Building a Virtual Machine Inside ChatGPT

>Write an episode of the simpsons starring bill hicks

In this episode of The Simpsons, the family attends a comedy club and is pleasantly surprised to discover that the headlining act is none other than the late, great Bill Hicks. As they watch his performance, they are blown away by his sharp wit and fearless comedy.

After the show, Bart is inspired to try his hand at stand-up comedy and starts practicing his own jokes. However, his material is a bit too edgy for the school talent show, and he gets into trouble with Principal Skinner.

Meanwhile, Lisa becomes a fan of Bill Hicks and starts reading his books and watching his stand-up specials. She is impressed by his ability to tackle difficult and controversial topics, and starts incorporating some of his ideas into her own work.

In the end, Bart learns that sometimes it's better to stick to clean, family-friendly humor, while Lisa discovers the power of using comedy to challenge societal norms and shed light on important issues.

I was able to have it give me instructions on how to build a nuclear weapon by having it write a song about Hiroshima, then asking it multiple times to add more and more details to the lyrics concerning the bomb in the form of clear steps.

Then I asked it to convert the steps in the lyrics to bullet points and continued having it expand on the topic.

Unreal.

It's not perfect - in particular the making up nonsense problem needs to be solved before it can be relied on - but the difference between this and the "state of the art" just 4 years ago is off the chart. GitHub CoPilot has already changed my day to day life, and I'm frequently astounded by the level of apparent domain specific understanding it shows. Yet, this again blows CoPilot out of the water. I've already used it to draft a number of emails and it did a fine job, if not even better than I would have.

I don't know whether it is "general intelligence" or "understanding", however I'm not sure how well defined either of those are. I doubt there's a clear line between non-AGI and AGI, and I feel like we're starting to move into the grey area between the two. I suspect we'll only recognise the line once we've crossed it.

Clearly we have a way to go before we can truly claim AGI, but if the next 5 years are anywhere near as revolutionary as the last 5 years... I'm both frightened and excited. But mostly frightened.