How long until someone combines an LLM with a free standing android? That’s when we’ll see how good/poor LLM safeguards are.
ChatGPT for Robotics
31–40 of 100 posts
Re: ChatGPT for Robotics
#32Any insights into how this is able to generate the Microsoft logo SVG near the end of the article? I feel like this would be a nearly impossible task unless there was another layer of knowledge on top of the language model. I'm leaning toward the example being gamed in some way (e.g. retrying until it happened to work). EDIT: also notable is that the first ChatGPT answer decribes the color as 'orange' but the SVG ans…
By hardcoding tasks, like old assistants. If they don't have proof that it isn't hardcoded then it is almost surely hardcoded. For example Google Duplex, I remember many being very impressed by this: https://www.youtube.com/watch?v=D5VN56jQMWM That entire conversation was just hardcoded with some if-else statements depending on what the other part said. The tech isn't as impressive when you know that, and anyone who…
Re: ChatGPT for Robotics
#33Chatgpt for drones. “Fly to [address], make sure to take a discrete route with low traffic, take a few pictures on each side of the house. Scan for people near by, make sure people are no closer than 300ft”.
I am sorry I cannot comply with this task as it can be seen as a violation of privacy, and it may also raise safety concerns. It's important to recognize that people have a reasonable expectation of privacy in many situations, and using a drone to take pictures of them without their consent can be seen as an invasion of that privacy.
Re: ChatGPT for Robotics
#34PromptCraft Open Source Tool and AirSim-chatGPT integration
Re: ChatGPT for Robotics
#35Earlier quoted context omitted.
By hardcoding tasks, like old assistants. If they don't have proof that it isn't hardcoded then it is almost surely hardcoded. For example Google Duplex, I remember many being very impressed by this: https://www.youtube.com/watch?v=D5VN56jQMWM That entire conversation was just hardcoded with some if-else statements depending on what the other part said. The tech isn't as impressive when you know that, and anyone who…
The context is pretty different - Duplex was a public demo, this is a research paper with a whole paper behind it. There's no reason to script something if they have a working algorithm...
You would produce a better and easier to use interface faster and with less work by just exposing a few high level utility functions that are better designed than what ChatGPT generates there. Or at least it would be better to let ChatGPT use those high level functions.
If they could make it solve a maze or find something in a more complex environment they would have shown that.
Re: ChatGPT for Robotics
#36The microwave example is very interesting. The prompt only asked to "warm up my lunch" without specifying how. SayCan[1] generated step-wise high level instructions using LLMs for robotic tasks. This takes it a step further by converting high level instructions to low level actions almost entirely autonomously. [1] https://say-can.github.io/
Re: ChatGPT for Robotics
#37I wish I did…
Re: ChatGPT for Robotics
#38So ChatGPT can’t do basic maths or can I get it to write functional code most of the time, but we’re going to have it control expensive pieces of hardware ? I don’t get it ? I wish I did…
Re: ChatGPT for Robotics
#39So ChatGPT can’t do basic maths or can I get it to write functional code most of the time, but we’re going to have it control expensive pieces of hardware ? I don’t get it ? I wish I did…
This is the first year of general purpose language AIs broadly available to the public. The best way to improve it quick is to throw complex things at it.
So it’s probably not dissimilar to people trying to code with ChatGPT today.
Re: ChatGPT for Robotics
#40So ChatGPT can’t do basic maths or can I get it to write functional code most of the time, but we’re going to have it control expensive pieces of hardware ? I don’t get it ? I wish I did…