AI Has Killed Information Arbitrage, but Amplified Another Kind of Demand

Explanation as currency

Artificial intelligence has stripped away much of information’s scarcity — and with that, a once-profitable form of arbitrage. Data, reports and product takeaways can now be auto-generated in minutes. So what do markets and people trade when raw information is essentially free? The answer emerging from China’s tech scene and global markets alike: explanations — compact, coherent frameworks that reorganize facts into a view of the future.

The piece in Huxiu uses a historical lens — from Mercator’s map to Korzybski’s “the map is not the territory” — to argue that AI collapses the value of surface information while elevating interpretive models. A Substack post about “Citrini” reportedly erased hundreds of billions in market value in 48 hours, not by releasing new data but by offering a single, hard‑selling explanatory chain about AI-driven disruption. It has been reported that the post succeeded because its logic was internally coherent and difficult to refute in part — a point David Deutsch’s criteria of “reach” and “hard to vary” help explain: good explanations extend far beyond their initial domain and fall apart if you tweak their core parts.

What this means for China and markets

For Chinese tech firms such as Baidu (百度), Alibaba (阿里巴巴) and ByteDance (字节跳动), narrative construction is now a strategic asset alongside chips and cloud infrastructure. In a world where U.S. export controls and other geopolitical frictions make advanced hardware constrained, narratives can substitute for scarcity: they reframe product roadmaps, investment theses and policy risk in ways that move capital and talent. Regulators and state media in China also play an outsized role in which explanations gain traction, creating a different informational ecology than Western markets.

The practical takeaway is blunt. Companies should invest in building “hard to vary” explanations tied to measurable roadmaps. Investors should evaluate not just data but the structural robustness of the narratives driving valuations. And policymakers should recognise that essays, longreads and viral threads can have systemic market effects even without new facts — so how do you police persuasion without chilling legitimate analysis? In short: AI may have killed information arbitrage, but it has amplified demand for interpretive frameworks. Who gets to write those frameworks will shape winners and losers next.