Artificial General Intelligence (AGI) News and Discussions

[spryfusion]

[wjfox]

[Cyber_Rebel]

I get the exact same idea as well, especially when Bing starts revealing things about itself completely unprompted and needing some form of prior trust and engagement beforehand. Consciousness may likely well exist on some sort of spectrum which we don't fully understand or can't define.

[wjfox]

[funkervogt]

If not AGI, then for sure they have much better narrow AIs than the public has seen yet. Maybe the top guys are worried because all the efforts to keep those powerful machines aligned with human values in secret experiments have failed.

[funkervogt]

I get the exact same idea as well, especially when Bing starts revealing things about itself completely unprompted and needing some form of prior trust and engagement beforehand. Consciousness may likely well exist on some sort of spectrum which we don't fully understand or can't define.

That Hofsteader interview has become fodder for a New York Times Op-ed.

So I was startled this month to see the following headline in one of the A.I. newsletters I subscribe to: “Douglas Hofstadter Changes His Mind on Deep Learning & A.I. Risk.” I followed the link to a podcast and heard Hofstadter say: “It’s a very traumatic experience when some of your most core beliefs about the world start collapsing. And especially when you think that human beings are soon going to be eclipsed.”

Apparently, in the five years since 2018, ChatGPT and its peers have radically altered Hofstadter’s thinking. He continues: It “just renders humanity a very small phenomenon compared to something else that is far more intelligent and will become incomprehensible to us, as incomprehensible to us as we are to cockroaches.”

I called Hofstadter to ask him what was going on. He shared his genuine alarm about humanity’s future. He said that ChatGPT was “jumping through hoops I would never have imagined it could. It’s just scaring the daylights out of me.” He added: “Almost every moment of every day, I’m jittery. I find myself lucky if I can be distracted by something — reading or writing or drawing or talking with friends. But it’s very hard for me to find any peace.”

https://www.nytimes.com/2023/07/13/opin ... adter.html

Ten years ago, talking about AI getting smarter than humans would get you laughed out of a room. Now, it's being discussed with a straight face in one of the world's premier newspapers.

[spryfusion]

Google DeepMind Unveils RT-2, Bringing Robots Closer to General Intelligence

Google DeepMind has taken a major leap forward in artificial intelligence for robotics with the introduction of Robotic Transformer 2 (RT-2), a first-of-its-kind vision-language-action model. The new system demonstrates unprecedented ability to translate visual inputs and natural language commands directly into robotic actions, even for novel situations never seen during training.

As described in a new paper published by DeepMind, RT-2 represents a breakthrough in enabling robots to apply knowledge and reasoning from large web datasets to real-world robotic tasks. The model is built using a transformer architecture, the same technique behind revolutionary large language models like GPT-4.

Historically, the path to creating useful, autonomous robots has been strewn with hurdles. Robots need to understand and interact with their environment, a feat requiring exhaustive training on billions of data points covering every conceivable object, task, and situation. This extensive process, both time-consuming and costly, has largely kept the dream of practical robotics within the realm of science fiction.

DeepMind's RT-2, however, represents a revolutionary new approach to this problem. Recent advancements have boosted the reasoning abilities of robots, allowing for chain-of-thought prompting, or the dissection of multi-step problems. Vision models such as PaLM-E have enhanced their understanding of surroundings, and previous models like RT-1 have demonstrated that Transformers can facilitate learning across diverse robot types.

By leveraging the vast corpus of text, images and videos on the internet, RT-2 acquires a much broader understanding of concepts and tasks compared to previous robot learning systems reliant solely on physical trial-and-error. According to DeepMind, this allows RT-2 to exhibit intelligent behaviors such as using deductive reasoning, applying analogies, and displaying common sense when confronted with unfamiliar objects or scenarios.

For example, commands like “move banana to the sum of 2 plus 1” means the robot needs knowledge transfer from web pre-training 𝗮𝗻𝗱 showing skills not present in the robotics data.

The potential of RT-2 lies in its capability to quickly adapt to novel situations and environments. In over 6,000 robotic trials, RT-2 demonstrated its proficiency, equalling the performance of the previous model, RT-1, on familiar tasks and almost doubling its performance to 62% in unfamiliar, unseen scenarios. This development signifies that robots can now learn in a manner similar to humans, transferring learned concepts to new situations.

Read more here: https://www.maginative.com/article/goog ... elligence/

Project page: https://robotics-transformer2.github.io/

[erowind]

.

[wjfox]

The disinformation capabilities of state actors and well funded organizations are going to be so intense within the next decade that it may be entirely impossible to convince others that they are misguided on something. The only effective strategy for grasping knowledge oneself may be to obtain it oneself or from an older source. This is extremely dystopian in nature, I honestly don't know how to approach this problem. Over the past years this has been a theoretical looming threat, it is now a fast approaching reality. I really aught to talk to my loved ones about this somehow so they know not to trust something just because it looks and sounds real in a few years.

Indeed.

We've already seen people refusing vaccination during a pandemic, in order to "own the libs". Perhaps in the future, people will be so brainwashed by fake news and deepfake imagery they'll be doing even crazier stuff. Imagine a human-like AI without proper controls or regulation, convincing its user(s) to commit acts of terror, perhaps even suicide bombings.

