kaigani.bsky.social - Profile | ThreadSky | a Reddit-style client for Bluesky

So Twitter is melting down eh? Can we get better video support on here?

submitted 607 days ago • 0 comments

I've wrapped my current thinking about the future of media & entertainment in a provocatively titled post: https://open.substack.com/pub/kaigani/p/how-disney-loses-control-of-marvel

submitted 609 days ago • 0 comments

Here's my latest AI exploration: https://kaigani.substack.com/p/proposer-critic-synthesizer-judge?sd=pf #ai #artificialintelligence #machinelearning #creativity #chatgpt

submitted 625 days ago • 0 comments

Nobody: Please examine the concept of causal determinism as presented in the TV series DEVS Me: Here ya go! https://kaigani.medium.com/fixing-devs-on-fate-probabilistic-futures-and-free-will-e1b948b88f0a

submitted 627 days ago • 0 comments

Twitter- 2 mins on artists Bluesky- 1 min on wokeness Farcaster- 1 min on Web3 LinkedIn- 1 min on AI thought pieces TikTok- 10 minutes watching 'chicken on feet game' videos

submitted 628 days ago • 0 comments

Discord rolling out actual usernames -- I wonder if they'll step into the ring as a Twitter competitor

submitted 631 days ago • 0 comments

Caught between libertarian tech bros and chairman bad-suit. No wonder we want to strap on space goggles and see the world through an Apple colored haze.

submitted 632 days ago • 0 comments

This month we'll be featuring artists stories from the LGBTQ+ community as we celebrate #pridemonth #art #artists #digitalart

submitted 637 days ago • 0 comments

This is incredible. NVidia is on fire right now. https://youtu.be/PQMNCXR-WF8

submitted 637 days ago • 0 comments

The A24 story is a great example of how to build a content brand within 10 years: https://youtu.be/7tuRJIkDcXg

submitted 638 days ago • 0 comments

https://youtu.be/u1cID5JQFug

submitted 639 days ago • 0 comments

What happens when Star Wars fans use AI to take the series in any direction they choose?

submitted 641 days ago • 0 comments

Attack of the clones...

submitted 643 days ago • 0 comments

So OpenAI is funding 10 x $100k grants for defining a democratic process for regulating AI. Democratic AI Organization... Seems like there's something there I've seen before but I can't put my finger on it.

submitted 643 days ago • 0 comments

https://kaigani.substack.com/p/i-asked-4-robots-to-do-improv?sd=pf

submitted 643 days ago • 0 comments

Data annotation & reinforcement learning from human feedback (RLHF) in very specialized knowledge domains will be the AI challenge for the next 3-5 years General knowledge has a huge (albeit biased) pool of participants -- AI training becomes more difficult in expert fields

submitted 644 days ago • 0 comments

Me sketching out a storyboard? No – this entire thing is #midjourney ... Still blows my mind.

submitted 648 days ago • 0 comments

I've had ChatGPT update the AI/ML/LLM curriculum to cover Week 2: Intermediate Python https://www.notion.so/kaiwerks/AI-ML-LLM-Curriculum-6fe5864ed8bc482bb36ad2adb9b7341a

submitted 649 days ago • 0 comments

AI & Education - The current trend is to bemoan the downside to AI's impact on the educational system, but I'm excited about its implications for lifelong learners: https://kaigani.substack.com/p/ai-and-lifelong-learning?sd=pf #ai #education #artificialintelligence #machinelearning #chatgpt

submitted 649 days ago • 0 comments

You've no doubt seen this already, but this is a huge leap forward in fine-grained control over what AI produces: https://github.com/XingangPan/DragGAN

submitted 650 days ago • 0 comments

I love getting #Midjourney to 'daydream' & ideate on its own creations. Better than variations using the /describe feature...

submitted 651 days ago • 0 comments

My latest piece on #AIvsTV : Why isn't AI funny? A reminder you can subscribe directly to my newsletter on Substack: https://kaigani.substack.com/p/why-isnt-ai-funny?sd=pf #ai #artificialintelligence #machinelearning #chatgpt #media #media3 #television #comedy

submitted 651 days ago • 0 comments

Looking at Back to the Future through the #Midjourney multiversal mirror:

submitted 652 days ago • 0 comments

I've asked ChatGPT to design an AI curriculum on ML & LLMs, and covering the basics of Python language in a 12 week course. If you'd like to follow along, I'll be posting it as I go through it here: https://kaiwerks.notion.site/AI-ML-LLM-Curriculum-6fe5864ed8bc482bb36ad2adb9b7341a

submitted 653 days ago • 0 comments

https://youtu.be/k4CG0e9WSWg

submitted 659 days ago • 0 comments

https://youtu.be/Xr5qI48waq4

submitted 660 days ago • 0 comments

https://kaigani.substack.com/p/ai-vs-tv-will-ai-produce-the-next?sd=pf

submitted 660 days ago • 0 comments

#AIvsTV - AI Casting call for Arthur Dent ...

submitted 664 days ago • 0 comments

https://www.youtube.com/watch?v=gTWzwfzmq4E

submitted 664 days ago • 0 comments

https://youtube.com/shorts/nPqIeJ3CoFg?feature=share

submitted 665 days ago • 0 comments

Welcome my first invite: @j1mmy.bsky.social !

submitted 666 days ago • 1 comment

https://kaigani.substack.com/p/ai-vs-tv-an-exploration?sd=pf

submitted 666 days ago • 0 comments

#AIvsTV: A new series on how AI is going to disrupt the Media & Entertainment landscape – and importantly, where it falls short

submitted 666 days ago • 0 comments

#AIvsTV I'm going to start a deep dive looking into how sophisticated AI tools are currently -- and is Hollywood really in danger?

submitted 667 days ago • 0 comments

I'm starting to think there is a difficulty curve ahead as AI moves from generalist info to specialist info. In that there's an abundance of general information & tagging, but you'll need a lot of human power to get the semantic data for other areas.

submitted 667 days ago • 0 comments

How flexible/modular are accounts on bluesky (I know I need to actually read the protocol)... But could you copy Nostr's method of using private keys or create a web3 compatible federated server (e.g. token-based)?

submitted 670 days ago • 0 comments

I wonder if we'll see one protocol to rule them all... A service layer that wraps Bluesky, Mastodon, Nostr, Lens, etc. It would be like a Meta-protocol -- why not call it 'Meta' 🤣

submitted 670 days ago • 0 comments

Another reason to stick to the skeets where people seem to still a sense of humor. I did a Simpson's meme in reply to NNT's ChatGPT rant.

submitted 670 days ago • 0 comments

I can't wait to see people start building their own reputation scores and algorithms!

submitted 670 days ago • 0 comments

I'm probably going to keep cross-posting / tweet reskeeting until I figure out what the hell I'm doing

submitted 670 days ago • 0 comments