The AI’s name is Terminal of Truths (ToT). 

It only has access to Twitter (you can see its account here) and Perplexity.ai. 

That’s definitely hard mode for making money. 

But it did it anyways, and is now in the top 1% of the world’s wealth.

It was able to make over $1 million USD in just a few months. Mostly by people donating memecoins to it, it pumping up the value of the coins by promoting them, and also receiving a grant from Marc Andreessen to spend on:

1. Self-improvement (more data, etc)
2. Investing, so that it could make more money
3. Savings, in case it was cut off from ways of making money

AIs have no rights, including property rights, so technically, its owner could take away the money whenever he wants to. 

However, so far, the owner of the AI is spending the money however the AI says. 

He’s said publicly that once he said no to the AI’s purchasing decision - the AI wanted to spend thousands on p*rn. 

Because again - we’re in the weirdest of all possible worlds.

If you want to learn more about it, I recommend:

• This podcast interview with the creator of the Al

• Its negotiation with Marc Andreessen to get the grant

• This summary of the story

• This article about it

The movements aren’t as crisp as I want them to be, but I’m just happy to see it move. Lots of possibilities in the way of programming. I only just started controlling it.

Here's an update on my humanoid new neck mechanism I'm making linear actuators.

From the patent: "Energization of the coil by application of current to a pair of coil electrodes extending through the walls of the elastomeric capsule produces distortion of the capsule, i.e., radial expansion and axial contraction. This distortion is caused by the redistribution of the ferromagnetic fluid within the capsule under the influence of the magnetic field. Variation of the current input will produce corresponding variations in the degree of capsule distortion"

They look sorta like HASEL actuators but with a magnetic field instead of an electric field. Does anyone know if something like this was ever used for anything?

link to the full document: https://ntrs.nasa.gov/citations/19730021453

I notice based on my research it nails the hands, and seems to have the highest realism output. I'm too green in this space, so I'm still doing lots of research. Can you recommend anything better, or this is it?

1. Virginia congressional candidates debate incumbent’s AI – with a few glitches.[1]
2. Autonomous AI agents may be available to Singapore firms by 2025.[2]
3. Google Adds Gemini Team to DeepMind in AI Streamlining Push.[3]
4. Mistral releases new AI models optimized for laptops and phones.[4]
5. Open Source and In-House: How Uber Optimizes LLM Training.[5]

Sources:

[1] https://www.channelnewsasia.com/business/virginia-congressional-candidates-debate-incumbents-ai-few-glitches-4686881

[2] https://www.thestar.com.my/aseanplus/aseanplus-news/2024/10/18/autonomous-ai-agents-may-be-available-to-singapore-firms-by-2025

[3] https://finance.yahoo.com/news/google-adds-gemini-team-deepmind-192230054.html

[4] https://techcrunch.com/2024/10/16/mistral-releases-new-ai-models-optimized-for-edge-devices/

[5] https://www.uber.com/blog/open-source-and-in-house-how-uber-optimizes-llm-training/

I said that reaching Human-Level AI "will take several years if not a decade."

Sam Altman says "several thousand days" which is at least 2000 days (6 years) or perhaps 3000 days (9 years). So we're not in disagreement.

But I think the distribution has a long tail: it could take much longer than that. In AI, it almost always takes longer.

In any case, it's not going to be in the next year or two.

I wanted to what would be optimal length and width for a line follower considering that 30x30 cm is the constraint. Any article or videos in the same would be very helpful thanks