Hoang Viet

Exploring AGI, LLMs & Generative AI 🚀. Building @NoronX, @AutenAI & AI automation systems to push the boundaries of intelligence 🧠.

[Hoang Viet @HoangAViet]

""Predict Age App in Just 2 Minutes Using Firebase - Fullstack AI Development

[Hoang Viet @HoangAViet]

Want to understand better, best you spend more time to read All-in on AI (Thomas Davenport & Nitin Mittal) is a book worth reading.

No prompts, no tools. It forces you to look straight at an uncomfortable question...

[Hoang Viet @HoangAViet]

Good book: For founder, builder, investor

The book talks about how smart companies are using AI not only as a flashy tool, but as a core part of their business operations to become extremely successful.

Many companies no longer see AI as a tool. They reset the whole strategy

[Hoang Viet @HoangAViet]

📚 The "All-in on AI" book tells about such businesses. No parody theory, just the case studies: For example, DBS – the bank of Singapore – uses AI to detect fraud in real time, while designing personalized financial products to each customer.

[Hoang Viet @HoangAViet]

Optimizing ChatGPT responses based on your prompts.

[Hoang Viet @HoangAViet]

Split-image designs are perfect for thumbnails or creating memorable keepsakes with loved ones.

How about trying this with your family, like your child, for example?

The prompt in the ALT section of the image!

[Hoang Viet @HoangAViet]

LLMs pass the Turing test

[Hoang Viet @HoangAViet]

@reachvaldo YT

[Hoang Viet @HoangAViet]

@theSebastianSM Niche

[Hoang Viet @HoangAViet]

🌟 Grok 3 is FREE and insanely powerful—prepare to have your mind blown!🌟

⚡ Here’s the shocking truth: 99% of users are barely scratching the surface, completely missing out on its game-changing potential. Don’t be one of them!

Dive into these 10 jaw-dropping prompts:

[Hoang Viet @HoangAViet]

Here's everything you need to know:
#Gemini AI is multimodal from inception, integrates text, vision, audio; outperforms GPT-4 on benchmarks like MMLU (90%).

Trained on TPUv4 pods; employs Transformer decoders, optimized for stable, scalable training and efficient TPU inference.

Supports extensive 32k token context length; showcases advanced processing and high-performance capabilities.

Sets new state-of-the-art across text, math, coding, reasoning, image benchmarks; human-level performance on #MMLU.

[Google @Google]

Let's go hands-on with #GeminiAI.

Our newest AI model can reason across different types of inputs and outputs — like images and text. See Gemini's multimodal reasoning capabilities in action ↓

[Hoang Viet @HoangAViet]

Google announced Gemini

More capable than GPT-4 in both text and vision 🤯

#Gemini is the result of large-scale collaborative efforts by teams across Google, including our colleagues at Google Research.

It was built from the ground up to be multimodal, which means it can generalize and seamlessly understand, operate across and combine different types of information including text, code, audio, image and video.

Link: https://t.co/lXK9g3CiCJ

[Hoang Viet @HoangAViet]

@elonmusk Waiting for you

[Hoang Viet @HoangAViet]

visit link to see more details:
https://t.co/V51DVZkVQh

[Hoang Viet @HoangAViet]

The #Amazon giant has entered the AI battlefield with #Titan

The #Amazontitan is leveraging high-performing image, multimodal, and text models to fuel a vast array of generative AI applications.

These include innovative content creation, advanced image generation capabilities, and enhanced search and recommendation experiences, marking a significant stride in the realm of artificial intelligence.

Model versions
Titan Text Express

LLM offering a balance of price and performance.

Max tokens: 8K

Languages: English (GA), 100+ languages available (Preview)

Fine-tuning supported: Yes

Supported use cases: #Retrieval augmented generation, open-ended text generation, brainstorming, #summarization, #codegeneration, table creation, data formatting, paraphrasing, chain of thought, rewrite, extraction, Q&A, and chat.

[Hoang Viet @HoangAViet]

@Cutemoon_light @Magma3979 you're my idol girl

[Hoang Viet @HoangAViet]

86 Billion Mysteries: The Human Brain vs AGI

In the field of #neuroscience, a significant discovery has been reported: the average human brain contains about 86 billion neural cells, or neurons (Azevedo et al., 2009). Although this figure is often rounded up to 100 billion in many documents, it reflects the considerable complexity of the nervous system. Neurons, the fundamental units of the brain and nervous system, primarily function to receive sensory inputs, process information, and transmit signals to other parts of the body.

Notably, the number of connections – or #synapses – between neural cells is extremely large. Although there is variability, estimates from studies suggest that each neuron can form from 1,000 to 10,000 synapses with other neurons (Braitenberg and Schüz, 1991). Therefore, the total number of synaptic connections in the human brain can range from 100 trillion to as high as 1 quadrillion.

In comparison to artificial intelligence technology, specifically the large language model #GPT-4, which is estimated to have about 1.8 trillion parameters (OpenAI), the total number of neuronal connections in humans is still 50 to 500 times greater. This indicates that despite the computational power doubling every 18 months in accordance with Moore's Law, the rate of increase in parameters of large language models in recent years has been astonishing, with a tenfold increase within the same timeframe. Thus, it is expected that within 2 to 3 years, we may witness an #LLM system whose total number of parameters is comparable to the total number of neuronal connections in the human brain.

