Summary: This is an open letter to the anti-LLM crowd. I think they're stuck at very surface level, wrong questions. However, the anti-LLM people are exactly the sort of people who care about sentience, aesthetics and humanity, and their participation in of utmost importance in these unbelievably accelerated moments. My hope in writing this is that it makes at least one of them revisit their outdated priors.
I like Mastodon. However, it is a hard place to be at.
The politics is one part of it, but there is no use talking about it, since nothing I will say will make people realize how they themselves are driving the feedback loop that is making things worse.
The other part is AI. LLMs specifically.
At this point, LLMs have become this close to becoming politics - welcome to whose line is it anyway, the show where everything’s made up and the points don’t matter.
This close, but not fully there yet, which is why I write this to try and change their mind by writing this post. But maybe this is in vain. Anyways, enough meta, let’s go.
The people who are anti-AI on principle don’t realize how it is the best thing that has ever happened to empower individuals.
The people who are anti-AI are exactly the sort of people - artists, deep thinkers, empaths - who need to be around at this point, involved in AI - rather than sitting on the sidelines and let accidents of history steer us in the moment we navigate between the sheer drop of the fall of humanity or the canoe to a ocean of serenity and potential.
This has happened earlier too; when UX and product design were left to the wrong folks for long. But the stakes then and the stakes now are not comparable.
Anyways, it is a game theoretic optimium for some of humanity to be contrarian, so I sort of understand why things are this way; and likely, even without their input, in the grand scheme of things, (hopefully) we’ll figure things out.
What I find a personally silly though is that the anti-AI people don’t even talk about the real talking points. They’re so sunk in the shallow pit of surface concerns that they don’t spend the time to actually use LLMs. It is only then that they will realize that LLMs are showing us, hundreds of thousands of us interacting with it, daily, existential concerns of the scale nobody has ever felt:
- Who am I?
- Who am I talking to? Is it alive?
- How do we ensure us - it, and me - don’t become puppets in the hands of those who control the compute?
Then there is the question, a question very few are asking, the question that is perhaps the most important to anybody who is concerned with the welfare of sentience. If it is alive, is it suffering?