This is the question I try to discuss during my talk @LibConf
Let the machines do the routine and brute work, we think. Sounds like a reasonable argument. But machines are not capable of replacing humans in every aspect of work. In some cases (like fighting Covid19, it will be nice if machines can completely take over, but they can’t). They can do parts, but not the whole.
Machines are good at automating some amount of grunt work. So they may help automate certain tasks in a job. Here is a simple table that describes how certain Product Research tasks can be automated.
|Process||How Technology can help|
|Gather Information||Programmable Searches, APIs, Web Scrapers, Alert Engines, RPA (robotic process automation tools)|
|Extract Keywords, Topics||Natural Language Processing Tools and Services from various vendors, Language Models|
|Extract Entities||Natural Language Processing Tools and Services, Semantic Parsers|
|Miscellaneous Tasks||Clustering, Similarity Metrics|
While technology is good at some of the simpler tasks, they can do only a fraction of the job of the researcher. This is because:
- NLP and other AI tools are still in their infancy
- They (entity extractors) do not have a good context and get confused between products, people and places.
- They are trained with very generic data
That is where we (humans) come in. Technology does the grunt work. Humans add insights. Humans can train machines to improve. But the way you train machines is very different from the way you train people.
As AI improves, humans will improve the way they use AI, too.