2021-06-26

tkgshn: hmmm…is there a theory that with machine learning becoming the default (assuming the notion of productive capacity becoming software is the default), the competitiveness of companies is swinging back to a layer of capital, rather than knowledge worker and so on? Is there a theory that the competitiveness of companies is swinging back to the capital layer, instead of knowledge worker and so on?

tkgshn: this isn’t structured like the one mentioned in “postcapitalist society”, is it? Have we passed it by now? This is a tweet to say.

nishio: yes there is. Machine learning seems to be an equipment industry by huge capital. Drucker is not an esper, so he could not have foreseen this.

tkgshn: I guess - I was trying to make do with my knowledge until now, but I feel like I’ve been shaking things back up with all the ML spinning and brute force, you know..,

nishio:spiral staircase: like spiral staircase, we’re not back to where we started, but one step up. In the past, Google researchers would not have been able to publish papers, nor would they have been able to publish their learned models. Once the world went through a human resource market dominated by knowledge workers, the equilibrium point moved because “if you prohibit the publication of results, you can’t hire the best researchers”.

nishio: maybe this is because information science is a cross-disciplinary discipline. In the past, the knowledge produced by corporate laboratories was tightly coupled to the business, and disclosing it would only be a foolish act that would benefit competitors. Nowadays, information disclosure is a public interest action because it provides benefits far and wide outside the business domain. - open innovation


This page is auto-translated from /nishio/機械学習は装置産業への揺り戻しか? using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.