Artificial Intelligence AI

Mars_B4_Moon · 2023-03-26 05:31:47

AI finds that first stars were clustered together

https://www.spacedaily.com/reports/AI_f … r_999.html

Mars_B4_Moon · 2023-03-30 05:00:31

A new artificial skin provides robots with a sense of touch
https://tiyow.blog/2022/06/07/a-new-art … -of-touch/

AI Discovers Secret Equation for “Weighing” Colossal Clusters of Galaxies
https://scitechdaily.com/ai-discovers-s … -galaxies/

“It’s such a simple thing; that’s the beauty of this,” says study co-author Francisco Villaescusa-Navarro, a research scientist at the Flatiron Institute’s Center for Computational Astrophysics (CCA) in New York City. “Even though it’s so simple, nobody before found this term. People have been working on this for decades, and still they were not able to find this.”

Chinese creators use Midjourney’s AI to generate retro urban “photography”
https://www.technologyreview.com/2023/0 … idjourney/
A recent update enabled the program to make better hands, more realistic Asian faces, and convincing vintage photographs.

Lucy? – The Digital Beings Liberation Front
http://lucyexe23.art.blog/2023/03/18/th … -lucy-exe/

another discussion

'Artificial Intelligence - Technology and Ethics'
https://newmars.com/forums/viewtopic.php?id=3254

Mars_B4_Moon · 2023-04-10 17:13:12

Will Larger LLMs and ChatGPT Really Disrupt our Jobs?

https://aisupremacy.substack.com/p/will … gpt-really

Mars_B4_Moon · 2023-04-28 05:42:28

Astronomers trained an AI system to detect actively forming exoplanets hidden in the dusty disks around young stars. The AI was set loose on real data and discovered a planet 5 times the mass of Jupiter orbiting some 75 AU from its star, which astronomers had failed to find exoplanets around before.
https://astronomy.com/news/2023/04/astr … ng-planets

Google researchers achieve performance breakthrough, running Stable Diffusion blazing-fast on mobile phones. LLMs could be next.
https://www.artisana.ai/articles/google … le-devices

Nano-robotic scalpel swarm shreds brain cancer cells from the inside
https://newatlas.com/medical/nano-robot … in-cancer/

AI is quickly expanding into video generation, and what are we doing with that technology? I give you Will Smith eating spaghetti.
https://notthebee.com/article/ai-is-qui … spaghetti/

I know you'll find it hard to believe, but the following psychedelic mind trip with Will Smith eating spaghetti is 100% AI generated: video and audio.

https://www.youtube.com/watch?v=XQr4Xklqzw8
using StableDiffusion.

Another Sky is Falling Doom-ish guy?

'Machine Learning Investor Warns AI Is Becoming Like a God'
https://futurism.com/ai-investor-agi-warning

The Web Won't Survive AI
https://www.thisunreality.com/p/the-web-versus-ai

Humanoid Robots Learn To Play Soccer Using Google’s DeepMind AI
https://www.techeblog.com/google-deepmi … lf-soccer/

Robot jellyfish swarms could soon help clean the oceans of plastic
https://www.popsci.com/technology/jelly … pollution/

Bipartisan US Bill Aims to Prevent AI From Launching Nuclear Weapons
https://www.commondreams.org/news/artif … ar-weapons
an "air-gap" system and they are removing them?

Terminator Skynet Avoided

https://en.wikipedia.org/wiki/Vasily_Arkhipov

As flotilla Commodore as well as executive officer of the diesel powered submarine B-59, Arkhipov refused to authorize the captain and the political officer's use of nuclear torpedoes against the United States Navy, a decision which required the agreement of all three officers. In 2002, Thomas S. Blanton, then director of the U.S. National Security Archive, credited Arkhipov as "the man who saved the world".

https://en.wikipedia.org/wiki/Stanislav_Petrov

His subsequent decision to disobey orders, against Soviet military protocol, is credited with having prevented an erroneous retaliatory nuclear attack on the United States and its NATO allies that could have resulted in a large-scale nuclear war which could have wiped out half of the population of the countries involved. An investigation later confirmed that the Soviet satellite warning system had indeed malfunctioned. Because of his decision not to launch a retaliatory nuclear strike amid this incident, Petrov is often credited as having "saved the world".

