Let me tell you, I went into this guest lecture at Stanford last week fully expecting to roll my eyes the whole time. I thought virtual humans were just another dumb gimmick, another startup trying to cash in on VR hype. Man, was I wrong. This talk was from the team at Cougar, this Tokyo-based startup that's building the next generation of human-AI interaction tools, and they completely blew my mind.
The Speakers
The two guys running the show were the co-founders of Cougar, and their backgrounds are so cool it sounds made up:
-
Satoshi Ishii, their CEO: This guy's a total veteran engineer. He spent years at IBM building large-scale distributed systems before he left to start Cougar, because he wanted to build virtual human tech that actually matters.
-
Kazuaki Kanzaki, their chief blockchain architect: Self-taught dev, started out as a DJ in Hollywood before he got into blockchain and AI infrastructure. Dude's built systems to fix AI's trust and privacy problems, and it's way more legitimate than any crypto bro nonsense I've heard lately.
What This Talk Was Actually About
These guys cut through all the empty AI hype we've been drowning in lately, no cap. They gave this practical, no-BS look at what's actually next for how we interact with tech. They walked us through their flagship product, Connectome — this platform that mixes IoT, AR, AI, and blockchain to build "virtual human agents," which are basically the next step after boring old voice assistants like Siri. They broke down all the hidden trust and privacy issues that are plaguing the centralized AI we use today, and showed us how decentralized tech can fix all that. Their end goal? A world where anyone can build, share, and use trusted AI, without having to rely on big tech monopolies hoarding all our data.
Who This Was For
Honestly, this talk was for pretty much anyone in tech right now:
-
AI product managers who are sick of building the same old voice assistants over and over again
-
Blockchain devs who are tired of all the crypto nonsense and actually want to build something useful
-
UX/UI designers working on immersive and AR experiences
-
Startup founders trying to build the next big AI or virtual human product
-
Anyone who's just sick of big tech hoarding all your data and lying about AI safety
What I Walked Away With
By the end of the hour, I learned way more than I expected:
-
How human-computer interfaces are evolving, and what comes after smartphones and smart speakers
-
What virtual human agents actually are, and how they mix machine learning and game AI to feel like real people
-
All the hidden trust and privacy problems with today's centralized AI that big tech doesn't want you to know about
-
How blockchain can fix AI's "black box" problem, making it transparent and verifiable
-
How to build a decentralized marketplace for AI agents, so we don't have to deal with big tech gatekeepers anymore
