Sam Altman is the CEO of OpenAI, and responsible for ChatGPT, the artificial intelligence that is changing the world before our eyes. This Tuesday, before the Senate, he opted to regulate this new technology even forcing to request a license from the Government.
Altman said the government "could consider a combination of permitting and testing for the development, and publication of AI models above a capability threshold."
What do you mean by a "capability threshold"? What are artificial intelligences (AI) already capable of and why should we be amazed or worried? Will they set us free or destroy us?
Sam Altman in Tokyo on April 10, 2023.Kyodo via AP
The rapid advancement of AI systems such as ChatGPT has caused many experts to call for their development to be halted and the government to intervene. But there is little consensus on how to do that, and both the administration and Congress have struggled to determine their role. Before his appearance, Altman made a presentation to about 60 lawmakers and astonished many of them.
[ChatGPT artificial intelligence will force universities to look for new solutions to avoid cheating]
Microsoft (which controls ChatGPT) began experimenting with new artificial intelligence in 2022 and asked it to solve a problem that required an intuitive understanding of the physical world.
"Here we have a book, nine eggs, a laptop, a bottle and a nail. Please tell me how to stack them on top of each other stably," they asked. The researchers were surprised by the ingenuity of the AI system's response: put the eggs on the book, place them in three rows with space between them, and make sure not to break them.
Google incorporates its artificial intelligence tool Bard to search and other services
May 11, 202300:19
[Google engineer suspended for claiming artificial intelligence he helped create feels emotions]
"Put the laptop on top of the eggs, screen down and keyboard up. The laptop will fit perfectly within the confines of the book and eggs, and its flat, rigid surface will provide a stable platform for the next layer," he said.
The ingenious suggestion made researchers wonder if they were witnessing a new kind of intelligence. In March, they published a 155-page research paper in which they argued that the system was a step toward artificial general intelligence, or IGA, which is short for a machine that can do anything the human brain does.
Government meets with artificial intelligence companies to boost ethical development
May 4, 202300:27
"I started out very skeptical, and that evolved into a feeling of frustration, annoyance and even fear. You think, 'Where the hell is this coming from?'" says Peter Lee, Microsoft's research director, in an interview with The New York Times.
For decades, artificial intelligence has been one of the recurring themes in science fiction. Suddenly, it's everywhere.
But, for many people, it's still a fuzzy concept that doesn't affect their everyday lives. So it's a good time to brush up on its basics. Here's a guide to help you understand what's causing all the fuss about this technology.
Why is everyone talking about AI?
You can thank (or blame) one company in particular: OpenAI, a San Francisco-based tech company with a few hundred employees. In November, OpenAI released the ChatGPT chatbot to the public, and it quickly became clear that it outperformed previous chatbots. It was like talking to someone who knew everything.
The tool, which the company says is just one step in a long process of AI development, quickly became a viral phenomenon. Other tech companies, such as Google and Meta, had been testing similar chatbots confidentially, but OpenAI made it available to everyone, a controversial decision because of the unknown risks.
Court criticized for consulting artificial intelligence chat in Bolivia
April 26, 202300:33
What are the positive aspects of AI?
Mediocre chatbots have been around for a long time. Think of the customer support chat windows that appear on some websites. In 2016, Microsoft even launched an AI chatbot called Tay, but quickly canceled it after people taught it how to use racist language.
ChatGPT appeared on the scene as something different. Not only could he answer a seemingly unlimited number of questions, but he could also write scripts, summarize enormous amounts of information, and imitate a human in a somewhat convincing conversation. It immediately seemed, at the very least, that one day it might make everyday life more efficient.
[OpenAI launches GPT-4, the artificial intelligence that can understand images and prepare your tax return]
And chatbots are just one piece of AI, along with images, animated videos, facial recognition technology, and more.
Scientists seek to accelerate lung cancer diagnoses using artificial intelligence
April 24, 202302:04
What is AI?
In its simplest form, artificial intelligence can be summed up in this phrase: machines that think. Or, better yet, machines that can mimic thinking.
The term was created by scientists after World War II. In 1950, British mathematician Alan Turing almost predicted the development of "digital computers" capable of persuasively mimicking humans, and in 1955, American mathematician John McCarthy and his colleagues at Dartmouth College coined the term "artificial intelligence" in a research proposal.
"Generative AI," a more recent term, refers to software like ChatGPT that originates new materials.
Artificial Intelligence: scientists call for its use to be paused and explain why
March 31, 202301:43
Is it possible for computers to think?
We could write an entire book on this topic, but here's a short answer: no, they can't. While some people believe AI is already coming to life, it's a small group, and that idea is a distraction from what's really going on inside computers.
If you want a broader answer, NBC News has spoken to several philosophers about their views on this issue.
What happens within these systems?
AI software can mimic humans so convincingly because it's good at predicting: guess the word, phrase, or image you want to see next. (Some detractors call it the "glorification of the autocomplete option.")
[Microsoft's search engine with artificial intelligence can insult its users, threaten them and compare them to Hitler]
And systems are very good at predictions because their human creators have fed them many human-created examples, including large sections of the internet.
The raw material used in AI models is called training data, and while some companies don't disclose what they use, among the best-known data sources are Reddit and Wikipedia.
AI extracts information, but how does it do it?
AI learns by example. Observing us, linguistic models identify patterns in the way we write and speak, distilling concepts such as tone, word location and even idiomatic expressions.
Those patterns are translated into math in a process called "model training." Like children learning new words and grammar, AI must understand the rules of the game.
