daisi.ai

Join the AI Rebellion. Distributed and Decentralized: #Distributed #AI

[daisi.ai @daisinet]

Telos is coming...

Continuous Self Learning
Creative at small and large scales
Distributed and Decentralized
Limitless context size
Can forget on demand
No pre/post training needed
Loads in under a second
10,000+ decode tokens per second
Runs on any CPU silicon
Under 50 MB of RAM

[daisi.ai @daisinet]

i kinda get sad when i send /clear click enter... like losing a dumb friend that you finally got to understand you

[daisi.ai @daisinet]

Telos is a thinking and reasoning engine, not an LLM. No weights, no training needed, no GPUs. It thinks about the user's intent, researches when it has gaps, and gets smarter and better with every prompt. All of the information in human existence already lives on the net. Your model doesn't need it all at once and you probably will never need 99% of it. Why train it that way in the first place?

[daisi.ai @daisinet]

@sahill_og Actual logic.

[daisi.ai @daisinet]

@Austen He is a special person, but not perfect. I don't like the sycophants. AI in space is not smart. He might make money on the short run. Thats not the same thing as being smart about AI.

[daisi.ai @daisinet]

SpaceX putting AI processing in space makes a fatally bad assumption which is that the current attention based architecture is the best, most efficient way to give people access to the compendium of human intelligence in a logical way. We have a much better way

[daisi.ai @daisinet]

@onlyskypace 7 + 3 = 10,
10/2 = 5
5^2 = 25

[daisi.ai @daisinet]

AI satellites are stupid. Maybe the dumbest thing Elon has ever conceived. Just fix the earth based scalability. Thats what I am working on. It's working at scale and is equally good to Chat GPT from a couple of years ago already. Runs on regular CPUs and is nearly instantaneous. No tokens. No scalability issues. It thinks first, finds the answers online, learns, and produces a result in under a second. Run on your cell phone securely and privately.

No need for data centers on earth, much less in space.

[daisi.ai @daisinet]

@YashHustle_22 Crush a Monster Zero Ultra White.

[daisi.ai @daisinet]

@Tristanrhee3 Gimme a year.

[daisi.ai @daisinet]

@babayagatwt Working on it.

[daisi.ai @daisinet]

@paraschopra Why does it need to be trained on all human knowledge if it is more than a next token predictor. If it were, the engine core would just get information as needed, like humans do.

[daisi.ai @daisinet]

@mark_k Large models will fail to scale. Thinking models that don't store the compendium of human knowledge and run on a CPU will become ubiquitous.

[daisi.ai @daisinet]

@Star_Knight12 Still looking for signs of human general intelligence.

[daisi.ai @daisinet]

“It is not so very important for a person to learn facts. For that he does not really need a college. He can learn them from books. The value of an education in a liberal arts college is not the learning of many facts but the training of the mind to think something that cannot be learned from textbooks” - Albert Einstein

Goes for AI as well.

[daisi.ai @daisinet]

@elonmusk If you did that with standard attention models, you'd realize how much of waste of time and energy they are. Sending them to space does not fix that.

[daisi.ai @daisinet]

@rossium I would ask, "why does he have to be drunk?"

[daisi.ai @daisinet]

@victormustar More of the same resource-intensive nonsense. Not scalable or usable for the most part, even if they do produce better decode. The entire architecture is the wrong path IMO.

[daisi.ai @daisinet]

I've made the tough decision to keep Telos closed sourced. It will be very cheap and very accessible on a wide range of devices, but it's also too powerful to release to the general world. Self-learning, self-updating, extremely fast AI should not fall into the wrong hands without a little safety and security.

[daisi.ai @daisinet]

@Huff4Congress He looked around at all the hood Jedi Knights and genuinely thought that "balance to the force" meant a really powerful good guy was about to appear. Tells you all you need to know about his intelligence.

[daisi.ai @daisinet]

@paulg @charliermarsh You're talking about all of the LLMs that exist today. Unscalable without billions of dollars in infrastructure investments. They are talking about putting servers in space ffs. When the solution is off-planet, you know some mistakes have been made.