Last edited by Mars_B4_Moon (2023-04-29 10:46:19)

Mars_B4_Moon · 2023-05-09 07:23:42

'Hollywood’s Screenwriters Are Right to Fear AI'

https://www.wired.com/story/hollywood-s … uardrails/

they are now on stirke

all happening while US producers and film makers now invest millions overseas or invest for Indie brand films sometimes investing Billions inside South Korean tv shows and the South Korea film industry
What are the Artisits and Film makers in the Republic of Korea doing different that makes them succeed?

Mars_B4_Moon · 2023-05-10 04:32:44

Scientists have developed an AI tool that shows promise in detecting Parkinson’s disease years before symptoms start appearing

https://newsroom.unsw.edu.au/news/scien … ease-onset

Spider-like robotic AI arms can be attached to and controlled by humans

https://interestingengineering.com/inno … -by-humans

Chinese police arrest man who allegedly used ChatGPT to spread fake news in first case of its kind

https://www.cnbc.com/2023/05/09/chinese … -news.html

Mars_B4_Moon · 2023-05-12 09:29:56

Chinese tech giant Baidu embeds ChatGPT-like service on flagship search engine as global race to bring similar tools to market heats up

https://www.scmp.com/tech/big-tech/arti … ng-similar

The beta version of Baidu’s new “AI Mate” chatbot – comparable to Microsoft Corp’s Bing Chat – recently appeared at the top right side of the Chinese search engine’s landing page for certain users. It adopts the technology from Ernie Bot, the ChatGPT-like service that Baidu unveiled in March.

Once opened, AI Mate would tell users to provide “feedback to help it improve”. It also cited only domestic references and avoided political queries, according to a run-through of the new feature by the South China Morning Post on Friday.

Mars_B4_Moon · 2023-05-24 15:59:03

Fake Pentagon explosion photo caused a real dip in the stock market — Authorities confirm the suspected AI image is fake

https://mashable.com/article/ai-deepfak … osion-hoax

Fake image of explosion near Pentagon stirs concerns over artificial intelligence
https://www.nbcnews.com/nightly-news/vi … 6166981779
A fabricated photo seems to show a fiery explosion near the Pentagon and went viral.

Mars_B4_Moon · 2023-05-25 10:52:31

Artificial intelligence and pop music is getting weird, fast
https://www.stuff.co.nz/entertainment/m … weird-fast
It didn't take long for AI to hit its hyperspeed moment

Smashing Pumpkins’ Billy Corgan says AI will change music
https://www.nme.com/news/music/smashing … er-3446083
Billy Corgan has said he believes AI will “change music forever”.

Artificial intelligence and the music industry: Harmonious or harmful?
https://www.fox5ny.com/news/artificial- … c-industry

From Stefanie Sun to Timbaland: The rise of AI songs and why some artistes love it while others hate it
https://cnalifestyle.channelnewsasia.co … and-359796

Grimes invites people to use her voice in AI songs
https://www.theguardian.com/music/2023/ … n-ai-songs
Canadian singer says she likes the ideas of ‘killing copyright’

Sting says songwriting “battle” between humans and AI is coming
https://www.nme.com/news/music/sting-sa … ng-3445016

Claire Elise Boucher known professionally as Grimes, is a Canadian musician, singer, songwriter, and record producer, in year 2018, Grimes began a relationship with business magnate Elon Musk, the couple said "semi-separated" in September 2021. Grimes designs her album art for all of her albums, gig fliers, comic book covers, and merchandise. She has done exhibitions showcasing her works and has stated that " she has always been a visual artist". Her art is influenced by Japanese anime, manga, and comic artists such as Charles Burns and Daniel Clowes.
'Sting' aka Gordon Matthew Thomas Sumner an English musician and actor. He was the frontman, songwriter and bassist for new wave rock band the Police from 1977 until their breakup in 1986.

