Your observation "More people will start to use AI, but the less they know, the harder it will be for them to critically assess the quality or originality of the output" is something I have been struggling with for the past year ... using genAI shifts the user's role away from creating to evaluating.
But that presupposes that the user has the knowledge/skill to evaluate.
This I think is a real concern for people who haven't gone through the painstaking process of learning a skill. There are no free lunches. A key challenge for education and L&D is not to simply teach users how to use genAI but also how to hone and develop skills without genAI so that they can properly evaluate and "critically assess" the generated output.
I am just not sure how this can be done in an expedient way.
1. By default, yes. But just like some people prefer to use maps nowadays, there is nothing that can prevent individuals from refusing to be dependent on it (well, mundane utility and social pressure sure provides incentives though).
2. Debatable, though it's hard to argue no as a general trend. I'm a student in computability, and using AI won't make me less of one.
3. Technically yes, but I don't necessarily see it as a problem. Compare with losing the know-how of knitting and barrel-making.
4. Debatable, but I would lean no. AI can be used for human augmentation. See cyborgism, AIS by debate, IDA, etc. It's unclear whether it will negatively affect society at large though I would lean yes.
5. Depends on the subject, but for most mundane practical applications, the answer is clearly no for now because AI raises the floor of competence more than the ceiling. It might or might not change as AI becomes superhuman at these tasks, but there is yet no clear trend in scientific opinions.
@Jurgen Gravestein Interesting. Just read the post by @Ben Dickson on research by UCLA on the impact of GenAI on creative writing.
Should complement yours.
Feel feee to link it below! Love to read it 🤝
You point to an important problem.
Your observation "More people will start to use AI, but the less they know, the harder it will be for them to critically assess the quality or originality of the output" is something I have been struggling with for the past year ... using genAI shifts the user's role away from creating to evaluating.
But that presupposes that the user has the knowledge/skill to evaluate.
This I think is a real concern for people who haven't gone through the painstaking process of learning a skill. There are no free lunches. A key challenge for education and L&D is not to simply teach users how to use genAI but also how to hone and develop skills without genAI so that they can properly evaluate and "critically assess" the generated output.
I am just not sure how this can be done in an expedient way.
1. By default, yes. But just like some people prefer to use maps nowadays, there is nothing that can prevent individuals from refusing to be dependent on it (well, mundane utility and social pressure sure provides incentives though).
2. Debatable, though it's hard to argue no as a general trend. I'm a student in computability, and using AI won't make me less of one.
3. Technically yes, but I don't necessarily see it as a problem. Compare with losing the know-how of knitting and barrel-making.
4. Debatable, but I would lean no. AI can be used for human augmentation. See cyborgism, AIS by debate, IDA, etc. It's unclear whether it will negatively affect society at large though I would lean yes.
5. Depends on the subject, but for most mundane practical applications, the answer is clearly no for now because AI raises the floor of competence more than the ceiling. It might or might not change as AI becomes superhuman at these tasks, but there is yet no clear trend in scientific opinions.
I do not have any better answer to this question than you have already provided, but I have a few questions:
1. Are we going to become overly dependent on AI, like we are on GPS and calculators?
2. Are we going to have superficial knowledge, as we may not fully understand the underlying principles?
3. As we rely more on it, will we stop building the tacit knowledge that only comes with experience?
4. Will our overreliance on AI impact our ability to think critically?
5. Does the gap between less experienced and experts widen in the long run if both use AI?
All these questions are related and probably have the same answer, ‘Yes,’ but I will look for your thoughts.
I’m afraid the answer on all of those is Yes.