My Notes About AI (as a wannabe AI Engineer).

This was a comment I randomly made on Reddit but I may as well retrofit it as content given that Reddit hates me and I have zero clue if it even posted.

Coming from a liberal arts/art background, I used to feel scared about generative AI but, after getting my MS in Data Science, I know it’s just matrix multiplication. No sentience. As objective as can be when you discount human bias in the training data.

I also think every group debating AI concurrently has misconceptions and good points.

• Those who think generative AI job replacement will occur don’t know their history but, at the same time, are right to be concerned about what work will look like.
• Those who are explicitly anti-generative AI don’t understand that human accountability comes first, whether it be for algorithmic bias or for curating how to integrate generative AI outputs into their work. However, they also make great points about why we should be concerned about generative AI’s impact on society.
• Those who think that generative AI will be the panacea to the world’s problems are pretty fanatical and brutally ignore current realities of the job market, but do bring the positive side of AI to the forefront.
• Overall, cautious optimism and the recognition that generative AI will be shaped by informed humans is the way that we should move forward. I can’t stand listening to gimmicks or fearmongering about AI, and I especially don’t agree with the premise that AI will just disappear (that is not how technology works). I can only listen to suggestions for how to build and implement generative AI models and tools safely and ethically.
• Please stop saying AI when you actually mean generative AI. AI literally includes basic classification models.

Personally, I use generative AI for a lot of things, but usually for helpful suggestions to guide my thinking (not replace it). I also use it to build things with code that I never could have imagined building just by curating prompts that build and refine tools, which has inadvertently made me want to learn how to engineer software even more (I’m starting another MS in SWE soon). I actually enjoy reading the “Thinking” that Claude Code, Perplexity, etc. provides since it helps me with refining my instructions and to understand how I should maybe be thinking about putting the code together.

I think AI literacy is important in the same way that media and data literacy are as well. You don’t need to use explicit AI tools to learn this given that AI algorithms are embedded in all major search engines and social media platforms (whether they state it outright or not). For instance, this comment could be way shorter, but I love writing and processing my thoughts myself, so I don’t use generative AI to shorten it (apologies).

I rarely use generative AI to create art, though I generally believe that it’s better to exercise your creative juices than not (whether you choose to generate images using generative AI or create art manually). Read this study from Swansea University on how AI may positively impact creativity.

Whenever someone tells you that generative AI is going to do catastrophic things, such as destroy humanity, this is generative AI hype that generative AI companies want you to believe in order to scare ordinary people into submission and to gain more shareholder profit. Please understand that this is utter bullshit. I genuinely want to see the downfall of these profiteering generative AI companies. Also, note that Steve Bannon literally supports these initiatives. That should give you a whole clue as to who this ideology is for.

Overall, I am not an “AI-pilled” tech bro, I just believe in epistemic integrity in how we approach generative AI. Furthermore, in the same way that we can change governance and society for the better, we can still change the destiny of how generative AI interfaces with and helps humans. Losing hope now only lets the capitalists take the win when open-source generative AI is still possible in the same way that Bluesky and Mastodon have shown that big tech doesn’t need to hold the only options for social media.