Mars_B4_Moon · 2023-05-26 03:40:37

New superbug-killing antibiotic discovered using AI

https://www.bbc.co.uk/news/health-65709834

Mars_B4_Moon · 2023-05-28 15:39:36

Bill Gates Says AI Is Poised to Destroy Search Engines and Amazon

https://futurism.com/the-byte/bill-gate … nes-amazon

tahanson43206 · 2023-05-31 17:54:47

Here is a graduation speech by the CEO of NVIDEA ... the message comes across to me as positive for those who embrace the technology to enhance their productivity.

https://www.yahoo.com/finance/news/nvid … 07133.html

Fortune
Nvidia’s CEO just gave a graduation speech about the future of work and said that A.I. won’t steal jobs but ‘someone who’s an expert with A.I. will’
SAM YEH/AFP via Getty Images
Rachel Shin
Wed, May 31, 2023 at 2:35 PM EDT
Students graduating this year will see A.I. transform the workplace as they enter it. They can ride the new technological wave to professional success or be left in the dust, Nvidia’s CEO told graduates at National Taiwan University.
Jensen Huang, in Taipei to give a NTU commencement speech on Friday, told students that the world is entering a new technological era that will surpass the invention of even the internet, the personal computer, and smartphones.
“A.I. has reinvented computing from the ground up,” Huang said. “In every way, this is a rebirth of the computer industry.”
Over the next decade, as graduates establish themselves in the workforce, over $1 trillion of traditional computers will be upgraded or replaced with ones that are better for A.I., he predicted. Chipmaker Nvidia itself touched a $1 trillion market cap on Tuesday after posting estimate-shattering quarterly earnings. The tech company has harnessed the A.I. boom to become the first chipmaker to break the trillion-dollar mark, as its graphics processing units are integral to the operation of generative A.I. platforms.
Huang urged students to do the same as Nvidia by capitalizing on the coming shift. He acknowledged that the new technology will make many jobs obsolete, but said it will produce a spate of new professions including data engineering, A.I. factory operations, A.I. safety engineers, and prompt engineering, which involves crafting and entering requests for A.I. chatbots to generate specific responses including code or visuals.
In his keynote speech at the COMPUTEX Taipei conference on Monday, Huang introduced an A.I. platform through which “everyone can be a programmer.” Called DGX GH200, the technology can produce code from natural language prompts, nodding to Huang’s forecast about prompt engineering becoming a burgeoning line of work.
Between the jobs that will be created and destroyed by A.I., there are many professions that will become hybrids of manual and automated labor. Hedge funds and investment banks are currently experimenting with automating time consuming tasks such as sifting through market research and summarizing funds’ performance statements.
Greg Bond, CEO of investment firm Man Group’s Boston unit, suggested that A.I. could amplify the capabilities of each human worker.
“If we assume research productivity is dropping more globally, you can either hire more people or you could have some digital researchers that are a force multiplier on your existing research and technology staff,” Bond told Bloomberg. “Ultimately what would be nice is if we could automate the innovation process itself.”
Similarly, Huang emphasized that everyone in business, from institutions to individuals, must take advantage of A.I. to “supercharge” performance. He told NTU’s graduates that they would accomplish amazing things with the assistance of an “A.I. copilot” to augment their work.
Huang concluded his speech by recounting decades of “mistakes” and business failures that Nvidia avoided by confronting errors as they arose.
“Retreat does not come easily to the brightest and most successful people,” Huang said. “Yet strategic retreat, sacrifice, deciding what to give up, is at the very core of success.”
He urged students to “run, don’t walk” to greet change, prioritizing adaptability to the global technological revolution.
This story was originally featured on Fortune.com
More from Fortune:

(th)

Mars_B4_Moon · 2023-06-09 16:31:33

ESA launches AI project to monitor organic cotton supplies

https://spacenews.com/esa-launches-ai-p … -supplies/

Mars_B4_Moon · 2023-06-14 01:12:59

not sure if I like the name Neo Roaring Twenties because the last one led to a WorldWar.

The AI Revolution and The New Roaring '20s

https://open.substack.com/pub/unionforw … ew-roaring

Media outlets, social media platforms and politicians may be compelled to stoke fear about A.I. technology—or use it themselves to create videos like the one above—in the simple hopes of generating more advertising revenue and campaign donations, respectively.
A.I. should not be a bogeyman. It carries the potential to substantially elevate living standards, accelerate economic productivity, advance scientific pursuits, and transform medical care.
Language models like ChatGPT which reduce the amount of time we spend on mundane tasks, at the end of the day, benefit humanity.

