@Austin Thanks so much!
This is a great point and something we went back and forth about a lot. I am going to post in more places today and hopefully we'll see a bit more traffic. If we are seeing consistently high traffic in those windows we will extend the times where it's playable.
But because you need other people to be online while it's being played, and currently we're not seeing very high traffic even in those periods, then hopefully this format at least maximises the chance that people can play every day.
I like the idea of adding a share button to the homepage. There is currently one after you complete a game, but it could be good to incentivise people to share so that they can play straight away.
On the point of playing vs a friend, this might undermine the Turing test to some extent. I think if you know that the other person is your friend you have a lot of insider knowledge that would allow you to beat even a very good LLM agent (or another human)!
Thanks very much for your support and these suggestions!
@cameron
$0 in pending offers
Comments
camrobjones
18 days ago
camrobjones
25 days ago
The site is finally back up at turingtest.live.
The new site uses a 3-party format where you chat with both a human and an LLM at the same time, and your task is to decide which is which. This setup is closer to Turing's original idea, and we think it will be much harder for the models to pass.
We've also added a range of models, including GPT-4o, Claude, and LLaMA, along with new prompting techniques that allow the models to interact in different ways.
The site will be live every day from 12β1 PM and 8β9 PM GMT (8 AM and 3 PM ET, 5 AM and 12 PM PT). We've done this in the hope of increasing the density of users online at the same time.
Thanks so much to everyone for your help, patience, and support while getting this back going. Please let me know if you have any comments or thoughts on the site as I'll be continuing to make updates. And please feel free to share the site as it works best when we get enough traffic for many people to be live at the same time.
camrobjones
about 2 months ago
Weβre planning to relaunch the site soon and weβre running a pilot test on Friday at 8pm GMT (3pm ET, 12pm PT).
You can access the new site here: 3p.turingtest.live. The pilot will be accessible here on Friday.
Any feedback you can provide about how the site and your interactions work would be greatly appreciated!
If you have any questions or comments, please let me know at cameron@ucsd.edu.
camrobjones
4 months ago
Hi all,
Thank you for your patience and apologies for the lack of updates. I have had to focus on other things over the last months including finishing my PhD and starting a postdoc. However, I am now able to put my full focus on this for the next couple of months. I'm hoping to have the updated experimental design finished by the end of the month and to start collecting data in October.
Thanks again and please let me know if you have any questions!
Cameron
camrobjones
9 months ago
@dominic Thanks very much, Dominic! I'm glad you had a chance to try it out and I appreciate the support!
camrobjones
9 months ago
@AntonMakiievskyi Thanks so much Anton! I really appreciate the support.
1. Participants are randomly assigned to be witnesses or interrogators. The lack of humans online is a definite issue, as there were periods where a player would be repeatedly matched with AI if no humans were online. I'm considering only making the game available for ~1hr a day to maximise the density of humans online while keeping costs down.
2. I have applied for OpenAI but didn't hear back. Will try Claude & OpenAI again.
3. This could be a good backstop if we run out of credits again. I'm a little nervous about handling the data but I'm sure there's a secure way to do this.
Yes, I will probably take a couple of weeks to make changes to the site and then hopefully update just before we relaunch. Thanks again and let me know if you have more questions or would like to chat more.
Transactions
For | Date | Type | Amount |
---|---|---|---|
Manifund Bank | 8 months ago | withdraw | 1999 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +234 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +500 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +20 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +250 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +150 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +200 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +25 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +420 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +100 |
Run a public online Turing Test with a variety of models and prompts | 9 months ago | user to user trade | +100 |