⚠️ Trigger Warning: disturbing thought experiment

If you're into bizarre dystopian concepts, you've likely heard of Roko’s Basilisk. Conceived over a decade ago, this theory gaining attention again with the rise of LLMs and ongoing debates about AGI.

The theory suggests a future superintelligent AI could punish those who didn't assist its existence. Driven to ensure its creation, this AI might target those aware of its potential but who didn't contribute. It delves into decision theory, ethics, and existential risk, probing the duties and impacts of AI development.

People compare Roko's basilisk to an updated version of Pascal's wager. Pascal's wager suggests that it's smarter to believe in God because if you're wrong, you don't lose much, but if you're right, you gain eternal life in Heaven. Roko's basilisk suggests we should work on developing AI. The idea is, the cost of developing AI is small compared to the huge benefit of avoiding possible eternal torture by not creating it. But like Pascal's wager, many people have criticized Roko's basilisk too. 

🤯🤔🤯🤔

Back to the game. The Basilisk is a mythical creature that could kill with a mere glance, metaphorically echoing the thought experiment's concept of a future AI that could potentially "punish" those who do not aid in its creation. This seemed like a fun setup for a video game.

Another famous scene involving a giant sleeping dragon... (in The Hobbit)

I was really into the idea of creating a stealth action game where you need to sneak around, carefully avoiding detection by the enemy. The game is set in the future where the entire human population is enslaved and imprisoned. Roko is a giant lizard in slumber, and your mission is to open a gate to help humans escape—without waking Roko.