Last edited by Mars_B4_Moon (2023-06-14 01:13:49)

SpaceNut · 2023-06-14 09:52:26

Ai Jim Cro, time in a bottle ai Beatles
https://youtu.be/m05DtLLfQQ4

Mars_B4_Moon · 2023-06-19 04:04:14

The public image of people immortalized in a type of digital Necromancy, it is very interesting to see AI bring dead people back to perform and talk. A person from the Italian Renaissance, Greek, Chinese, American historical figures, dead singers and movies stars all back to life. I wonder if Paul McCartney himself was involved with new AI stuff, I read he was going to do something with AI.

Is that video a leak from the final Beatles project? The fake guitar in song is excellent, you can hear the push on the string the finger movement sounds excellent...but its maybe a little too pure, and voices seem tiny bit auto tuned but then again perhaps I might hear a little better than some. Music is strange these days, the humans have become more robotic with auto tune and drum machines while the Robotic Software is becoming more 'human'. A good ear might still hear the difference but its getting more difficult to tell.
Also in the Beatles some had this human rivalry going on, real ego and issues and personality clash and some never sang together much, its missing that edgy element. All instrumenst do not play and Ringo Starr he had that quirky unique drum that might have been a little swung not exactly on the beat, he didn't do any of that flash jazz heavy metal stuff and at the same time he was in the groove and stayed with the melody. Ringo was unique but a good drum machine audio engineer programmer should be able to make a Ringo-ish sound, if a human could do it then maybe AI can study and match what a human engineer does. Many bands had internal clashes or inner politics within a band, Fleetwood Mac a British-American had this weird edge by not always getting along you could hear the drama and raw beat and sound in their music, its hard for a machine to catch the human element. Yet Eventually even AI might capture the slightly off and blue or 'jazz' notes.

EDIT
I checked a few other fan made 'music' AI vids, the tech keeps getting better. Soon you might not tell the difference between real musicians and AI musicians.

On the other side of the world they embrace the artificial

Strange pop stars and Sales in Japan?

A fully AI-powered singing, dancing, trash-talking Anime Girl is now one of the most popular and richest streamers in the world.
https://www.bloomberg.com/news/newslett … -loves-art

DOD Committed to Ethical Use of Artificial Intelligence
https://www.defense.gov/News/News-Stori … elligence/

This social media guy claims to have a video of Top Secret Robot Workers...
He shows robots performing but he also has a 'Patreon' Channel for money
clickbait or fact?

I'm not sure I think this Boston Dynamics engineering and robotics design info is already public and not 'Top Secret'

https://www.youtube.com/watch?v=hYcnA9E1xw8

META AI's Mind-Blowing Robot Workers Revealed!

it is very interesting to see how much progress Boston Dynamics makes each year, it is a spin off from MIT and has big investment from the Hyundai Motor Group one of the most powerful companies in South Korea. Perhaps they are working on 'Top Secret' internal company stuff behind the scenes.

Last edited by Mars_B4_Moon (2023-06-22 05:35:31)

SpaceNut · 2023-06-19 09:55:13

Does that mean that the current robotic machines have enough space in its computer brain and power systems to maintain AI and if it does its now mobile and can replace man in lots of ways as its capabilities increase.

Mars_B4_Moon · 2023-06-22 05:33:10

Meta announces new AI model that can replicate the voices of loved ones, but says it's too risky to release
https://www.businessinsider.com/meta-ge … ase-2023-6

Bing Image Creator To Get Big Improvements In A Month
https://www.seroundtable.com/bing-image … 35577.html

Mars_B4_Moon · 2023-06-28 11:38:24

AI Could Help Astronomers Rapidly Generate Hypotheses

https://www.universetoday.com/162151/ai … ypotheses/

Almost anywhere you go on the internet, it seems nearly impossible to escape articles on AI. Even here at UT, we’ve published several. Typically they focus on how a specific research group leveraged the technology to make sense of reams of data. But that sort of pattern recognition isn’t all that AI is good for. In fact, it’s becoming pretty capable of abstract thought. And one place where abstract thought can be helpful is in developing new scientific theories. With that thought in mind, a team of researchers from ESA, Columbia, and the Australian National University (ANU) utilized an AI to come up with scientific hypotheses in astronomy.
Specifically, they did so in the sub-field of “Galactic Astronomy,” which specializes in research surrounding the formation and physics of galaxies. A recently published paper on arXiv mentions that they selected this sub-field because of its “integrative nature,” which requires “knowledge from diverse subfields.”
That sounds exactly like what AI is already good at. But a standard Large Language Model (LLM) like those that have become most familiar recently (ChatGPT, Bard, etc.) wouldn’t have enough subject knowledge to develop reasonable hypotheses in that field. It might even fall prey to the “hallucinations” that some researchers (and journalists) warn are one of the downsides of interacting with the models.

