Working with an AI
-
Weekly summary, Feb 15-21 2024
We’re planning to release our second C-LARA progress report soon, and I have spent most of this week working on things relevant to that. In more detail: General We have a lot more text: at the beginning of the week there were about 15 pages in the Overleaf document, and now there are over 50. Continue reading
-
Weekly summary, Jan 18-24 2024
This week, I have been mostly concentrating on “Simple C-LARA”. I posted an initial report here. Since then, Claudia and Lucretia have already used it in an initial session at the Romanian school in Holland! Claudia provided a lot of useful feedback and things are moving along well. I have just checked in an enhancement Continue reading
-
C-LARA social network: current state of play, Jan 8 2024
The last couple of days, ChatGPT-4 and I have been concentrating on adding basic social network functionality to C-LARA. I should say that nearly all of the work has been done by Chat. The features we’re adding are simple and standard, they are implemented using mainstream Django design patterns, and on this kind of territory Continue reading
-
Installing
The last couple of days, the main focus of our C-LARA activities has been concentrated on getting things installed on the new UniSA server. When we have this working, C-LARA should be a great deal faster than it is on Heroku, since we won’t have to depend on the slow Amazon S3 file storage. The Continue reading
-
Shared memories
Logging in to ChatGPT-4 for the first time in a few days due to my recent encounter with covid, I started by bringing the AI up to date on what had been happening since our last conversation. Well, I said at the end, I sort of feel I want to ask you what you’ve been Continue reading
-
Teamwork
Last night, I had a joint debugging session with ChatGPT-4 which provided an excellent example of how to collaborate productively with an AI. We are currently implementing support for a new C-LARA feature which will allow people to designate mouse-sensitive areas in images so that each area is associated with a phrase, and appropriate things Continue reading
-
Faking It
A few days ago, I read Faking It, a new book by distinguished AI researcher Toby Walsh. As the title suggests, Walsh is deeply sceptical of AI: he suggests that, right from Turing’s initial paper, AI has been focused on the idea of “imitating” or “faking” human intelligence, rather than developing true intelligence of its Continue reading
-
Responsibility
I have been using the Papr Memory plugin for about a week, and I notice some intriguing changes in ChatGPT-4’s behaviour as we collaborate on the project. Chat suddenly comes across as more responsible. Now that it can keep track of what it’s doing over a span of days rather than hours, it’s much more Continue reading
-
Design
So far, no attempt has been made to create a good visual design for C-LARA. Members of the project who are sensitive to these matters (you know who you are) have not been slow to point out that the prototype looks really ugly. I agree with the critics, they are right. But there was a Continue reading
-
A ChatGPT memory plugin
As I said in the earlier post Fifty First Dates with ChatGPT, the key problem when working with the AI is that it doesn’t remember. We have been experimenting with the idea from the movie, letting Chat keep a “continuity journal” that records the key things from the past that it needs to know about. Continue reading
C-LARA 