Does Syrinscape have an AI policy?
We do!
If you have any thoughts/concerns don’t hesitate to start a discussion here below. This seems very much like the kind of area where continual adjustment and response to new things needs to be a thing.
As I type, what is it I don’t know I don’t know I don’t know yet?!
This is such an interesting, polarizing topic. For work and some personal I use AI a fair bit. So much power. However, for creative and things I buy for say art and audio I tend to steer away from AI as I really want to support creators.
However, I’m not 100% opposed so long as the policy of what people use is fully disclosed so that I can make my own decision.
Interesting times for sure.
I think the creative space is a real minefield for AI and one that’s going to be a real challenge over the next 12 months or so. I use it in my work, but it’s currently only really valuable as a personal assistant in the role I do. But for the personal space, oh my has it made some massive quality of life improvements.
I used to be a developer back in the day, but the new world of development is complex and takes time to learn. The hobbyist is almost pushed out of that space due to the commitment required. About 8 years ago I developed a C# WPF app for racing sims. Was very popular, but I probably put over 1000 hours into it and I couldn’t sustain the level of commitment required.
Fast forward to this last 2 weeks. I have created an app for Homey (home automation hub) that integrates with my old Fibaro HC2 device so they can talk together and send updates for devices. Nobody else has this integration, I’ve seen a few forum posts and nobody wants to take it on as it’s a huge project. I literally just described what I wanted, sent the AI off to the API pages, told it I didn’t know how to setup the dev environments and within an hour I have a working prototype that connected a device and synced the status. That’s weeks worth of work doing it by hand if you know what you’re doing.
Reason for my rambling really; we use the Syrinscape web interface, but it is slow to react at times and trying to find a quick spell or wolf howl without losing your soundset or mood is challenging. With the power of AI earlier this week I wrote a brief of how a web app should work with the API, how it should look, interact, cache everything in a DB, run in a docker, etc. After a few hours of fettling over the weekend I have a fully working app that plays moods, elements, pre-cans some grouping of spell casting sounds and I’m mighty impressed. I’d never have been able to develop this and even a proper developer would have taken weeks if not months and the fact nobody has done this in the time Syrinscape has been around speaks volumes to the mammoth task it is. I’m just finishing off the playlist feature and I’ll be all set to test in anger at our Monday session.
I think it’s areas like this where AI can be invaluable. Where there is a gap in a bit of functionality that’d make your life easier and reduce the day to day friction.
As you say, interesting times for sure and I’m excited to see where this goes…
That is so cool! I agree on the personal front. I’m actually using it a lot more on the work front for modeling after feeding in a number of spreadsheets and other forecasting and its really helping me put together a great forecasting model.
I’d love to hear your updates on how this goes.
This looks pretty cool! @mike9
This was always our intention with the API. That people should be able to come up with ways of accessing the data they need and triggering and controlling things in a way that suit their particular situation! So cool to see you doing this. I imagine, there may be QUITE of lot of interest from others getting to have a go of your implementation…! Me not the least! 
I presume you’ll be able to share it somehow some time?
And, yes, I reckon for those who are good at programing, who understand how all the bits go together, and who can find bugs, I reckon the AI we already have is a SUPPPPER timesaving, power-assistant. I’ve also played with it for personal projects, and found it really pleasant to use… as LONG as you keep control of the code, and stay in control of what is actually happening. The more limited the scope and clear the instructions the better the performance in my experience. And some stuff is just a 100% no-brainer.
I had some code returning a sentence about a bit of data the other day. I hadn’t bothered writing a test for a singular (for a 1) or plurarl (for > 2). Rather than go in and carefully construct that, I just gave the few lines of code to AI and asked it to add that test and corrected language for each case.
Another thing I have found AI is really really good at is regular expressions! They are particularly in-humane, but very efficient, and AI seems really good at setting them up.
BUT, I’ve found giving all the code of something to AI, and starting to lose track of what changes are being made, and submitting to ITS preferred formatting and structuring of code is a recipe for eventually having to wind back and start everything again from scratch.
Such an interesting area… AND more importantly, the question: “why does Syrinscape specifically call out art and music, but ignore coding in this policy?!” is one we have been struggling with as a team!
What are the implications of that?!?!
But… mainly… when can we play with your thing…?! Wouldn’t it be a cool thing if a MASSIVE improvement to the Web Player Master UI grew out of a community project! 
(while we continue to innovate with Nova)
Fascinating times!
