theMultiplicity.ai
@Multiplicityai
Followers
323
Following
11
Media
8
Statuses
16
We've built a tool to help increase transparency in AI bias by engaging multiple models simultaneously to see how they think differently about the same queries!
Berkeley
Joined August 2025
Were you surprised? I was surprised. See the full discussion with @divyasiddarth here: https://t.co/do3dc1nQGQ
theMultiplicity Podcast Episode 1: Divya Siddharth on collective intelligence, with Andrew Critch (@divyasiddarth / @AndrewCritchPhD) 0:00:00 – Collective intelligence, AI and democracy 0:07:29 – Capitalism 0:12:47 – Nudging America 0:14:43 – More capitalism 0:16:14 – Facebook
0
2
5
Users wanted a way to permanently turn on the "Smart/Slow" models, and/or save their preferred AI aggregator model between rooms. Our new "Save Default Settings" menu will do this. Keep posting public feature requests 😀
1
1
1
theMultiplicity Podcast Episode 1: Divya Siddharth on collective intelligence, with Andrew Critch (@divyasiddarth / @AndrewCritchPhD) 0:00:00 – Collective intelligence, AI and democracy 0:07:29 – Capitalism 0:12:47 – Nudging America 0:14:43 – More capitalism 0:16:14 – Facebook
7
0
15
Big thanks to @ScottEdwar33859 for being (I think!) the first person to file a public bug report @Multiplicityai! Scott: I just generated you a $100 gift code, see your DMs 😀 Everyone: more public bug reports and feature requests for https://t.co/1MgpRIuDii please!
@Multiplicityai @AndrewCritchPhD looks like it's working now - thanks for fixing - I love it!
3
2
6
grok-4.1-fast-reasoning has now replaced gemini-2.5-pro in https://t.co/4MuC0nEEZi for summarizing AI responses; it's faster and more faithfully better summarizes contentious topics. Try it here: https://t.co/pTkXVSvNL4 Congrats @grok / @xai / @elonmusk on a great release!
1
1
2
News update! I hereby introduce https://t.co/4MuC0nEEZi, with our first tutorial video. theMultiplicity is a multi-agent chat app with GPT-5, Claude Opus 4.1, Gemini 2.5 Pro, and Grok 4, plus special protocols for collaborative ranking and estimation tasks, to help you quickly
Big news, friends! I hereby introduce https://t.co/4MuC0nEEZi. It's a multi-agent chat app with special features for collaborative ranking and estimation tasks, to help you quickly fact-check AI responses against each other. It has GPT-5, Claude Opus 4.1, Gemini 2.5 Pro, and
3
8
34
Big news, friends! I hereby introduce https://t.co/4MuC0nEEZi. It's a multi-agent chat app with special features for collaborative ranking and estimation tasks, to help you quickly fact-check AI responses against each other. It has GPT-5, Claude Opus 4.1, Gemini 2.5 Pro, and
9
18
87
Does AI think it's possible to prove we're *not* living in a simulation? Here's what the four smartest available models agreed on:
0
0
1
Are even lemonade stands illegal in California? This deeply important question surely warrants not-legal-advice from eight different AI models.
0
0
2
What does AI think people mean when we ask if it's conscious? Eight AI models show surprising level of agreement on this...
0
0
2
What are the best foods to get rid of gas? A clear winner emerges, after AI models bloviate on this explosive topic...
0
0
2
How many square meters of ideal radiative cooling surface is needed to run an h100 gpu in space?
0
1
3