So I’m playing with AI (https://chat.openai.com/) and I wonder how we could use it for sociocracy practice.
I’m seeing a lot of use for it in particular for information flow, like generating useful summaries of what other circles are doing.
In my community, we were thinking that it’s a bit like having a very knowledgeable new circle member as well, at least for generic questions.
What are you thinking?
Up until explorations, I’d be happy to welcome them as a circle member – with no consent rights though (it doesn’t get that consent is not about preferences…)
I noticed it typically resists opinions but it does have opinions about a few things - for example, climate change. Although, from my point of view, that’s not an opinion but scientific consensus so I guess I agree with it.
I entered the game you proposed and I asked AI:
Is sociocracy useful?
Here is the response:
Sociocracy is a governance model that has been used successfully in a variety of settings, including businesses, nonprofits, schools, and community organizations. It is designed to facilitate collaboration and participation among all members of an organization, and to allow for decentralized decision-making and adaptive management.
There are several potential benefits to using sociocracy as a governance model. Some of these benefits include:
Promoting collaboration and inclusivity: Sociocracy’s consent decision-making process is designed to ensure that all voices are heard and that decisions reflect the needs and concerns of all members of the organization. This can help to create a more collaborative and inclusive decision-making process.
Encouraging participation and engagement: By giving all members of the organization a voice in decision-making, sociocracy can help to foster greater participation and engagement among all members.
Improving effectiveness and efficiency: The consent decision-making process is designed to be efficient and effective, and can help organizations make decisions in a timely and efficient manner.
Promoting transparency and accountability: Sociocracy is based on principles of transparency and accountability, which can help organizations to be more open and accountable to their stakeholders.
That being said, it’s important to note that no governance model is a one-size-fits-all solution, and it’s important for organizations to carefully consider their specific needs and goals when deciding on a governance model. It may also be helpful for organizations to consider a variety of different models and approaches, and to choose the one that works best for them based on their unique context and circumstances.
AI can collect meaningful information about a subject, but it is limited to the algorithm provided by its programmers. He has no personal opinions and brings nothing new beyond what has already been said by someone, sometime, somewhere … Anyway, it can be fun to interact with… 
So imagine this:
AI is always reading all notes; one circle talks about topic XY, and somewhere else in the org, another circle talks about something similar. Could we use AI to let us know that? AI would pop up and say “hey just so you know” and give the circle a summary of what the other circle thought? That would be an INSANE game changer for decentralized orgs
I also want AI to “:notice” when there are patterns, like all circles talking about being understaffed, to name something plain and simple. It could then say. “Hey just so you, many circles are currently struggling with being understaffed. Maybe something to look at?" and then it recognizes that that’s a GC issue and suggests to me to put it on the backlog.
When I plan the meeting, it might add info to the backlog items to tell me how “hot” they are currently.
EDIT TO ADD: If this worked really well, one could even do without aims/domains in an organization. The only reason aims and domains exist is because our minds are limited and we need to `categorize’ certain topics into domains. If every circle talked about all kinds of stuff, we’d lose track of who decides what. So the nested domains are useful as a constraint to make domains easier to track and avoid doubling and gaps between in domains. Now if AI reliably recognized related topics and could give real-time feedback to groups and individuals, telling them what other groups’ topics they are currently touching, then we could make do without domains. This is wayyyy into the future, but I think it’s possible.
@thomas.kemps @jerry.koch-gonzalez @alex.rodriguez @Julien.Gaillard you might like this thought.
I wrote a proper LinkedIn post on it here that lays out the idea a little better: Ted Rau on LinkedIn: #openai #decentralizedorganizations #sociocracy #holacracy…
What would be game-changing is software that automatically transcribes and summarize circle meetings, than properly classifies information for other circles and roles to digest, i.e. news feeds.
Step 2 would be analysis of power dynamics my measuring contributions and influence which would translate into adressable tensions.
Having AI nudge suggestions could be nice but could also be a dangerous path if the members do not have a say in the algorithms used.
We could have tried something already – such as looking for recurring themes in meeting agendas – if Open AI would not have had this limitation:
Do you learn new information?
As an artificial intelligence, I do not have the ability to learn new information in the same way that humans do. I was designed to be able to generate responses to a wide range of questions and prompts based on the information that I was trained on, but I am not able to learn new things or acquire new knowledge on my own. My knowledge and understanding are limited to the information that was included in my training data, and I am not able to browse the internet or access new information that has been published since my training cutoff date.
However, after this answer I was a little confused 
because it does remember a few things: for example, I asked OpenAI to draft a business plan for a European nonprofit working on sociocracy; it also suggested a name for this organization and after a few questions I asked to write a Facebook post about the sociocratic selection process. It used the same name it suggested earlier in the business plan.
This maybe means that within the same chat we could provide the AI with a few info and get something out of it, streaching a little the answer it gave me to the ‘Do you learn new information?’ question. So let’s continue playing with it! 
Yes, the “news feed” function seems like it would be especially useful. Having a way (even if somewhat clunky) to read summaries of not only what groups are working on what, but how they are feeling about the work, if tensions are surfacing in multiple places, etc. I also wonder how this might be designated as a domain within an org, a “news studio” within it that is tracking these things and actively asking people questions…