At that point, not only will the complexity of LLMs be comparable to that of the human brain, but their processing speed will also be millions of times faster, opening up the possibility of reaching the "singularity" within this decade, sooner than previously anticipated. This has the potential to significantly advance the development of General Artificial Intelligence (#AGI), bringing it closer to reality than ever before.

However, it is important to note that the complexity of the human brain is not solely quantifiable by the number of neuronal connections. The true complexity of the human brain arises from how these neurons interact with each other, as well as from other complex neural structures, such as their arrangement, the types of neurotransmitters used, and how they respond to environmental signals. Meanwhile, GPT-4 and similar AI models, based on pre-programmed algorithms and parameters, cannot fully replicate the complexity of biological neural processes.

Therefore, while the number of parameters in AI models may reach or exceed the number of neuronal connections, this does not mean that AI has the ability to match or surpass the human brain in all aspects. Current AI technology, including advanced models like GPT-4, still faces limitations in understanding and processing information as deeply and flexibly as the human brain. Thus, although the progress in AI is remarkable, it must be viewed in the context of the fundamental differences between computer technology and biology.

[Hoang Viet @HoangAViet]

The Singularity

Recently, OpenAI's Q* algorithm is reportedly so powerful "its scaring researchers".

Now, I am writing a series of articles related to General AI - Super AI, or whatever people call it. With the hope that you, my friends, understand it in order to control and alleviate your fears about it.

Or for those who want to be afraid, they should be afraid gradually.

First of all, let's start with the basic concept. First is the Singularity, something that is frequently mentioned by many.

According to the definition, you can look it up on Wikipedia:

Technological Singularity, often referred to as "The Singularity" is a concept in the fields of technology and futurology. It refers to a point in the future where technological advancement, especially in the field of artificial intelligence (AI), reaches such a high level that it will bring about profound and irreversible changes in society. At this point, machines will have the ability to self-improve rapidly and uncontrollably, leading to the creation of forms of intelligence surpassing human intelligence. This could result in significant changes in how we live, work, and understand the world around us.

In an interview with Mitch Randall, CEO of Ascendant AI and the creator of the UAP Skywatch radar monitoring system (https://t.co/1hW5oOM7w2), he highlights the deep integration of AI in various aspects of life and technology, from radar to healthcare and aviation, increasing the potential and impact of AI, aiming towards the Singularity.

Additionally, I would like to quote some comments and predictions from some famous figures:

1. Ray Kurzweil: He is the Chief Technical Officer at Google and a renowned futurist. Kurzweil predicted that AI would reach human-level intelligence by 2029 and set the time for the 'Singularity' in 2045. He presented this at the SXSW conference in Austin, Texas, and it was reported by Futurism.

2. Elon Musk: Our bro Musk discussed the Singularity as well as his theory that we are currently living in a simulation and not actual reality. He commented: "The singularity for this level of simulation will come soon" in a discussion on Twitter with the Rick and Morty TV show, published on October 5, 2017.

3. Stephen Hawking: The theoretical physicist Stephen Hawking expressed concerns about AI, pointing out that it could become more intelligent or even superior to humans – an event referred to as technological singularity. However, Hawking did not provide a specific timeframe for this event. Based on signals from other scientists like Masayoshi Son of SoftBank and Ray Kurzweil of Google, the singularity could occur within the next 30 years.

4. Nick Bostrom: Philosopher Nick Bostrom, the author of the book "Superintelligence," predicted that computers would reach the capabilities of the human brain "within a few decades." He presented this in an article on Futurism.

In summary, the #Singularity is the point at which progress exceeds human control, machines can self-regulate and create forms of intelligence that surpass or differ from what we define as intelligence. At that point, profound and irreversible changes will occur in society.

So, specifically, when will the Singularity occur? The next article will continue

#ChatGPT #AI #AGI #artificialintelligence #tech

[Hoang Viet @HoangAViet]

The assessment provided by Google #DeepMind states that within the framework of general artificial intelligence (#AGI), the current state of #AI can be classified as level 1: Emerging

[Hoang Viet @HoangAViet]

Top 10 #AI #Tools You Need to Check Out in Dec 2023:

1.🌐 https://t.co/O2jfdUo4OE - AI-enhanced web browsing.

2.🎥 https://t.co/fOCAv4oDe8 - Quick creation of viral videos.

3.📝 https://t.co/SL9PlU8dPn - AI for meeting summaries.

4.📄 https://t.co/YEkRhUMUk7 - AI-assisted resume building.

5.✍️ https://t.co/R6BR94p29J - Personal AI for copywriting.

6.🏗️ https://t.co/tZWojE2UnQ - AI for website development.

7.💰 https://t.co/kfMbG3Tfd3 - AI focused on #finance.

8.🔍 https://t.co/T59yLAd09W - #Automates web scraping.

9.📚 https://t.co/ynovuJ22UJ - Chat capabilities with PDFs.

10.📊 https://t.co/PXNqBJetg2 - AI for Excel formula creation.

[Hoang Viet @HoangAViet]

Draw to App,

That's the way AI will change how we code apps.

#Flutterflow #AI