Chinese factories adopt agentic AI while skipping earlier digitalisation steps

By Splitfeed EditorialLast updated Oct 12, 2026, 13:10 UTC

On October 12, 2026, reports from the South China Morning Post and IndexBox described a deployment pattern taking shape in Chinese manufacturing, where factories are adopting agentic AI systems without having gone through the conventional digitalisation stages that Western manufacturers typically used as stepping stones. A company called Black Lake is identified as a central enabler of this shift, building and deploying AI agents that handle tasks autonomously inside Chinese factories. The reports also indicate that these deployments are being extended to international markets, though specific countries, contract values, and operational details were not established in the available coverage.

The leapfrogging framing is what gives this story its broader significance. In most industrialised economies, factory digitalisation followed a recognisable sequence, moving from basic data collection and enterprise resource planning through increasingly sophisticated automation before arriving at AI-driven autonomy. What the October 12 reports describe is Chinese manufacturers skipping much of that ladder and landing near the top. This raises genuine questions about whether the conventional sequence was ever necessary, or whether it was simply a product of the historical moment in which Western factories modernised. It also has competitive implications, since a factory running agentic AI without legacy enterprise software debt could, in principle, move faster and at lower cost than one that accumulated layers of earlier technology.

Because no reactions from the Pro-AI, Anti-AI, or Middle Ground camps had been published at the time of this story, what follows reflects what each camp would typically argue rather than anything anyone said. A Pro-AI camp would likely treat this as evidence that AI deployment can be democratised across industrial contexts and that the technology delivers practical value even where prior infrastructure is thin. An Anti-AI camp would typically raise concerns about deploying autonomous systems in environments that lack the governance structures and data foundations that earlier digitalisation would have built, arguing that skipping those steps stores up risk rather than eliminating it. A Middle Ground camp would probably focus on Black Lake's role specifically, asking whether the leapfrogging works because of something particular about that company's approach, or whether it signals a genuinely replicable model.

The clearest things to watch going forward are any operational performance data or independent assessments from the factories where Black Lake has deployed, and whether the international expansion produces comparable results outside China's specific manufacturing context. Those details would go a long way toward settling whether this is a durable industrial shift or a pattern that depends on conditions unlikely to travel.