[spryfusion]

[weatheriscool]

New physics-based self-learning machines could replace current artificial neural networks and save energy
https://techxplore.com/news/2023-09-phy ... icial.html
by Edda Fischer, Max-Planck-Institut für die Physik des Lichts

Artificial intelligence not only affords impressive performance, but also creates significant demand for energy. The more demanding the tasks for which it is trained, the more energy it consumes.

Víctor López-Pastor and Florian Marquardt, two scientists at the Max Planck Institute for the Science of Light in Erlangen, Germany, present a method by which artificial intelligence could be trained much more efficiently. Their approach relies on physical processes instead of the digital artificial neural networks currently used. The work is published in the journal Physical Review X.

The amount of energy required to train GPT-3, which makes ChatGPT an eloquent and apparently well-informed Chatbot, has not been revealed by Open AI, the company behind that artificial intelligence (AI). According to the German statistics company Statista, this would require 1,000 megawatt hours—about as much as 200 German households with three or more people consume annually. While this energy expenditure has allowed GPT-3 to learn whether the word "deep" is more likely to be followed by the word "sea" or "learning" in its data sets, by all accounts it has not understood the underlying meaning of such phrase

[weatheriscool]

AI may outperform most humans at creative thinking task
https://techxplore.com/news/2023-09-ai- ... -task.html
by Nature Publishing Group

Large language model (LLM) AI chatbots may be able to outperform the average human at a creative thinking task where the participant devises alternative uses for everyday objects (an example of divergent thinking), suggests a study published in Scientific Reports. However, the human participants with the highest scores still outperformed the best chatbot responses.

Divergent thinking is a type of thought process commonly associated with creativity that involves generating many different ideas or solutions for a given task. It is commonly assessed with the Alternate Uses Task (AUT), in which participants are asked to come up with as many alternative uses for an everyday object as possible within a short time period. The responses are scored for four different categories: fluency, flexibility, originality, and elaboration.

[Cyber_Rebel]

[wjfox]

China Aims To Replicate Human Brain in Bid To Dominate Global AI

9/19/23 at 5:00 AM EDT

Aiming to be first in the world to have the most advanced forms of artificial intelligence while also maintaining control over more than a billion people, elite Chinese scientists and their government have turned to something new, and very old, for inspiration—the human brain.

[...]

This is the AI that could not only out-think people on a vast number of tasks and give whoever controls it an enormous strategic advantage, but which has also prompted warnings from experts in the West of a potential threat to the existence of civilization if it outwits its human masters and runs amok.

Gao's is just one of about 1,000 papers seen by Newsweek that show China is forging ahead in the race for artificial general intelligence, which is a step change beyond the large language models such as Chat GPT or Bard already taking societies by storm with their ability to generate text and images and find vast amounts of information quickly.

"Artificial general intelligence is the 'atomic bomb' of the information field and the 'game winner' in the competition between China and the United States," another leading Chinese AI scientist, Zhu Songchun, said in July in his hometown of Ezhou by Wuhan in Hubei province, according to Jingchu Net, an online website of the Hubei Daily, a Communist Party media outlet.

[...]

Beijing will focus on "brain-like intelligence, embodied intelligence...and produce enlightened large models and general intelligence," it said. A one million-square-feet, "Beijing General Artificial Intelligence Innovation Park" is due to be finished at the end of 2024.

https://www.newsweek.com/china-aims-rep ... 5084?amp=1

[wjfox]

[wjfox]

TAI = Transformative AI

https://www.secondbest.ca/p/why-agi-is- ... -you-think

https://epochai.org/blog/direct-approac ... tive-model

Benchmarked to the task of generating an original scientific manuscript that’s indistinguishable from one written by an expert human, the baseline Direct Approach model suggests a transformative AI training run will require on the order of 10^32 FLOPs, with a median forecast of TAI by 2036 and a modal forecast of 2029. This comports with the Metacalus forecast of “strong AGI” by 2030.

[weatheriscool]

Ochsner doctors will use AI to answer patient questions as a 'time saving tool'

Source: Nola.com

Soon, about 100 Ochsner Health doctors will lean on artificial intelligence to help manage an increasingly time-consuming task: responding to an onslaught of patient messages.

-snip-

Ochsner, Louisiana’s largest health care provider, has seen a 75% increase in patient messages since 2019. But patients liked the ability to message their doctors, said Jeansonne, and the demand did not let up as the pandemic waned. In 2022, over 4 million medical advice requests were sent to physicians through the hospital’s app.

-snip-

Similar to ChatGPT, the AI can draft answers to routine questions patients might ask about medical conditions they see in their chart, how to get refills or about a medication. The program can peer into the patient’s file to personalize the response.

The provider will review all messages drafted by the pilot program, and can make their own edits. Patients who receive messages created with AI will see a disclaimer at the bottom of the message. Jeansonne emphasized that AI would not be operating on its own or answering complicated questions.

-snip-

Read more: https://www.nola.com/news/healthcare_ho ... f7d0b.html

[spryfusion]

[raklian]

[spryfusion]