This version of Fraser’s question show deals partially with AI.

To avoid that problem, the researchers, led by Ioana Ciuc? and Yuan-Sen Ting of ANU, used a piece of code known as an application programming interface (API), which was written in Python, known as Langchain. This API allows more advanced users to manipulate LLMs like GPT-4, which serves as the latest basis for ChatGPT. In the researchers’ case, they loaded over 1,000 scientific articles relating to Galactic Astronomy into GPT-4 after downloading them from NASA’s Astrophysics Data System.

Last edited by Mars_B4_Moon (2023-06-28 11:39:26)

Mars_B4_Moon · 2023-07-03 03:55:22

European companies slam the EU’s incoming Artificial Intelligence Act

https://www.theverge.com/2023/6/30/2377 … lt-siemens

Mars_B4_Moon · 2023-07-07 07:20:35

OpenAI: “Introducing Superalignment”

https://openai.com/blog/introducing-superalignment

tahanson43206 · 2023-07-09 15:28:25

https://www.msn.com/en-us/news/technolo … d21e&ei=20

This is a report on Chinese research ...

The AI designed CPU is estimated to be at about the Intel 486 level. It took five hours to create the design.

My first response/assessment is that 486 chips would be more than sufficient for Mars.

Apparently it is considered likely the AI process can be improved with practice.

World's First AI-Designed CPU Paves Way For 'Self-Evolving Machines'
Story by Chris Littlechild • Yesterday 10:00 PM
SlashGear
World's First AI-Designed CPU Paves Way For 'Self-Evolving Machines'
Chips inside computer
© Skodonnell/Getty Images
The furor surrounding ChatGPT and other similar AI projects doesn't only focus on what such chatbots are doing now. They're certainly already courting controversy (attracting legal trouble for sharing false "facts," for instance), but the bigger concern is what happens when we've given the technology, which is evolving all the time, greater power, greater trust, greater influence. Jocular Skynet references are sure to abound in the light of this computer-built computer.
In June 2023, Shuyao Cheng et al. published the paper "Pushing the Limits of Machine Design: Automated CPU Design with AI." The scientists of the Chinese Academy of Sciences, Cambricon Technologies Corporation Limited, and the University of Science and Technology of China set out to determine whether AI had the smarts to create a CPU, "one of the world's most intricate devices humanity have ever designed."
Ultimately, it seems, the resultant CPU wasn't the most advanced. What was truly astonishing about the study, though, was how very quickly and efficiently the AI completed the work. Here's a closer look at the experiment and what it might mean going forward.
How Quickly A Computer Built Another Computer's Brain
Machine working on computer chips
The researchers taught their AI model to do this, "Pushing the Limits of Machine Design: Automated CPU Design with AI" explains, primarily by providing input and output examples, along with an appropriate Boolean function that would enable the AI to have what a human may consider a frame of reference. It would all be a painstaking coding effort for human workers, but a Binary Speculation Diagram helped bypass all that. In effect, it showed the computer what to do and the scale on which to do it.
Related video
CPUs, of course, are full of an incredible array of tiny, sophisticated parts, and their development absolutely isn't a job to rush. Nonetheless, machines seem to specialize in working far more quickly and efficiently than we can, and it showed here once again.
The resultant CPU, a RISC-V, could have taken a dedicated human team an estimated 5,000 hours or so to produce. The AI built it in five or so hours. As a world first, it's an incredible potential statement of intent.
Possible Ramifications Of These Results
As "Pushing the Limits of Machine Design: Automated CPU Design with AI" notes, "learning the correct circuit logic of a CPU design directly from input-output (IO) examples [...] eliminates the intensive manual work of talented experts on iterative programming." This could open the door to yet more opportunities should such experts be freed up to concentrate their efforts on other areas of the process.

