The Genuine Intelligence
AI looks promising, but I prefer to see beyond the noise today
The bigger question recently is whether AI will replace us. It might be. It might not. It’s a little early to predict, and with no expertise in it, I can just have my opinion.
My take is that it will be part of our lives, but it will be like just any technological shift in the past, like the advent of internet or mobile devices. How deep it’s going to be integrated, that’s still a question I have in mind.
Here’s my other thought: adoption.
It’s very easy now to be optimistic or pessimistic, but it’s all based on the noise we see online, in the tech community, and in some ways, in developed world. Most of the people I follow on Twitter talk about AI, but again those who talk about it are mostly engineers or content creators.
If it’s going to be a big shift, it would need to be inclusive and accessible. It needs to matter in:
More economies — it needs to be beneficial not only in the developed world, but also developing and underdeveloped worlds. It needs to be used by high-tech companies, but it needs to be beneficial for the grassroots communities, and actually enhance their lives. Will it actually be used by moms in the villages? Will street vendors benefit from it? Or is it something that will just enhance capitalism?
Respecting cultural and social context — sure, it does support multiple languages. I can ask ChatGPT in Indonesian and it will feed me back in Indonesian. I am impressed. But will it understand the true Indonesian context? Not sure. Will it then understand cultural, social and religious nuances? Not sure.
The cultural context reminds me of this talk by Adri Reksodipoetro, who ran research for Google to improve search adoption in their Next Billion Users.
He went into one of the kampungs in Indonesia to understand how Google search is used by just any regular mother living in these communities. The mother was prompted to search on Google on fever medication for her child. Obviously, the mother never used Google search, and the way that she prompted Google was through voiceover, “My kid has been sick for a few days, I have given her X, what medicine should I give her?”
Obviously, this didn’t work for Google, as we usually just put few keywords like “fever medication kid” or something similar.
This brought a good reminder — the mother was actually prompting Google like we all engineers try to prompt ChatGPT:
Setting up context: “My kid has been sick for a few days… I have given her X”
Query: “What medicine should I give her?”
Of course, AI “prompt experts” would have added something like “Imagine you’re a doctor or pharmacist…” but really, this could be an entry point to make AI more accessible.
It is of course early days to demand inclusivity and accessibility, let alone adoption, but if we all think AI should be the next big thing, it better be a “big” thing for everyone and not just for capitals or privileged few.
It might also be possible that AI is not necessarily adopted by a wider audience, maybe it’s just another technological iteration. In this case, the hype now might fade away.
As a designer, should we be worried about it replacing our jobs? People say, it’s not AI that will replace our jobs, it’s the unwillingness to use AI to support our jobs that will. Fair point, but I honestly think it doesn’t mean that designers should actively learn prompt engineering now and follow the hype. Designers should always focus on the problem, the user and the context. AI is just a tool, and if it does help us focus on the problem, user and context, then by all means learn how to use it. Although I foresee that as we go forward AI will be seamlessly integrated in our workflow anyway, through apps that we already use (e.g. Figma, or anything in the future), or through methods (e.g. helping us analyse or affinity-map findings).