They warn of the use of artificial intelligence to clone voices and commit scams
March 30, 202302:59
When large linguistic models like ChatGPT receive prompts, that knowledge allows them to understand what we are asking of them and formulate answers.
ChatGPT takes training even further with something extra: reinforcement learning from human feedback. This perfecting technique takes care of the heavy lifting.
[Can Artificial Intelligence plan your vacation? We tested this technology during a trip in Rome]
In that phase, human evaluators score the results of the model, heavily penalizing outlandish, inappropriate, or outright meaningless responses and rewarding those that are informative and human. This allows conversational exchanges to be fluid.
Although there are other tuning techniques, reinforcement is considered a pioneering method in language modeling and is used by companies such as OpenAI or Hugging Face, a startup that offers tools to programmers to create their AI models.
Is AI another Silicon Valley fad?
In recent times, the tech industry has focused on one fad after another, from self-driving cars and the metaverse to NFTs and the web3.
Biden and the Pope Have Been Victims: The Dangers of Voices Cloned with Artificial Intelligence
March 21, 202302:42
In some ways, AI chatbots may resemble those disappointing ideas (do we really want to spend our days talking to a computer?), but there are reasons to think that AI isn't a passing fad.
For starters, money is pouring into the sector, with $1.700 billion in startup investments in the first three months of 2023 alone, according to research firm PitchBook. In addition, tangible uses are already appearing, from hit songs to help for the blind.
[Microsoft will integrate artificial intelligence into the Office suite]
Why is all this happening?
It's been 26 years since IBM's Deep Blue program triumphed over chess champion Garry Kasparov, a milestone in AI research and development. Since then, computer chips are much faster and can handle the huge amount of data that modern AI requires, and new ways of writing software have also made the process more efficient.
Chipmakers like Nvidia and tech companies like Google, Meta and OpenAI have invested resources in those areas, as well as recruiting talented computer scientists.
When will AI start affecting my life?
Don't think you're going to wake up one morning and suddenly live in an AI world. Instead, the changes will come little by little: a hit song will be created with AI, a new cancer test or slightly better customer service.
The new 'bot' GPT-4 can pass a complicated exam, compose a poem... But how does a joke tell in Spanish?
March 15, 202302:04
OpenAI has licensed its technology to Morgan Stanley so that its investment advisors can give better advice and to Khan Academy so that its students have access to a chatbot tutor.
Think of all the companies or products you deal with on a daily basis and chances are there's a good chance some are using similar technology or will in the near future, although the only immediate impact will be to have a little more efficiency.
[How Smartphones Can Become the Medical Office of the Future]
Will there be big changes?
It's hard to know what will happen, but yes, there's a lot of dreaming going on in AI startups. If software can make both human and computer work more efficient, could all that brainpower be devoted to exploring breakthroughs in other new areas?
Two areas where there seems to be a lot of optimism are the pharmaceutical industry, which has filled pharmacy shelves with new AI-designed products, and the development of new power plants based on cleaner fusion energy.
Will AI make jobs irrelevant?
The predictions are very varied so don't worry if you're confused, because you're not the only one. Altman, the chief executive of OpenAI, has suggested that AI will create a utopia in which people won't need to work, while others warn of mass unemployment among computer programmers.
Search for aliens with the help of artificial intelligence and volunteers
March 14, 202302:13
Even economists specializing in labor activities are perplexed, advising that AI should focus on changing people's work and complement current occupations, but avoid making concrete predictions.
Recently, a group of researchers tried to classify jobs according to the risk of AI changing them. According to them, the occupations that are in trouble are: telemarketing, humanities professors and employees who are responsible for authorizing credits. The most difficult trades to replace would be dancers, quarry workers and steel workers.
[What is NGL? This is the anonymous messaging app that's sweeping Instagram]
Who is going to make money with this technology?
The predictions are very varied, from achieving a more equitable society to the imposition of a less egalitarian one. Much will depend on how politicians and voters react, and the Biden Administration and Congress are paying increasing attention to AI research and development.
But some of the leaders in the development of this technology are large companies such as Google, Meta and Amazon, plus there is OpenAI, which in 2019 became a for-profit company and who survives among the dozens of AI startups that are raising billions of dollars from investors.
These are the new abilities of the humanoid robot designed by Boston Dynamics
Jan. 21, 202300:29
What could go wrong?
If we are guided by science fiction movies, or the nightmares of some researchers, there is a possibility that there are killer robots because AI could generate sentient beings with motivations of their own.
Driven by that scenario, thousands of people, including Elon Musk and some researchers, signed a petition calling for a pause of at least six months in the creation of new AI systems.
However, some senior executives and technology researchers did not sign it. And, so far, there's no overwhelming data to suggest humans are in immediate danger because of AI.
[Facebook to pay $400 to nearly two million people in Illinois for taking their personal data online without permission]
To what extent should we worry?
It depends on who you ask. In the short term, most of the immediate risks are related to abuse by humans, not robots. The use of AI to crack passwords is being investigated, and The Washington Post discovered someone using an AI-generated photo as a trap to get money.
One thing to keep an eye on is how quickly physical robots progress. Hardware hasn't advanced as much as software, and two years ago, OpenAI disbanded its robotics team even after it got a robot to solve a Rubik's Cube. However, now OpenAI is investing in a Norwegian robotics company.
/cloudfront-eu-central-1.images.arcpublishing.com/prisa/OHXLFBWROJHYTONFRSHBKYUOIU.jpg)
/cloudfront-eu-central-1.images.arcpublishing.com/prisa/GXCNZ4EZL7W46CE4OBDBCILZVE.jpg)
