TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Ask HN: Adjusting to a Future with Near-AGI

5 pointsby xarici_ishlerover 2 years ago
Dislaimer: my intent in writing this is not scaremongering. I&#x27;m sharing my view on this and want to see others&#x27; perspectives and advice.<p>Ever since trying out ChatGPT I&#x27;ve been getting increasingly anxious and even depressed. I have found my interaction with it to be more insightful, creative and knowledgeable than with the majority of people I&#x27;ve met. Even when it comes to software engineering, ChatGPT has shown more understanding, patience and insight than most of my colleagues (maybe that&#x27;s more descriptive of the people I&#x27;ve encountered throughout my career, YMMV). I&#x27;ve done &quot;pair programming&quot; with it with great success. It doesn&#x27;t matter whether it has consciousness or &quot;understanding&quot; (whatever that might be) of its input and output, the results it produces isn&#x27;t that different from that of a junior developer&#x27;s (again, from my experience).<p>Now, I don&#x27;t want to argue about ChatGPT&#x27;s merits and its future potential – I&#x27;m not an ML researcher and neither are probably the vast majority of people on HN. I&#x27;m genuinely convinced that there&#x27;s far more to come in the nearest future, and ChatGPT is just an inflection point on our path to AGI.<p>My question is: what action should I take, right now, an undergrad, a CRUD &quot;engineer&quot;, to prepare myself for the near future where I have to compete with not only my fellow flesh-and-bone software engineers, but now also a gigantic language model trained on the entire human body of knowledge by hundreds of smartest minds in the world? Should I run headlong into ML research and just try to join the minds behind the AGI of the future? Should I get into some esoteric niche software engineering area which is unlikely to be automated away in the near future (and if so, which)? Maybe I should quit software engineering altogether and aim for something hands-on, lab-based, like electrical engineering or physics research?<p>Thanks for any advice.

2 comments

thesuperbigfrogover 2 years ago
&gt;&gt; ChatGPT has shown more understanding, patience and insight than most of my colleagues<p>It is a computer program, not a person. It matches patterns and generates text, but it has no feelings.<p>It is no different than ELIZA (<a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;ELIZA" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;ELIZA</a>) or Clippy )<a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Office_Assistant" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Office_Assistant</a>), just more sophisticated.<p>&gt;&gt; what action should I take, right now, an undergrad, a CRUD &quot;engineer&quot;, to prepare myself for the near future<p>Stick to your studies and finish your degree. Take breaks and have fun. Make friends and spend time with them. The whole world is ahead of you. Don&#x27;t waste your time worrying about future unknowns and maybe what-ifs.
cloudkingover 2 years ago
ChatGPT is a tool that is as powerful as the mind that wields it.<p>I don&#x27;t think it would hurt to learn more about how these systems work, but I wouldn&#x27;t panic about it. There are many other advanced software tools out there that you may not know how to use,<p>Will the barrier to entry for these LLM powered tools be lower? Probably, but they still require someone with knowledge of the matter at hand to produce useful output. Someone who doesn&#x27;t know software development can write code with it, but it doesn&#x27;t mean they understand if the code works, if it&#x27;s optimal, how to deploy&#x2F;maintain&#x2F;debug it etc.<p>If the theory of &quot;AGI&quot; comes to fruition, then we probably won&#x27;t have to worry about working anyway since it will theoretically be able solve all the world&#x27;s problems. By that time we should be exploring deep space on the starship Enterprise and synthesizing anything we need :)