Hope and Hallucinations

What a challenging week it has been! There was a very difficult technical problem which I inherited but I was relieved that I managed to find a solution for it. I've always practiced gratitude in everything I do, and I am grateful this time that I did not run out of ideas to pursue. By getting the ego out of the way, ideas will always flow. That I must always remember.

Today is my chance to catch my breath a little and relax with some quiet written words. I see a lot of people getting all worked up by news and events that are happening in the world. I see everything with a bit a equanimity, listening to both sides of the story. I do not have any desire to indulge in virtue signalling. Having built a thriving social media service before, I have no desire to dwell in that world anymore. 

I'm still that old-fashioned bloke typing text into an outdated blogging platform, something I've been doing for 20 over years now. This blog is a record of my intellectual development over the last two decades. When I first started blogging in Jakarta, I did it mostly from an Internet cafe which I frequent on a daily basis because the only other alternative was using a slow modem dialup from my hotel room, which I did occasionally.

What happy days those were, living alone in the heart of Jakarta, within walking distance of good street food along Jalan Sabang, a good bookstore, the now defunct QBWorld, and a grand cinema, Theatre Jakarta.  Food, books and movies are what I live on. Can't think of anything more in life that I need!

I am grateful for what I have and had. I treat every experience in life as an opportunity to learn, as another set of data to improve my natural intelligence model. We learn the same way as those AI LLMs that are changing every aspect of our lives. People say that these LLMs are still imperfect because they have a tendency to hallucinate. But they forget, people also hallucinate all the time!  How many times have we encountered people who speak with false certainty about subjects that they have scant knowledge of? Isn't our susceptibility to conspiracy theories a form of hallucination too?

At least ChatGPT has the decency to apologise and correct itself whenever you point out its errors. How many times have you seen people doing that? If only everyone has the same balanced and measured response on every subject that ChatGPT has, the world would be a much better place!

The only difference between my brain and these LLMs now is that, mine lives in a body that moves around in the world, constantly absorbing new data and input. Whenever someone talks to me, I do not only receive word prompts, I also get visual, auditory, tactile and chemical ones. Even with these advantages,  I still cannot come up with better answers than ChatGPT. And I am often wrong too. So let's have the humility to accept that we too 'hallucinate'.

It is important that we feed our brains with good quality information, like how we try to consume nutritious food for health. Today, we are lucky that quality content is cheaply available everywhere. It is up to us to feed ourselves wisely and improve our own natural intelligence models. ChatGPT and other LLMs will be continuously improving over time. AI models are the ingenious products of human intelligence, which ends up augmenting our own in return. 

Our future depends on how well we assimilate new technologies into our lives, for the betterment of humankind. And key to that is the ability to break out of bad thinking habits, superstitious dogmas and our primal tendencies towards violence, territoriality and sexual dominance. Yes, there's still hope for the world if only we humans know how to break out of our mass hallucinations.