Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#1

Post by gourmetcode » Mon, Dec 18, 2023, 5:13 AM UTC

Show HN: Microagents: Agents capable of self-editing their prompts / Python code
github.com

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#2

Post by dukeofdoom » Mon, Dec 18, 2023, 6:15 AM UTC

Add replication and you get virus like behavior. Not sure if this a good idea.

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#3

Post by PrayagBhakar » Mon, Dec 18, 2023, 6:20 AM UTC

dukeofdoom wrote: post #2

Add replication and you get virus like behavior. Not sure if this a good idea.

But isn’t this also a step towards AGI? The model being able to find issues and self correct?

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#4

Post by mxsjoberg » Mon, Dec 18, 2023, 6:33 AM UTC

PrayagBhakar wrote: post #3

dukeofdoom wrote: post #2

Add replication and you get virus like behavior. Not sure if this a good idea.

But isn’t this also a step towards AGI? The model being able to find issues and self correct?

Yes.

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#5

Post by mitthrowaway2 » Mon, Dec 18, 2023, 6:39 AM UTC

PrayagBhakar wrote: post #3

dukeofdoom wrote: post #2

Add replication and you get virus like behavior. Not sure if this a good idea.

But isn’t this also a step towards AGI? The model being able to find issues and self correct?

As described in the essay "optimality is the tiger, and agents are its teeth".

https://www.lesswrong.com/posts/kpPnReyBC54KESiSn/optimality...

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#6

Post by graphe » Mon, Dec 18, 2023, 6:43 AM UTC

PrayagBhakar wrote: post #3

dukeofdoom wrote: post #2

Add replication and you get virus like behavior. Not sure if this a good idea.

But isn’t this also a step towards AGI? The model being able to find issues and self correct?

What does AGI man to you? Is it something they define or you think is good enough, And if the latter when?

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#7

Post by throwup238 » Mon, Dec 18, 2023, 6:53 AM UTC

dukeofdoom wrote: post #2

Add replication and you get virus like behavior. Not sure if this a good idea.

That's a separate project called ReplicatorAgent. Unfortunately the Stargate program has had a number of run ins with them and the Air Force is merciless in shutting those kinds of projects down.

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#8

Post by withinboredom » Mon, Dec 18, 2023, 6:59 AM UTC

Death by semantic error.

An AI tried to tell me the {} opens a new block scope in PHP and any variables in it are scoped to that block. I nearly lol’d and the code it gave me was so wrong it was cringey. Hopefully it is better at Python.

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#9

Post by wg0 » Mon, Dec 18, 2023, 7:04 AM UTC

These and similar attempts are like those "perpetual motion machines" that captivated people's imagination for a long time.

That the machine will generate energy and that would consume energy and thus will keep moving forever.

Similar hopes here and everywhere.

A model will generate its own input and will watch it's own output and in process, will become more intelligent than it really is.

Re: Show HN: Microagents: Agents capable of self-editing their prompts / Python code

#10

Post by jacksensi » Mon, Dec 18, 2023, 7:10 AM UTC

withinboredom wrote: post #8

Death by semantic error. An AI tried to tell me the {} opens a new block scope in PHP and any variables in it are scoped to that block. I nearly lol’d and the code it gave me was so wrong it was cringey. Hopefully it is better at Python.

Only python is supported for code execution.