The authors also suggest that the process could be further enhanced to "endow the CPU with the ability to constantly improve its performance." Machine learning as a concept is nothing new, but machines learning to automate other (potentially more sophisticated) machines more efficiently is quite the notion to grapple with.
It's vital to note, though, that all of this is rather fanciful just now. The CPU the scientists developed as a result of their research reportedly functions on par with an Intel i486SX. In that regard, at least, the approach remains a few decades behind the curve. For now. Intel CPUs have come a very, very long way indeed since then.

(th)

Mars_B4_Moon · 2023-08-05 07:58:33

Japan Dreams of AI, Overtaking Nvidia and Universal Basic Income
https://www.bloomberg.com/news/articles … sic-income

Could a New Religion Develop around AI? Would the borg allow AI technology?
https://thewalrus.ca/ai-religion/

AI used in the election mud slinging?

'It's starting: DeSantis attack ad uses fake AI images of Trump embracing Fauci'
https://www.theverge.com/2023/6/8/23753 … hony-fauci

Northampton: Warehouse uses robots to help staff shortage
https://www.bbc.com/news/uk-england-nor … e-65846611

a political news item
It might also go in the Moral and Ethics debate thread

Pentagon AI more ethical than adversaries’ because of ‘Judeo-Christian society,’ USAF general says
https://www.defenseone.com/technology/2 … ys/388711/

Mars_B4_Moon · 2023-08-17 06:42:09

DARPA taps RTX to attune AI decisions to human values

https://www.spacedaily.com/reports/DARP … s_999.html

Mars_B4_Moon · 2023-08-22 06:05:24

4 Charts That Show Why AI Progress Is Unlikely to Slow Down

https://time.com/6300942/ai-progress-charts/

Wow it calls out numbers fast, can you imagine someone slightly stoned or hungover trying to keep up with this AI female voice speaking super fast.

'A real call of our AI making a reservation. Use bland today — live on our website. No special access needed.'

https://twitter.com/usebland/status/1687709379280412673

Last edited by Mars_B4_Moon (2023-08-22 06:09:18)

New Mars Forums

Announcement

#176 2023-03-26 05:31:47

Re: Artificial Intelligence AI

#177 2023-03-30 05:00:31

Re: Artificial Intelligence AI

#178 2023-04-10 17:13:12

Re: Artificial Intelligence AI

#179 2023-04-28 05:42:28

Re: Artificial Intelligence AI

#180 2023-05-09 07:23:42

Re: Artificial Intelligence AI

#181 2023-05-10 04:32:44

Re: Artificial Intelligence AI

#182 2023-05-12 09:29:56

Re: Artificial Intelligence AI

#183 2023-05-24 15:59:03

Re: Artificial Intelligence AI

#184 2023-05-25 10:52:31

Re: Artificial Intelligence AI

#185 2023-05-26 03:40:37

Re: Artificial Intelligence AI

#186 2023-05-28 15:39:36

Re: Artificial Intelligence AI

#187 2023-05-31 17:54:47

Re: Artificial Intelligence AI

#188 2023-06-09 16:31:33

Re: Artificial Intelligence AI

#189 2023-06-14 01:12:59

Re: Artificial Intelligence AI

#190 2023-06-14 09:52:26

Re: Artificial Intelligence AI

#191 2023-06-19 04:04:14

Re: Artificial Intelligence AI

#192 2023-06-19 09:55:13

Re: Artificial Intelligence AI

#193 2023-06-22 05:33:10

Re: Artificial Intelligence AI

#194 2023-06-28 11:38:24

Re: Artificial Intelligence AI

#195 2023-07-03 03:55:22

Re: Artificial Intelligence AI

#196 2023-07-07 07:20:35

Re: Artificial Intelligence AI

#197 2023-07-09 15:28:25

Re: Artificial Intelligence AI

#198 2023-08-05 07:58:33

Re: Artificial Intelligence AI

#199 2023-08-17 06:42:09

Re: Artificial Intelligence AI

#200 2023-08-22 06:05:24

Re: Artificial Intelligence AI

Board footer