TThe AI Apprentice Model

The current debate about AI and jobs is mostly about whether AI replaces seniors or juniors. The framing is wrong. AI does not particularly replace either; it replaces the work that juniors used to do as part of becoming seniors. The first-draft document, the basic research, the entry-level analysis, the routine code, the simple letter, these were not just jobs. They were the apprenticeship through which juniors became professionals. AI does them faster and cheaper, and as a result, fewer juniors are being hired to do them. The juniors who would have grown into the seniors of 2035 are not being hired in 2026. This is the more serious problem, and it is barely being named.

A profession is not just its present practitioners. It is the pipeline that produces its next generation. If the pipeline breaks, the profession ages out in one generation. We have seen this before in trades that lost their apprenticeship infrastructure, the senior masters retired, and there was nobody competent enough to replace them. The work moved overseas, or it stopped being done at all, or it was performed badly by inexperienced people who never had the years of contextual learning that the apprenticeship would have provided.

AI is doing this to white-collar work right now. Slowly, quietly, the entry-level positions that used to be the start of professional life are being absorbed by tools. The senior practitioners are still there, still capable, still working. The class behind them, who would have been the seniors of the 2030s, is not being created at the rate it needs to be. Nobody is in charge of fixing this, and the fix is older than the problem.

¶What apprenticeship actually did

Apprenticeship, classically, did three things that no curriculum could replicate. First, it transferred tacit knowledge, the small judgements and intuitions that a master makes without thinking and that a junior absorbs by being in the room. Second, it built the senior-junior relationship that makes the rest of a career possible, the mentor who remembers you, the network that comes with belonging to a master's lineage. Third, it produced public confidence, clients trusted the apprentice because they trusted the master who trained them.

Modern professional formation has lost most of this. Universities replaced the first part badly, treating tacit knowledge as if it could be lectured. Hiring practices replaced the second part poorly, treating mentorship as optional. Credentialing systems replaced the third part loosely, with degrees that no longer carry the lineage they used to. The whole edifice has been creaking for thirty years. AI is the wind that may finally bring it down.

¶Why entry-level work mattered

The work juniors did was not, despite the popular framing, "grunt work" that was a waste of their talent. The simple document was the way you learned what good documents look like. The basic research was the way you learned what good research feels like. The routine analysis was the way you absorbed the structure of the field. Replace that work with AI and the junior is left with no on-ramp. The senior says "I would have done that in my first three years; that's how I learned." The junior in 2026 does not have that three years to learn from, because the work that taught it is now happening in milliseconds inside a model.

The wrong answer is to argue that the work must be preserved. It will not be. The work is going to AI, for good economic reasons, and that train has left the station. The right answer is to design new on-ramps that produce, in different ways, the tacit learning the old work used to produce. This requires deliberate effort. It will not happen by accident.

¶What an AI-era apprenticeship looks like

An AI-era apprenticeship is a hybrid. The junior still works alongside a senior, but the work is no longer the routine output that AI now produces. The work is the judgement: which of the AI's drafts is closer to right, why, and what would be the second pass? The senior shows the junior how to evaluate the AI's output the way the senior used to evaluate the junior's output. The tacit knowledge transfer continues, in a different register.

This requires the senior to spend time with the junior even though the senior no longer "needs" the junior's labour for production. This is a hard ask. Most senior professionals have stopped doing it because the immediate productivity gain is gone. The country has to make explicit, structural arrangements that re-establish the senior-junior pair, paid for by somebody, because the market alone will not do it.

The arrangements could look many ways. Public funding for apprenticeship slots in critical professions. Tax incentives for firms that commit to junior development at fixed ratios. Professional community standards that require seniors to take on apprentices as a condition of continued certification. Each of these is unglamorous policy. All of them are infrastructure for the next professional generation.

¶The Indian opportunity is to do this on purpose

India has the unusual fortune of recognizing this problem early enough to design around it. The country has, by some measures, the largest population of young, ambitious, AI-curious workers in the world. If those workers are absorbed into structured apprenticeships rather than thrown into a job market that has stopped hiring at entry level, India produces, by 2035, the largest skilled-AI workforce in the world. If they are not, the country wastes a generation and the workforce ends up replaced by software.

The community is, again, the natural unit. A profession's apprenticeship cannot be coordinated by individual firms; firms have the wrong incentives. It has to be coordinated by the professional community, the doctors, the lawyers, the engineers, the teachers, the civil servants, operating as a sector. The community sets the apprenticeship standards. The firms participate because their seniors are members. The juniors apprentice across firms, not just within a single one. The whole structure is older than the modern firm. It is what professions did before the firm became the unit of work.

¶What to do this year

If you are a senior in any field, take on one junior intentionally this year. Not for the productivity gain. For the judgement transfer. Show them how you evaluate AI output. Tell them why you discard the drafts you discard. Sit with them through three real decisions. The hours are real. The value to the profession is large.

If you are a junior in any field, apprentice yourself deliberately. Find the senior whose judgement you want to absorb. Offer to help in ways that have nothing to do with model output. Earn the right to be in the room when the hard decisions are made. The market will not give you this. You have to construct it.

Bharath.CLUB exists, in part, to make this construction easier. The community is the natural matchmaker between mid-career seniors who would mentor and early-career juniors who would apprentice. The chapters are the natural locations for the pairings. The tables are the natural rooms in which tacit knowledge moves. The apprenticeship that AI has disrupted has to be rebuilt. The community is where the rebuilding happens.