Пророчество Джобса: как «потерянное интервью» описывает наш 2024-й

Picture 1995. Steve Jobs is in a kind of creative exile: Apple threw him out the door, NeXT is struggling, and Pixar hasn't yet released Toy Story. At this moment, he sits down in front of the camera and spends an hour laying down the foundation that, thirty years later, seems not just a prediction, but a detailed blueprint for what's happening right now in Silicon Valley. This "lost interview" was accidentally found in a garage years later, and today it's worth revisiting for anyone trying to understand where large language models are taking us.

Jobs touched on a concept that has become the foundation of the AI revolution today: a colossal productivity gap. In the ordinary world, the best taxi driver drives 20% better than average. But in intellectual work, according to Steve, the difference between an average specialist and an "A-class player" is 50 or 100 to 1.

This used to seem like hyperbole, but look at modern engineers using Cursor or GitHub Copilot. One person with the right stack of AI tools today is truly worth an entire department from the nineties. We're entering an era where individual mastery multiplied by algorithms devalues the very idea of massive headcount.

The most ironic and simultaneously frightening thing in Jobs' words is his diagnosis of corporate death. He explained how companies turn into zombies: when they become monopolies, the marketing and sales department starts steering the process, while the people creating the product get pushed to the background. In the end, "process" replaces "results."

If you look at how sluggish giants try to catch up with OpenAI, you'll see exactly this picture. They spend months on ethics committees and font alignment approvals while small teams release models that change the world. AI simply highlighted this rotten foundation, making inefficiency too expensive a luxury.

Jobs also predicted the emergence of what we now call agents or personal assistants. He spoke of software that would be not just a tool, but a "reflection of our thoughts." In 1995 this sounded like science fiction, but today GPT-4o and Claude 3.5 do exactly that—they mimic our thinking style, complement it, and allow us to delegate cognitive load. Steve understood that a computer is a "bicycle for the mind," but he clearly sensed that soon this bicycle would acquire a jet engine and autopilot.

Special attention should be paid to his "death sentence" for middle management. In a world where information is transmitted instantly and AI can coordinate tasks, a layer of people whose only job is to pass papers from top to bottom becomes ballast. Jobs believed in flat structures where everyone understands the essence of the product. Today we see startups of three people valued at billions of dollars. This is the realization of his dream of maximum concentration of talent without bureaucratic impurities. AI here acts not as a killer of professions, but as a powerful filter, washing away those used to hiding behind processes.

Rewatching these footage, you understand that we're not inventing anything radically new in terms of business philosophy. We're simply finally getting the technologies that allow us to realize ideas from thirty years ago. Jobs didn't know the words "transformer" or "diffusion model," but he certainly knew that the one who could close the distance between idea and implementation would win. Now that distance has shrunk to a single prompt, and it's damn inspiring, though a bit frightening.

The key point: those who win are not those with more graphics cards, but those who have preserved the ability to see the essence behind the process. Are you ready to become that very "100-to-1 player," or is your role in the company just bureaucratic noise that AI will soon zero out?