What's new

"Generative AI is the greatest risk to the human creative class that has ever existed"

New law. Only one android per 3 humans in a live performance. (Speakers, recordings etc count as androids).
 
It’s probably worth linking my ex-colleague Bill Joy’s article from April 2000. For anyone who hasn’t read it, it created a stir at the time it was published and was considered controversial (enough so that our stock price was negatively affected). He makes the point that some technologies are potentially dangerous enough that an equal amount of resources should be devoted to developing risk mitigation - and we are not doing that.

Why the Future Doesn’t Need Us
 
Hello,

A full and very interesting study on the use of artificial intelligence in the music sector conducted by Goldmedia and commissionned by SACEM (french PRO) & GEMA (german PRO) has just been released. It's in English :)
We learn that:
- by 2028, 27% of music creators’ revenues will be at risk due to generative AI. This corresponds to an estimated potential damage of €950m in 2028 alone and a cumulative total damage for the period 2023-2028 of around €2.7 bn.
- the share of music applications is estimated at $300m in 2023. This corresponds to 8% of the total market for generative AI with already $3.7 bn in revenue in 2023. The market for AI music solutions is expected to increase more than tenfold by 2028, with an average annual growth rate of around 60% to over $3 bn for music AI alone.This means that in only a few years the market will reach a size that corresponds to 28% of global music copyright collections in 2022.
- 35% of of the 15k GEMA/SACEM surveyed for this study have used AI technologies of some kind in their work with music and creation in general (
I didn't except this percentage would be so high)
 
I'm far less worried about the creative aspects of AI than I am about the destructive aspects, like spreading misinformation via deep fake audio and visual information. We are already at a point where many people don't believe ANYTHING anymore—including cold, hard facts—and AI is bound to make this much, much worse.

From a creative aspect, I think AI is far less dangerous than the current hysteria over it suggests. And since I use FAKE INSTRUMENTS every day of my life, it would probably be hypocritical of me to complain about anything that replaces real people. It's okay to replace an orchestra and all of its players, but not okay to replace the guy who writes for it?

I, personally, use AI as a tool. It helps me write code for Reaper scripts when my own coding knowledge is rusty. So, like any new technology, there is good and bad in it. But people who code for Reaper for pay might not agree that this aspect of it is good.
 
Hello,

A full and very interesting study on the use of artificial intelligence in the music sector conducted by Goldmedia and commissionned by SACEM (french PRO) & GEMA (german PRO) has just been released. It's in English :)
We learn that:
- by 2028, 27% of music creators’ revenues will be at risk due to generative AI. This corresponds to an estimated potential damage of €950m in 2028 alone and a cumulative total damage for the period 2023-2028 of around €2.7 bn.
- the share of music applications is estimated at $300m in 2023. This corresponds to 8% of the total market for generative AI with already $3.7 bn in revenue in 2023. The market for AI music solutions is expected to increase more than tenfold by 2028, with an average annual growth rate of around 60% to over $3 bn for music AI alone.This means that in only a few years the market will reach a size that corresponds to 28% of global music copyright collections in 2022.
- 35% of of the 15k GEMA/SACEM surveyed for this study have used AI technologies of some kind in their work with music and creation in general (
I didn't except this percentage would be so high)
35% of 15,000 GEMA/SACEM composers have used AI in their music creation?

Maybe I’m living under a rock, but I literally don’t even know of any AI tools yet for composers. What are these people using or what is being defined as ‘AI technologies?’

All the figures in that study seem very alarming at face value. But if composers are using AI to create music how can anyone be surprised that the next logical step is to just get rid of the composer all together.
 
35% of 15,000 GEMA/SACEM composers have used AI in their music creation?

Maybe I’m living under a rock, but I literally don’t even know of any AI tools yet for composers. What are these people using or what is being defined as ‘AI technologies?’
Maybe they count something like Ozone?
 
Maybe I’m living under a rock, but I literally don’t even know of any AI tools yet for composers. What are these people using or what is being defined as ‘AI technologies?’
You can find all the details in the study!

To be honest, I think everybody should read the study. It's an interesting and full insight (192 pages) into what we can expect from the future.

Maybe they count something like Ozone?
Yes indeed (see Page 14)

But I think it's a mistake. As far as I know Ozone is not based on artificial intelligence.

I noticed that there's often a strong confusion induced by techs claiming retrospectively and falsely to use AI to look trendy while they just use algorithms (I like this definition found on Linkedin: "AI can change its outputs based on new inputs, while an algorithm will always generate the same output for a given input")

Creative people are not threatened by algorithms; creative people are threatened by AI because of its capacity to learn by itself from experience.
 
Last edited:
You can find all the details in the study!

To be honest, I think everybody should read the study. It's an interesting and full insight (192 pages) into what we can expect from the future.
192 pages? 😖 This will not tell us what to expect from the future.😣
 
Yet another reminder that stock investors are largely dumb animals that are easily fooled by smooth-talking charlatans.

All one has to do to see that AI is still largely in the pre-monetization phase outside of a few novelty applications (e.g. ChatGPT's monthly subscription for CGPT4/DALL-E) is to use it. It becomes very evident, very quickly that it's not world conquering, "game changing" technology yet. It will get there, but you're going to have to wait some time longer before the learning models hit the critical mass necessary to stake your financial future on it.

Then again, if you're dumb enough to heavily invest in stocks based on current AI output and expect returns any time soon, you're probably dumb enough to be suckered in by what ChatGPT does to think it's smarter than it really is.
 
Yet another reminder that stock investors are largely dumb animals that are easily fooled by smooth-talking charlatans.

All one has to do to see that AI is still largely in the pre-monetization phase outside of a few novelty applications (e.g. ChatGPT's monthly subscription for CGPT4/DALL-E) is to use it. It becomes very evident, very quickly that it's not world conquering, "game changing" technology yet. It will get there, but you're going to have to wait some time longer before the learning models hit the critical mass necessary to stake your financial future on it.

Then again, if you're dumb enough to heavily invest in stocks based on current AI output and expect returns any time soon, you're probably dumb enough to be suckered in by what ChatGPT does to think it's smarter than it really is.
I personally think the best stocks right now are chip making and gpu making stocks. If and when AI gets there, it will largely be due to the gpu/chipmakers producing enough for them. *not financial advice
 
Here’s an interesting article from the Hollywood reporter on AI art.


Visual art is kind of the front lines in the regulation of ai art, these kinds of wins are what will ultimately define the environment we will be in once it could do the same with music.
 
Top Bottom