And we can all be like "noone is owed a job doing traditional art, this happened to millions of people during the industrial revolution too", but in that case we have to be honest and not tell art-folks that they are overreacting.
AI graphic/video art hasn't found its Marley Marl yet, but I believe she's out there and is going to do stuff that will blow our minds.
"Prompt engineering" exercise generally involves a few lines of text. It's hard to see how that's going to result in a new kind of highly skilled artist. AI produce the same level of greatness and/or mediocrity for essentially all of their users.
This thread is like fifty variations of "it will work out (waves hand at past)"
Synthesizer instruments, yes. Sequencers, trackers, and modern DAWs like Ableton, absolutely not. They’re entirely different animals and are very much a musical analogue to “prompt engineering.” They have also enabled some incredible art that would be impossible if musicians were limited only to what they could play into a recorded (or midi) tape.
I suspect that the same will prove to be true of generative AI, and that those artists that embrace it will see a golden age of creativity. But there will also be many artists who see their craft diminished in value, much like what has happened to studio drummers. And I could be completely wrong, but either way the genie is out of the bottle, so we’re going to find out.
These kinds of tools are democratizing visual expression; with that, we should place even more importance on what is being told and expressed. In some ways I see it as a new form of writing. We're in our duchamp readymade era.
It had never entered into my thinking that the code I was writing, the screens I was designing, and the system I deployed would do that. Ever since that project, I think about that. You have to think not just about what the software will do for the people you intend to use it, but what it will do to them as well.
With our headlong rush to deploy more generative capacity, there's zero consideration for this. The reasons we're ignoring this are understandable, if sad. There's too much money to be made being the first or best on the block to deploy this capability. There's too much potential control if you're the winner of this race to stop. The downsides will be felt by others, not the ones deploying the software... until they put some of their friends out of work, and even then, they won't look up and notice until it's too late.
The main thing to do would be to give people a heads up so they can try to get out in front of it. Which is something AI companies are doing at every level.
The message I keep hearing time and time again is "Our aim is not to replace but to empower". The same companies with this message are also saying that their goal is to create a superintelligence that does everything better than humans.
As a junior developer, I have no idea what I should do, where to pivot, and what to expect in the next 6-12 months. I'm confused and lost. I don't know which skills I should be picking up and which ones will be worthless. All I know is that just a few months ago, when I expressed this concern, I was told by seniors that I, a clueless junior, have nothing to worry about, that it's just hype, that I'll never be replaced. Now the general sentiment is that juniors are indeed doomed. The sudden shift in tone felt like a slap in the face.
Learn to use the new tools. And become an expert in the business domain where you work. Developers who understand business issues are far more valuable than those who can only code to detailed specifications written by Product Owners or Business Analysts. Are you reading the trade press articles? Attending non-technical industry webinars? Networking with colleagues outside the engineering team?
Ignore these hyped claims by AI entrepreneurs. These are made to hustle money from investors. :-)
Comparison with past inventions do not take into account a few things:
1- AI may create new jobs, but it can also do those new jobs especially if they are digital. This is something that never happened before in human history. An easily available machine that can do what a human can do anywhere any time.
2- Cost and capitalism is a huge issue. If the cost of AI is the same as a human, but it can do the job 24/7 with no holidays, no sick days, no unions, and can execute and follow the whims of the prompter, then human will go for AI always.
3- Barrier to entry into any affected job will be low. People with minimal training or expertise will enter industries traditionally reliant on human expertise. This trend may lead to a proliferation of individuals claiming expertise in fields such as visual effects (VFX), which will blur the distinction between skilled professionals and amateur practitioners.
There is nothing like that for AI. The fact that something thematically similar happened in the past does not mean it's applicable here. Now, there are no new supply chains to build, this silicon's getting manufactured one way or another. No new jobs, just replace the department with an individual.
But I'm in my 30s, which means on average I have another 40-50 years of existence to get through.
The future contains AIs like SORA and Gemini 1.5. Gripe about it if you want - but that won't change anything. So why be grim about it? There are so many opportunities to use these new tools to make new things that have never existed before. Obviously there's a danger society itself crumbles, but if not - well, what an exciting transition to live through.
I look forward to being an old man and getting to tell the stories of what life was like when movies and software were made entirely by hand. What fun.
where's the hope in that? especially because it could be taken away at any time.
In terms of careers, children have years where they are expected to be unproductive and old people are no longer expected to be productive.
The ones hit hard are those who need to start from the bottom again learning a different skillset, and many have gambled putting off drawing an income for years for what they thought was a moat that AI is chewing through rapidly.
Potentially the future might be great, we don't have jobs... we have hobbies. Think Sisko's dad from DS9, replicators can create anything for everyone... but he runs a restaurant with handmade food because that's his passion.
But we might end up with a dystopia with corporate government, centrally controlled resources probably metered based on a social credit score.
yes, as I see it this will go one of five ways, ordered by probability
0) it's all hype: AI plateaus/fizzles out, maybe minor productivity improvements
1) techno-feudalism: ruling class of AI company trillionaires exercise complete domination of 99.9999% of humanity
2) new dark age: billions of newly unemployed formerly middle class workers riot worldwide targeting DCs and fabs. AI CEOs/VCs suffer the fate of the French kings
3) techno-utopia (Earth in Star Trek)
4) AI takeover
personally I'm hoping for 0
It feels like it's too late to switch plans and careers (and even if I do, what would that be?). At the same time I have another 30-40 years to live, a family, 20 more years of mortgage etc. so the technology has real consequences for a long time of my life, I can't get excited about it.
But that's what we use and it's not getting better. It's getting worse. It's not just OpenAI either, but that's a fairly prominent example that's germane here.
With that attitude, you're sure to live a very long time.
“I am 68 years old, on the opposite spectrum of life’s journey. At your age, I had the same passion to be a writer. But when personal computers became available just a few years later, that passion changed to becoming a computer programmer. I’ve been doing that ever since.
My point is:
Don’t lose your passion. Learn everything you can about this new technology and figure out how to use it to make your dream come true.
Be open to life’s changes. It may seem like nothing will ever change your mind, but no one knows the future. People change, the world changes, and as unbelievable as it may sound, you may one day become passionate about something else.
When Sam Altman announced the new GPTs and Assistants technologies, everybody who had been developing RAG applications for the previous year freaked out. I didn’t. I, as a RAG (sometimes called “Wrapper”) developer saw it as a unique opportunity. And so far, I haven’t been wrong.
So, SORA might just be your opportunity. Use it to CREATE and show that love in every piece you create.”