I am not an AI hater like some people. I work with it everyday. Not only because I have to, but because it does actually make my life easier.
For example, sometimes if I am having difficulties replying to an email I will use AI to help me. Often because I am unable to respond in a professional matter. “Do your damn job, Becky” is not really suitable for work but, “As per my last email…” Is a little more…diplomatic. I also use it as a brain storming tool, to help provide a starting place or jumping off point for other work.
That said, I completely understand why people — especially creatives — have turned against AI. Designers, illustrators, and writers didn’t consent to having their work scraped and fed into these systems. Anthropic, OpenAI, and others did it anyway. So yes, people are pushing back, and they have every right to.
In fact I recently played a great video game, the Long Dark, which starts with a strong statement against the use of generative AI.
I am in a fortunate position where I can see some of the benefits of AI. For example, I learned of a project where AI was reviewing satellite photos to look for ‘ghost nets‘ and then reporting the location to boats which would then go and pick them up. Or advancing cancer research with the use of AI.
However there is something to be said about the exploitation of the generative AI models. Not only that, but we know that AI is going to increase productivity exponentially. So it is imperative that we enact legislation now to protect peoples’ copyrights and ensure that everyone benefits from the use of AI. If I am twice as productive, shouldn’t I be paid twice as much? The company has already worked profit into the calculation. I am doubling the profit, in fact more than doubling since they aren’t doubling my benefits or time off.
It seems to me that most technical revolutions put the benefits in the hands of a few and the costs in the hands of many and AI is no exception.
Corporation, n. An ingenious device for obtaining individual profit without individual responsibility. ~ Ambrose Bierce
AI has a serious environmental impact, and the companies profiting from it should foot the bill — not taxpayers, not local communities, and not the planet. They created the problem, they can pay to fix it.
The scale of this problem is staggering. Data centers powering AI models consume enormous amounts of electricity and water. Training a single large language model can emit as much carbon as several transatlantic flights. And that’s before you factor in the ongoing cost of running these models at scale, billions of queries a day, every day. Microsoft, Google, and Amazon have all seen their carbon emissions increase since making major AI investments — in direct contradiction to their own stated environmental commitments.
So what do we do about it? A few things come to mind:
- Mandatory green energy investment. If you want to run a data center, you fund the renewable energy infrastructure to power it. Full stop. No offsets, no carbon credits — actual clean energy.
- Punitive electricity pricing. AI companies should pay a premium rate for grid electricity, the kind of rate that makes frivolous use economically painful. Want to generate a thousand AI images of cartoon cats? Fine — but it should cost you something real.
- Transparency requirements. Companies should be legally required to disclose the energy and water consumption of their models. Right now that data is largely hidden. Sunlight is the best disinfectant.
- Environmental impact fees. Similar to how polluting industries pay into environmental remediation funds, AI companies should contribute to a fund dedicated to offsetting their footprint.
Sure, some people will say this kills innovation. I say: good. Not every innovation deserves to be cheap. If companies had to charge the real cost of generating a throwaway AI graphic or a spam email campaign, a lot of people would decide it isn’t worth it. That’s not a problem — that’s exactly how a healthy market should work.
The companies building these tools are extraordinarily profitable. They can afford to do this right. The question is whether we have the political will to make them.
the unfortunate reality for intellectual property holders is that the legal system will be basically powerless when it comes to stopping ai technology from digesting and mutating and spitting out other people’s IP, for the same reason that it is powerless at stopping filesharing technology from being used to copy and distribute other people’s IP. unlicensed copies used to be a rare local phenomenon that could be traced back to a central source, then came personal computers with modems and cheap digital storage and suddenly anyone could make infinite copies of things at no cost share them with millions of people all over the world, enforcement of copyright law suddenly became a mathematical impossibility.
now with the advent of ai, it will not only be easy to copy and share things other people have made, it will also be easy to reverse-engineer and mutate and recombine and reproduce things other people have made – authorship itself will no longer be a localized rarity that can be traced back to a single original author. machine-learning will be used to analyze the publicly-visible parts of things and use predictive algorithms and vast libraries of training data and brute force trial-and-error to figure out the missing pieces of the puzzle with inhuman speed, source code and manufacturing processes and other trade secrets will become public knowledge and all this information will then be incorporated as training data and fed into more machines that blend it all up with everyone else’s ideas and spit out seemingly-original works that are impossible to recognize as unlicensed clones/amalgams of other people’s work. not only will copyright law be unenforceable, patent law will also be unenforceable, there’s not enough courts and judges in the world to deal with this kind of caseload.
in my opinion, intellectual property was always kind of a bullshit idea to begin with – the idea that information could be a form of private property. it was only because of archaic technological limitations that we were ever able to treat information as private property but as technology evolves and information becomes easier and faster to read/write/copy/modify/transmit, it becomes clear that the idea of intellectual property makes unrealistic demands of reality and reality is not going to accomodate them forever. it will only ever become easier to do things with other people’s ideas and there’s not really anything that can be done about that other than think about ways to adapt our thinking to a world without intellectual property, a a world where information, knowledge, ideas, creative works, all these things instantly become part of the public domain the moment they are exposed to the public.