik Christensen, director of the Contextual Robotics Institute at UC San Diego, put it, “Just as pocket calculators transformed everyday mathematics, pocket AI will become an essential assistant integrated into daily life.”

Healthcare is another field where AI is making massive strides away from the public eye. Augmented imaging, diagnostics, and enhanced surgical robots are some of the examples of areas where AI is already enabling better accuracy and superior outcomes. However, the area where AI could have the most profound impact is drug discovery. Currently, approximately 90% of drug candidates fail during clinical trials. Any technology that tilts those odds in favor of drug developers represents a massive win for society, as it would represent more resources going to actual treatments instead of unused research. 

Recent partnerships, such as the one between OpenAI and Ginkgo Bioworks, have integrated models like GPT-5 into autonomous labs, allowing the AI to not only propose experiments but run them at scale, while learning from the results. This has already brought protein production costs down by roughly 40%. Another example of this is a 2025 McKinsey study that estimates long-term savings in clinical trials of 30%–50%, highlighting the revolutionary potential of this technology.

These benefits do not come without structural challenges. Energy consumption has moved to the forefront of the conversation. As one example, the state government in Georgia is proposing a pause on data center expansion in the state until their effect is better understood. Challenges like these emphasize the importance of accessing the segment with a research-lead approach like that of the ROBO Global Artificial Intelligence Index (THNQ) that takes all these things into consideration. 

Beyond spotting trends, it is key to look at the companies tackling the industry’s biggest hurdles. A prime example is the proactive effort by the industry to solve the energy bottleneck, which is being approached from two directions.

On top of the use of mainstream renewable energy sources, there is a massive push for clean technologies like nuclear small modular reactors (SMRs). Albeit in their early stages at this time, SMRs allow data centers to decouple from the grid while maintaining a zero-emission profile (read more

• Edge AI: Smaller models operating at the source (on-device) remove the processing load from centralized data centers. THNQ constituent Ambarella’s (AMBA) latest system on chip (SoC) is specifically designed for multi-modal edge computing and provides a 2.5x performance jump over the previous generations.
• Optical transfer: Moving data using light instead of copper wires increases speed while reducing cooling demands. Companies like Lumentum (LITE), which was added to the THNQ index back in Q3 2025, are at the forefront of this technology. 
• Advanced materials: The use of gallium nitride (GaN) and silicon carbide (SiC) allows for power delivery with nearly 99% efficiency. This enables server racks to quadruple their power density while reducing energy loss by 30%. Infineon Technologies (IFX) last year announced it was on track to produce 300mm GaN wafers at scale, bringin GaN closer in cost to traditional silicon.

The picture we have now suggests that AI itself can be part of the solution to its own energy constraints. It wouldn’t be surprising to see energy consumption continue to climb, but data centers are often focused on securing their own power apart from the grid and using clean sources. This would allow data centers to significantly scale their output while remaining sustainable. After all, while AI is undeniably energy-hungry, its primary value is its ability to increase productivity across the board.

The race to address these structural challenges is just as intense as the race to build new models and new implementations of the technology. For investors, this dynamic environment creates opportunities for those who make decisions based on deep technological and industry knowledge. This conviction is why the ROBO Global Artificial Intelligence Index (THNQ) strategy is underpinned by a strategic advisory board of world-renowned researchers and industry experts. After all, the AI industry now encompasses much more than just the big names developing models or making chips. THNQ captures companies across the entire value chain, including those deploying world models, crafting the next generation of agents, and pioneering energy efficiency, from advanced semiconductor materials to data center power delivery. 

THNQ is the underlying index for the ROBO Global Artificial Intelligence ETF (THNQ) and the L&G Artificial Intelligence UCITS ETF (AIAI.LN).

The Hidden Robot

Data Centers Embracing Nuclear, SMRs for AI Needs

Looking for regular updates? Subscribe here for weekly insights on robotics, AI, and healthcare technology, delivered straight to your inbox.

For more news, information, and analysis, visit our Disruptive Technology Content Hub.

VettaFi is the index provider for THNQ ETF and AIAI.LN, for which it receives an index licensing fee. However, THNQ ETF and AIAI.LN are not issued, sponsored, endorsed, or sold by VettaFi. VettaFi and its affiliates have no obligation or liability in connection with the issuance, administration, marketing, or trading of THNQ ETF and AIAI.LN.