The state of the realm¶

The technology landscape currently resembles Ankh-Morpork during one of its periodic enthusiasms, where everyone is simultaneously convinced that enormous fortunes are to be made, that the current situation cannot possibly continue, and that they personally must act immediately or be left behind permanently. These three beliefs are not entirely compatible. This has never stopped anyone.

Artificial intelligence has achieved the status of transformative technology while remaining mostly experimental. Large language models can generate plausible text, write functional code, and convince investors to write cheques for billions of euros, which are admittedly impressive capabilities even if not quite the artificial general intelligence that press releases sometimes imply. The technology works, within limitations that vendors discuss quietly and customers discover loudly.

Meanwhile, quantum computing continues its long tradition of being five to ten years from practical utility regardless of which year you ask. The metaverse is discussed less frequently now. The Patrician considers this progress.

Where the money is flowing¶

Venture capital is still pouring into AI startups with the enthusiasm of prospectors who have heard there is gold in the hills but have not personally verified the reports. Funding rounds for companies with impressive demonstrations but uncertain business models reach valuations that would make traditional financial analysts reach for the smelling salts.

The established technology companies are investing even more heavily, which is either validating the AI enthusiasm or demonstrating that nobody wants to be the executive who declined to invest in the next transformative thing. Whether the thing was transformative becomes clear later. The reputational damage from missing it is immediate. The incentives are not subtle.

Hardware manufacturers are enjoying unprecedented demand. Nvidia’s market capitalisation has increased by hundreds of billions of euros based largely on demand for AI accelerators. The entire supply chain is experiencing boom conditions that will either prove justified by sustained demand or result in expensive overcapacity when the cycle turns. Both outcomes are historically normal.

The capability question nobody can answer¶

Whether AI capabilities continue improving at current rates or are approaching diminishing returns determines whether enormous infrastructure investments are prescient or premature. Nobody knows. Substantial capital is being deployed based on assumptions about which it is, which is either high-stakes investing or very expensive guessing, depending on how events unfold.

The scaling hypothesis, that larger models trained on more data produce better results, has held well so far. This is why companies keep training ever-larger models consuming ever-more resources. In January 2025, a Chinese laboratory called DeepSeek produced a model of competitive capability for approximately six million euros, at a fraction of what the industry assumed necessary. The hypothesis may still hold at the frontier. It no longer holds that frontier-level results require hyperscaler resources, which is a significant revision to several business models that had not pencilled in this possibility.

Geopolitical complications¶

The comfortable assumption that technology would remain globally integrated is not holding up. Export controls restrict Chinese access to advanced chips. China is investing heavily in domestic alternatives. Europe is building technological sovereignty. The result is a fragmenting global technology landscape that is less efficient but possibly more resilient, depending on your definition of resilience and your position in the fragmentation.

Standards battles are emerging where geopolitical blocs promote competing technical standards. These battles are partly about technical merit and partly about strategic positioning. They produce duplication, incompatibility, and increased costs. They also provide employment for standards committee members, which may explain why there are so many of them.

The Patrician’s assessment¶

The Patrician concludes that uncertainty is permanent, that grand predictions should be discounted heavily, and that the current moment is characterised by unusual uncertainty about which technologies will prove transformative and which will prove expensive distractions. AI might revolutionise knowledge work or might prove to be sophisticated automation of routine tasks. Both outcomes are possible. Honest observers would acknowledge they cannot confidently distinguish between them from current evidence.

The wise approach is watching where the serious money goes rather than listening to what people say, remembering that all parties have incentives to shade their descriptions toward their interests, and maintaining scepticism that comes from having watched previous enthusiasms resolve in ways that consistently surprised everyone involved.

This approach lacks the inspirational quality that makes for compelling conference keynotes. It has the considerable advantage of occasionally being correct.