Reposted from Lonely Brain

Language is humanity's greatest invention and also humanity's biggest bottleneck.

When large language models demolish the Tower of Babel, what exactly will they bring?

Humans built civilization through language, formed societies, and developed commerce. Humans monopolized language, and friction and misunderstanding between languages gave birth to most job positions.

Large models arrived, and in an instant, reduced language friction.

One: The essence of an enterprise is a language organization. What truly gets restructured on the B2B side is the enterprise as a language machine itself.

Two: The underlying code of commercial civilization is not currency—it's language. Past software automated "computation"; large models automate "expression" and "understanding."

Three: AI doesn't replace people; it replaces friction between people. Once AI becomes a high-bandwidth, low-friction intermediary layer, many people will discover they were merely "friction toll collectors."

Four: The essence of the Shuai Chonghu phenomenon is information asymmetry and unverifiable ability. AI will pierce through these.

Five: When language friction reaches zero, Coase's theorem will rebalance the equation. Coase said enterprises exist to reduce transaction costs; future companies may consist of only two types of people: decision-makers and AI auditors.

Six: The most expensive cost in human society is not ignorance—it's "understanding." Large models compress the "marginal cost of understanding" to near zero, and "helping others understand" faces devaluation.

Seven: Large models possess superhuman capabilities humans lack: being immersed in all contexts simultaneously. Humans can deeply engage in at most three to five projects at once. AI can simultaneously ingest an entire company's codebase, customer emails, contract terms, and meeting records, establishing connections between any two nodes.

Eight: The endgame of B2B competition is not a battle of models but a battle for process sovereignty. The key is entering mainstream processes, accessing core data, mobilizing real resources, and triggering action at critical nodes. The truly formidable capability of large models isn't point-level intelligence but end-to-end capability. Don't insert AI into a single step of existing processes as a helper; let AI redefine the processes themselves.

Nine: From "cost reduction and efficiency gains" to "dimensional strike," from SaaS to LaaS (Labor as a Service).

Ten: Individuals acquire "micro-corporatized" organs. Through AI, a person grows corporate "organs": externalized super memory, automated execution modules, round-the-clock customer service answers, precise data analysis. This will birth a new species of super individuals.

Eleven: Superhuman integrated intelligence. The human brain can only process 7±2 variables at a time, while large models can simultaneously integrate thousands of trivial elements (emails, reports, competitors, policies, customer complaints) to gain superhuman insights.

Twelve: In the future, what's most valuable is not "the ability to speak" but "having something worth saying." Firsthand experience, unique judgment, high-quality taste, and structured cognition formed through long-term practice—these things large models cannot replace for you.