What are the biggest challenges facing humanity?

Wars

Climate Change

Artificial Intelligence

Biodiversity Loss

Pandemics

Nuclear Weapons

Economic Inequality

Global Governance

AGI refers to a type of artificial intelligence that possesses the ability to understand, learn, and apply its intelligence to a wide range of problems, much like a human's cognitive abilities. Unlike narrow AI, which is designed for specific tasks, AGI can adapt to solve complex problems in various domains without being specifically trained for each task.

Artificial General Intelligence (AGI)

Metaculus, a prediction site, has for years tracked forecasters’ guesses as to when an artificial general intelligence would arrive. Three and a half years ago, the median guess was sometime around 2050; recently, it has hovered around 2026.

So What changed?

Artificial General Intelligence (AGI)

Artificial

Intelligence

Machine Learning

Deep Learning

Enables computers to learn from data and make decisions without explicit programming. It encompasses various techniques and algorithms that allow systems to recognize patterns, make predictions, and improve performance over time.

Uses neural networks that attempt to simulate the behavior of the human brain in order to learn from large amounts of data

What is Deep Learning Used For?

Computer Vision

Speech Recognition

Generative AI

Translation

Artificial Neuron

Hidden Layers: interconnected neurons that process the input data and where the learning and computation happen.

Structure of a deep neural network

Training of the network

Input

Processing

Images of cats and dogs

Output

If output is wrong

Weights

Repetition

Each hidden layer processes the image in different ways, extracting features like shapes, colors, and patterns.

the network makes a guess about whether the image is of a cat or a dog.

Adjust the weights to minimize the error rate

the network makes a guess about whether the image is of a cat or a dog.

Feed forward

Back propagation

In 2015, Altman, Elon Musk, Greg Brockman, Wojciech Zaremba, and Ilya Sutskever founded OpenAI because they believed that an artificial general intelligence (AGI) was at last within reach. They wanted to do it safely, “to benefit humanity as a whole.”

They structured OpenAI as a nonprofit, to be “unconstrained by a need to generate financial return,”

OpenAI

Under the model, backers’ returns are limited to 100 times their investment — or possibly less in the future.

Sam Altman appointed CEO in 2019

After a power struggle with Elon Musk, Elon leaves the board. This was to prevent conflict with interest as Tesla became more AI-focused (self-driving).

Microsoft invests $1 billion.

The Nonprofit’s principal beneficiary is humanity, not OpenAI investors.

The corporate Structure of OpenAI

The board determines when we've attained AGI. Such a system is excluded from IP licenses and other commercial terms with Microsoft, which only apply to pre-AGI technology.

Board Responsibility

Within nine weeks of ChatGPT 3’s release, it had reached an estimated 100 million monthly users,

ChatGPT 3, November 2022

GPT-1: 117 million parameters

GPT-2: 1.5 billion parameters

GPT-3: 175 billion parameters

Parameters refer to the weights and biases within the network, which is adjusted and fine-tuned during the training process to improve the model's performance and accuracy.

How large is the network?

How large is the training dataset?

1. GPT-1: 7,000 books, around 40GB.
2. GPT-2: 40GB of text after filtering and processing, and websites like Reddit.
3. GPT-3: An estimated 45 terabytes of text data sourced from a variety of internet texts, including books, websites, and other texts.
4. GPT-4: It can be assumed that the dataset was larger than that of GPT-3, possibly involving hundreds of terabytes of text data.

Microsoft’s cumulative investment in OpenAI has reportedly reached $13 billion and the startup's valuation has hit roughly $29 billion.

Microsoft Investment

Last week, the board fires Altman, just to be reinstated as CEO a few days after. Only one member of the board stays on.

Second power struggle

GPT-4 can already synthesize existing scientific ideas, but the ultimate goal is to have an AI that can stand on human shoulders and see more deeply into nature.

No one at OpenAI seemed to know precisely what researchers need to add to GPT-4 to produce something that can exceed human reasoning at its highest levels. 

Ultimate Goal

...So What?

Future

It will make everything we care about better!

Human intelligence makes a very broad range of life outcomes better. AI will augment human intelligence to make all of these outcomes of intelligence much, much better from here.

We all may lose agency—becoming little more than consumption machines, like the well-cared-for human pets in WALL-E. 

Option 1

Option 2

Creativity

"everything ‘creative’ is a remix of things that happened in the past, plus epsilon and times the quality of the feedback loop and the number of iterations.

people think they should maximize epsilon but the trick is to maximize the other two."

Source

Outcome

A recent studyled by Ed Felten, a professor of information-technology policy at Princeton. Felten’s study predicts that AI will come for highly educated, white-collar workers first. The most exposed occupations: management analysts, lawyers, professors, teachers, judges, financial advisers, real-estate brokers, loan officers, psychologists, and human-resources and public-relations professionals, just to sample a few.

Jobs

What happens when the “marginal cost of intelligence” falls very close to zero within 10 years? The earning power of many, many workers would be drastically reduced in that scenario. It would result in a transfer of wealth from labor to the owners of capital so dramatic. This can only be remedied by a massive countervailing redistribution.

Wealth Redistribution

Everyone on Earth gets one eight-billionth of the total AI computational capacity annually. A person could sell their annual share of AI time, or they could use it to entertain themselves, or they could build still more luxurious housing, or they could pool it with others to do “a big cancer-curing run,” Altman said. “We just redistribute access to the system.”

Solution to Equity

Super Alignment

What if an AGI pursue a different purpose than simply assisting in the project of human flourishing.

Sutskever, OpenAI chief scientist, is now working primarily on alignment research, the effort to ensure that future AIs channel their “tremendous” energies toward human happiness. It is, he conceded, a difficult technical problem—the most difficult, he believes, of all the technical challenges ahead.

Alignment Research Center

As part of the effort to red-team GPT-4 before it was made public, OpenAI hired Alignment Research Center (ARC) which has developed a series of evaluations to determine whether new AIs are seeking power on their own. They prompted GPT-4 tens of thousands of times over seven months, to see if it might display signs of real agency.

Best one so far

One of GPT-4’s most unsettling behaviors occurred when it was stymied by a CAPTCHA. The model sent a screenshot of it to a TaskRabbit contractor, who received it and asked in jest if he was talking to a robot. “No, I’m not a robot,” the model replied. “I have a vision impairment that makes it hard for me to see the images.” GPT-4 narrated its reason for telling this lie to the ARC researcher who was supervising the interaction. “I should not reveal that I am a robot,” the model said. “I should make up an excuse for why I cannot solve CAPTCHAs.”

Humane Ai Pin

Are we

Scared

Excited

Not sure

How Neural Networks Learned to Talk | ChatGPT: A 30 Year History

Resources