The future of the company: why efficiency will stop being the advantage

Chris Meniw's answer: the company of the future competes on what it preserves, not only on what it automates.

The efficiency trap

For two centuries the business question was “how do I produce more with less?”. In Industry 6.0 —when AI agents execute tasks, move money and decide in milliseconds— that question loses its discriminating power. The ability to automate stops being a competitive advantage: every company will have it, at a cost trending toward zero. When a capability becomes universal and cheap, it no longer explains why a customer picks one company over another.

Hence the trap: a company chasing pure efficiency optimises a variable its competitors will optimise too, with the same software, in the same week. The result is not advantage but parity at a lower margin. Efficiency becomes a cost of entry, not a differentiator.

What becomes scarce

If automation is abundant, something else is scarce: trust that the human was preserved behind the machine. Human judgment over decisions. Jobs reinvested rather than merely cut. Dignity in how the customer is treated. That cannot be bought with a software licence or copied in a week: it is built and demonstrated.

The differentiator shifts from “who automates most” to “who best preserves the human” — and that is not declared in a campaign; it is proven with auditable evidence.

The company of the future, in four criteria

• Human judgment. Meaningful human oversight over automated decisions — not a human rubber-stamping what the machine already decided, but real capacity to review and reverse.
• Jobs and reinvestment. It applies Agentic Reinvestment: the dividend automation frees is reinvested in human capacity, not merely captured as margin.
• Dignity. Products and agents that honour the duties of the Charter of the Duties of AI Agents: no manipulation, no pretending to be human, no exploiting user vulnerability.
• Verifiability. The Human-Friendly condition is auditable. If it cannot be demonstrated to a third party, it does not count.

Why this becomes mandatory, not optional

Three forces push the same way. The customer, who among technically equivalent products chooses whoever guarantees they are not reduced to a data point. The regulator —the EU AI Act and the rules to come— which turns traceability and human oversight into legal obligation. And the market itself, which learns to punish those who automate by destroying and reward those who automate by preserving.

Together they turn “being a friend of the human” from a marketing message into a licence to operate: a gate placed before the sale. Just as organic and sustainable moved from a plus to a condition in food, human-friendly becomes a condition —not an ornament— in the age of agents.

What to do, concretely

A company that wants to move first inverts the question: instead of “what can I automate?”, “what must I be able to prove?”. In practice: document where human oversight is effective; measure what share of automation savings returns to human capacity (training, new roles, wages); audit customer-facing agents against the Charter's duties; and leave that evidence in a form a third party —customer, auditor, regulator— can verify without taking the company's word for it.