AI: A View From Silicon Valley

CIO Tim Rocks sat down in early July with Raymond Tong and Harry Morrow of Loftus Peak to get an update on the state of play with AI. Harry had recently spent five weeks in the US, which informed much of the discussion.

Is it a bubble? Demand keeps outrunning supply

Harry and Raymond do not buy into the “end of the AI trade” pessimism. Having met around a hundred companies across five weeks, Harry found no trace of the doom-and-gloom mood you might pick up from the financial press. Instead, firms are quoting demand signals stretching out to 2028 and 2029, with one CFO of a multi-hundred-billion-dollar company calling those signals “crazy.” There are big differences from the dot-com era: unlike the dark fibre that sat unused for years, the data centres being installed now are monetised and put to work almost straight away. Anthropic was the headline proof point, with revenue climbing from roughly $10 billion at the end of last year toward a possible $100 billion run-rate by the end of this year. Supply constraints in chips and memory are actually holding any bubble in check, forcing discipline on spenders rather than letting them over-build.

Where we are in the rollout: still early, and mostly agentic

We remain early in the rollout story for AI. AI is progressing from early chatbots, to reasoning models, to today’s agentic systems that carry out whole multi-step workflows on a user’s behalf and make their own decisions along the way. But while the technology is arriving fast, actual deployment is still early. Most firms are using AI at a fairly basic layer rather than through deeply integrated, recurring workflows. Very few have so far been able to genuinely train and implement AI agents. This will change and create a long runway for AI usage.

Beyond the chips: physical AI, China, healthcare and energy

The next theme will be where the next legs of growth sit once you look past the data-centre names. One enormous area will be physical AI split between autonomous driving (Waymo expanding into the UK and Japan, Tesla ramping self-driving, Amazon’s Zoox) and robotics, where Amazon has passed a million warehouse robots and China leads on humanoids through firms like Unitree and UBTech. China has emerged as a distinct story: an ageing, shrinking workforce plus top-down government backing is driving unusually fast adoption and a genuine race with the US. Healthcare and life sciences will be another focus, with dramatic speed-ups in tasks like genome sequencing. Both Anthropic and OpenAI have named science and medicine as their most exciting near-term breakthroughs. Finally, the energy and electrification theme, including batteries paired with data centres, ties the whole picture back together.

Recent Podcast 

Words on the Next Frontier of AI

In this episode of Words on Wealth, Evans and Partners CIO Tim Rocks is joined by Harry Morrow and Raymond Tong, Portfolio Managers at Loftus Peak, to unpack the current state of AI and where it’s headed. Fresh off five weeks in Silicon Valley, Harry shares what companies are really saying about demand, and why the on-the-ground confidence doesn’t match the headline pessimism. The conversation covers semiconductor supply constraints, the explosive revenue growth of frontier AI labs, and why this infrastructure buildout looks nothing like the tech bubble of the past. The conversation also looks ahead to the next frontiers of AI, from agentic workflows and physical robotics to breakthroughs in health and life sciences. Tune in to find out more.

This episode is also available on Apple Podcast.