that AI will be the death of learning & so on; to this, I say NO! My student brings me their essay, which has been written by AI, & I plug it into my grading AI, & we are free! While the 'learning' happens, our superego satisfied, we are free now to learn whatever we want
This is a quote allegedly by Žižek, which I stumbled across recently on Twitter, where it circulated heavily. I don’t know if it’s genuine. Who knows, it might be AI-generated; after all, it is a response to the uneasiness that the skills of the chatbot GPT-3 have produced in some academics who have found out that the AI can produce better course essays than what they would reasonably expect from their students (and within seconds). Students will be deceiving us; and machines will replace us anyway!
Maybe. Clearly, machines are better and better at doing things we have not imagined them to be able to do, and faster and faster. That is just a matter of time. However, so much of what we do has the importance it has (in fact is the thing we do) because it is we who does it. AI can write a course essay for a student, but what it cannot do for the student is learning the supposed lesson from the essay. That is, the AI itself perhaps can learn from it (I have no idea), but it cannot make it happen that the student knows or has learned. To an extent, it will still be up to our students whether they want to learn something or whether they want to deceive us (and I am far from arguing that there is a universally valid answer that one or the other is per se the more right thing to do – sometimes we learn best by deceiving). Just as it has always been; here, AI is not a revolution, it plays a similar role to that of schoolmates others used to pay for writing their course essays for them (a very pre-AI phenomenon). Admittedly, AI is vastly more efficient and more democratic in approach.
Some say that very soon, artificial entities will be able to perform any cognitive task better than people. Perhaps so. And yet… It is now clear that computers can beat even the best human chess or go players. And, surprisingly, people still play chess and go – not only for fun as amateurs, but also professionally and still get their nice prize monies. (Perhaps a specially constructed running machine could run faster than Usain Bolt – just as cheetahs and ostriches can – and yet we are still having athletics competitions…) As I said earlier, the meaning of “some things (that we do) being done well” seems to involve the importance of the “that we do (them)”. Being a good husband to my wife or a good parent to my child is certainly a cognitive task too, to an extent (which is one of the reasons for my questionable performance). Could a machine do it better than me? In a way, it could, absolutely, but at the same time, it would not be doing this task anymore. This can change, too, but it’s not a matter of its performance, but of a shift in our form of life.
What machines will probably “take away from us” will be some of the jobs that people are doing nowadays. However, this is, I believe, not a problem for a philosophical analysis of the devilishness of AI, but a problem of political economy. The real villains might be the corporate moguls who would otherwise pocket the surplus value generated by automatisation – but there is nothing philosophically complex or interesting about their villainy. Rather than losing jobs in general, people may only lose the interesting jobs, while shitwork, as usual, is not going anywhere – to end by quoting another recent internet meme:
In important respects, the revolutionary threats that AIs keep bringing often turn out to be versions of something that we have known even before, to end (now truly) by quoting yet another meme, this time not so recent, but about the poignancy of which the advent of super-efficient AIs has changed pretty much nothing:
