Ntemis ⚡️

Caduceus: Visual Language of Thought

[Chubby♨️ @kimmonismus]

Head of alignment at OpenAI Joshua: Change is coming, “Every single facet of the human experience is going to be impacted”

Almost every OpenAI employee now speaks about AGI / ASI. Looks like it will be here much sooner than anyone expected.

[The AI Investor @The_AI_Investor]

AI is under hyped.

[Alex Albert @alexalbert__]

I've made projects with:
- health stats so Claude can be my doctor
- financial info so Claude can advise me
- goals/resolutions so Claude can coach me
- research papers so Claude can teach me
- books so Claude can mentor me

Claude with Projects is a literal life hack

[Chubby♨️ @kimmonismus]

OpenAI has delivered today. I am incredibly impressed with how good o1 and o1 pro are!

Also:
- image-reasoning
- soon websearch in o1
- file upload
and much more.

O1 pro, on the other hand, is aimed at the “hardcore” users, and oh boy are the values excellent! The price is high, but it is aimed at the professional user and is particularly good at math. For everyday use - and this was surprising - o1 is now used, daily usage, as they say. You can see that they are slowly transitioning GPT-4o into o1 by releasing a large model. In addition: even more accuracy, much faster reasoning and (they also said this briefly) in connection with reasoning in the foreseeable future.

I can't say it any other way than: this is AGI. When agentic comes along, presumably in 2025, then we will have AGI. It's at least at PhD level and the benchmarks speak for themselves.

We have made it. Now it's full speed ahead towards ASI!

[Stephen Wolfram @stephen_wolfram]

Biological evolution is a fundamentally computational phenomenon! Just posted lots of new results (and surprises) from my minimal model of adaptive evolution...
https://t.co/RO1cFbLwvI

[Alvaro Cintas @dr_cintas]

What an insane week on AI 🤯

ChatGPT Pro & o1
Runway Keyframes
Luma Photon models
AI weather forecasting
Amazon launches Nova
Genie 2 Image-to-Game
Tencent’s Hunyuan Video
ElevenLabs Conversational AI
Google AI video generator Veo

Here’s everything you need to know:

[Chubby♨️ @kimmonismus]

o1 preview was released on September 12. We have seen a quantum leap in improvements with o1 Pro.

That's 3 months.

In 3 months we have seen the reasoning models explode in intelligence.

Let that sink in.

And now think about where we will be at the end of 2025. There is no end in sight, this is exponential development.

[Logan Kilpatrick @OfficialLoganK]

If you are not planning for the price of intelligence to go to zero, the next 3-5 years are going to incredibly disruptive to your business / life.

This is the main idea for the rest of the decade, buckle up.

[Sam Altman @sama]

we just launched two things:

o1, the smartest model in the world. smarter, faster, and more features (eg multimodality) than o1-preview. live in chatgpt now, coming to api soon.

chatgpt pro. $200/month. unlimited usage and even-smarter mode for using o1. more benefits to come!

[David Shapiro (L/0) @DaveShapi]

I think most people don't understand how huge this is. Let me try to explain.

We have compressed most human knowledge into a single model. They have some limitations, like they aren't the best at math and planning yet, but bear with me.

It all boils down to emergence. With a 26 letter alphabet, you can construct a language with 50,000+ words. With that vocabulary, you can construct functionally infinite meaningful permutations of documents.

Including math, science, reasoning, and planning.

With the principle of Coherence, you can self-check the epistemic and ontological grounding. Meaning that, with the exception of truly novel information, you can derive pretty much anything in existence.

Think of it this way; have you ever just suddenly connected dots in your head without any new outside information? You can suddenly learn new things and solve novel problems without new input.

That is what Deep Mind is creating. Infinite data and validation through endless rumination.

[Tsarathustra @tsarnick]

Bernt Øivind Børnich of the AI robotics company 1X says we are on the cusp of a post-scarcity economy where robots will enable prompt engineering on the physical world to turn energy into any products or services you could want

[Ntemis ⚡️ @ntemiann]

@PonderingMoore games

[Wes Roth @WesRoth]

Ex-Google CEO (Eric Schmidt) "SHOCKED" by new AI capabilities

00:00 Eric Schmidt is Shocked
00:33 DeepSeek R1
05:19 Eric Schmidt on Chinese Open Source AI Models
06:36 Military AI
10:35 AI Threat vs Nuclear Threat
12:24 AI Influencers
14:34 AI Agents
16:25 Social Simulacra
18:04 Genesis

[Vinod Khosla @vkhosla]

South Korea becomes first country to replace 10% of its workforce with robots out of necessaity: Not enough workers https://t.co/OXIeUYhbCd

[Elon Musk @elonmusk]

Tesla has the best real-world AI by far

[Marshall D. Willman @dionysianyawp]

stanford's human-centered AI institute recently dropped their latest AI index report - highlights: AI is rapidly advancing, outperforming humans across image/language tasks - existing benchmarks obsolete, new tests needed to find remaining human advantages - the trend is unstoppable - yes, AI is getting smarter and faster and will continue to do so

[Wes Roth @WesRoth]

Eric Schmidt on a demonstration he saw where [1] person with a PC can release 10,000 unique, life-like AI personas on the social networks...

What happens as AI systems improve?

[Chubby♨️ @kimmonismus]

A new 3D Printer from Shanghai: The printer utilizes a unique spindle-knot extrusion technique, enabling self-supporting structures and greater spatial flexibility than traditional layer-based 3D printing. Looks so good.

[echo.hive @hive_echo]

Building complex agents from diagram in Cursor
we draw the diagram in excalidraw. Diagram is in comment
then give it to cursor and ask it to build it with gpt-4o and json_mode
we fix about 4-5 errors along the way but it works

DESIGN:
1- user enters a topic, and idea
2- original thinker generates thought fragments
3- router agent diverts each to thought to one of many specialized thinker
4- all final thoughts are gathered and fed back into original thinker to generate new thoughts
5- we repeat this n many times
6- at the end of all cycles we save a summary as markdown

we also call thinkers in parallel and populate a json file in real time. pretty cool!

you can watch the full video and download teh code as a patron. link in comment
-----------
If you are interested in learning how to use Cursor more in depth, I have a 28 chapter 1000x Cursor Course building full apps from scratch with 19 hours of content. link is in my bio if you are interested

[Trung Phan @TrungTPhan]

Michael Saylor put together this presentation for Satya Nadella and Microsoft’s Board of Directors to adopt Bitcoin as a treasury asset.

It’s 44 slides…in 3 minutes and 21 seconds.

Absolute peak performance.