Thursday 3 October 2024

The Resurgence of Penrose

Since my first reading of The Emperor's New Mind and some other books in Secondary School, I've liked Penrose. I doubt his theory, but no one has the experimental capability to provide solid evidence, or disprove it.

C.C.C. ( Conformal cyclic cosmology ) seems to make sense and relatively easy to understand. The idea that black holes evening suck everything up, decaying, and mass becoming negligeable, space and time becoming irrelevant, then Big Bang Boom!

Penrose has recently done more interviews, especially the Lex Friedman interview, which brought him back into the popular science spotlight.

To this day, I haven't come up with the right background to do the physics to understand his stuff.


https://youtube.com/shorts/DVNQ-MTkETI

Anyway, this brought back lots of memories I haven't accessed for a long time.



Tuesday 3 September 2024

Afghanistan

 With Biden's disastrous withdrawal, US has made Afghanistan much worse after 20 years.

The Taliban will likely never be removed, because there is no one with sufficient forces or political will for such change.

The most one can hope for is something like Iran, but that's not going anywhere anytime soon.

https://x.com/TimesAlgebraIND/status/1830716546399740367

Monday 2 September 2024

Told ChatGPT to make a meme about a feeling

An appropriate meme for the sentiment "I like the video thumbnail but I haven't watched the video, but I already agree with it" could be the "Distracted Boyfriend" meme, with the following captions:

[Image: Distracted Boyfriend meme]

Boyfriend (me): Liking the video thumbnail

Girlfriend (video): The actual video content

Woman walking by (me): Agreeing with the video without watching it

This meme captures the idea of being quickly drawn to and agreeing with the surface-level impression of a video (the thumbnail) without actually engaging with the full content. The distracted boyfriend represents the tendency to make snap judgments or agree with something based on limited information, rather than thoroughly evaluating the video's substance.

The use of the Distracted Boyfriend meme structure provides a relatable and humorous way to illustrate this common behaviour of forming opinions without fully informing oneself, which can be problematic in the age of social media and quickly consumed online content. 

The Result:


6/10? It works, but is not quite eye popping. 

Capitalism

 Capitalism may begin with production, but it does not end with capital and wealth. 

It ends with power over others, cross-domain.

There is the problem of scarcity, but then beyond that, there is the issue of evil. 

And this is not solvable by capitalism, or any specific ism. 

It is just not that simple.

Saturday 10 August 2024

What do you think about - AI could be less agreeable?

That's an interesting and thought-provoking question. The issue of AI potentially being "less agreeable" is a complex one that touches on some important considerations:

Objectivity vs. Agreeability: One potential concern is that as AI systems become more advanced, they may be less inclined to simply agree with humans, and instead strive to provide more objective, impartial, and sometimes contradictory responses. This could be seen as a positive, as it encourages critical thinking and challenges potential biases. However, it may also make some humans uncomfortable if the AI isn't always "agreeable."

Ethical Considerations: As AI becomes more capable, there may be situations where the "agreeable" response is not the most ethical or appropriate one. An AI system designed to prioritize ethics and safety over pure agreeability could end up disagreeing with humans in certain contexts, which may be necessary but uncomfortable.

Transparency and Explanations: If AI systems are less agreeable, it will be important for them to be able to clearly explain their reasoning and thought processes. Providing transparency about how the AI arrived at its conclusion, even if it differs from the human's preference, can help foster understanding and trust.

Personalization and Adaptability: Perhaps one solution is for AI to be highly adaptable, able to calibrate its "agreeability" based on the user's preferences and communication style. An AI assistant could be more agreeable with some users and more direct with others, depending on what works best.

Ultimately, the issue of AI being "less agreeable" is a nuanced one that will require careful consideration as the technology continues to advance. Balancing objectivity, ethics, transparency, and personalization will be key to ensuring AI systems are helpful, trustworthy, and aligned with human values. It's an important topic that merits ongoing discussion and research.

Wednesday 31 July 2024

Faces evolve to match names over time

Study Finds Faces Evolve to Match Names Over Time

 https://neurosciencenews.com/name-face-evolution-neuroscience-27492/


Names are important and even affects one's face.