I have been using Microsoft's Chat AI to find jobs that best suit my skill set, and so far it has been interesting. Based on the information available, I find it great for brainstorming an idea when you don't have your peers to talk with. I have only started looking at this in the past week, and a few days ago had email from my old uni offer an opptunity to do a fully funded Masters in AI. The qualification would be more as an AI Consultant being able to show have AI could be intergrated into a business. One thing I would say about AI, for good or bad it's here, just like computers and the web (Silken_thread) it's not going away, so we need to learn and understand it and how it can improve/enhance our lives. What is your experience of AI chat systems
Ive been highlighting some of the weekly use cases here https://discord.com/channels/368557034305486871/1092623524545429554 if you want to read on the EMC discord. I’ll say the AI Rundown newsletter is a good one to read weekly/daily I'll post more thoroughly in the weekend And to answer your question, it will enhance life, but it will change society as we know it once it is mass adopted as well as the need for regulation.
There's a video online where someone used AI (not sure if it's a plugin, or an application running in the background) where it took all the video footage they were working on in their video editor, and pieced it all together. It made jumpcuts, new layers, spaces, and so forth, based on the recorded footage. The person just sat back and watched the AI do all the work for them in realtime in front of them. This part of AI, I do not mind. I'm a bit curious where this will lead since I may want to get my hands on that and try that out for myself!
Way back in the 1970’s I spoke to software engineers who told me that no one person actually understood how their mainframe operating system worked as everyone knew only their own bit. And that’s not even remotely AI. While the likes Linux is at least understood if one takes the community as a whole because it’s open source not all these AI systems will be out in the open and watch out where they begin to re-write themselves. We already have the web filled with fake news and ChatGPT (I’ll use that to represent any of the LLMs) learning from that. We have students allegedly submitting essays written by ChatGPT. There was recently a case in the US of a lawyer submitting docs to court where they had used an AI and it created references to nonexistent caselaw - the lawyer apparently did not know the AI could lie. We are in danger of creating a web where nothing is actually correct but everyone believes what is written. (in fact we did this already regardless of AI!) I tried ChatGPT asking it to write some advertising for a thing and it did a good job. But when I asked it more in depth questions it produced wrong information. When challenged it tried again and again until finally it just agreed with what I told it. If I did not know it was wrong then would I have accepted the incorrect information? And there be dragons. Yes this is all just the LLM business and AI can / will do so much more. But puzzle me this. You generate an AI that can remodel a system to make it more efficient. It does a good job. Do you then unpick the system to find out how it works; or get the AI itself to document how it works; or just take it as read and never actually understand why whatever has been produced is better than what was there before? The other danger is there will be a perceived AI gap. They have one, we need one, let’s cobble one together / buy someone else’s. Hey let’s get their AI to create our AI. No, let’s not rush guys. As to regulation, well yes, but you just know that however it is regulated that only affects the guys obeying said regulations. There are various jurisdictions that will ignore any attempt to regulate anything so why do they care. AI gap… Your example of using it to find job matches is a good one because its a walled garden. You’d know if it offered you a hairdresser job (other professions are available) that its not a good match for you. It will be interesting to see where all this goes. As yes, news: ‘The US air force has denied it has conducted an AI simulation in which a drone decided to “kill” its operator to prevent it from interfering with its efforts to achieve its mission.’
I got my Bachelor's in AI, and am currently doing my Master's. Yesterday my Bachelor's study association had a reunion, and one of the talks was particularly interesting for me. We hear very often about this idea of 'AI winters', where there's little innovation, and public expectations are low. However, this speaker (Pieter den Hamer) distinguished all four seasons, based on four axes: there is an axis of innovation, going from very early stages to very advanced AI, and there is an axis of impact, going from negative to positive impact. Currently we are in an AI spring, meaning there is little innovation, but positive impact (this latter point is quite debatable, some people would argue AI currently has negative impact, but our speaker was optimistic). It is quite possible that developments stagnate, and soon we will find ourselves in another AI winter, where there is no positive impact and no innovation. If we find ways to develop AI further, while creating good regulations and ensuring positive impact, we might reach an AI summer. However, with current trends, it's also possible that we might reach new levels of innovation while also losing positive impact, and we'll reach an AI fall. Good post! This is in the wrong forum section, by the way. It should be in Miscellaneous, or in Controversial Topics if you want.
