[ExI] questions

[William Flynn Wallace]

Do not think of predicting the behavior of a group of 600 people.  Think of
predicting the behavior of an individual 600 times.  Even with all the data
on each person's actions and votes, it is very difficult to predict the
actions and votes on one issue by each person.  bill w.

On Wed, Apr 16, 2025 at 3:30 PM BillK via extropy-chat <
extropy-chat at lists.extropy.org> wrote:

> On Wed, 16 Apr 2025 at 20:32, Darin Sunley via extropy-chat <
> extropy-chat at lists.extropy.org> wrote:
>
>> At this point, anyone with any stake in US politics who doesn't have 600
>> deep-reasoning chatgpt instances running, one for each congressmen,
>> senator, senior elected official and cabinet member of the us government,
>> with a context window stuffed with every public utterance and obtainable
>> private communication, running a simulacrum of the the entire US
>> government, just isn't playing the game seriously.
>>
>> "They're dinosaurs. And they're going to be sitting on their asses on the
>> couches in October, watching the Boston Red Sox win the World Series." -
>> Moneyball
>>
>> ----------------------------------------
>
>
>
> I asked Gemini Pro 2.5 Deep Research about this possibility.
> Not feasible at present was the conclusion. Human behavior too
> unpredictable.
> It produced a long report, detailing all the problems.
> Here is the Conclusions paragraph.
> BillK
>
> In conclusion, while the concept of using 600 Deep Reasoning AIs to
> emulate and predict the behavior of the US government is theoretically
> captivating, it faces substantial hurdles with current and near-future
> technology. The inherent limitations of AI in understanding human-like
> reasoning, the unpredictability of human behavior in political contexts,
> the immense data and computational power requirements, and the significant
> ethical and societal implications all present formidable challenges. While
> AI can undoubtedly contribute to political analysis in more focused and
> specific applications, such as analyzing large datasets and identifying
> trends, the comprehensive emulation of a system as complex and human-driven
> as the US government remains a distant prospect. The continued importance
> of human expertise, critical thinking, and a deep understanding of
> political science principles will remain paramount in analyzing and
> interpreting governmental behavior. Future advancements in AI might make
> more sophisticated modeling feasible, but this will necessitate careful
> consideration of the ethical and societal consequences to ensure the
> responsible and beneficial application of such powerful technologies.
>
> ------------------------------
>
> _______________________________________________
> extropy-chat mailing list
> extropy-chat at lists.extropy.org
> http://lists.extropy.org/mailman/listinfo.cgi/extropy-chat
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20250416/ced46cc7/attachment.htm>