For decades, artificial intelligence has been fodder for sci-fi movies, philosophers, and sleep-deprived computer programmers, but suddenly it seems to be everywhere.

ChatGPT reached 100 million users at an unprecedented rate. bill gates recently declared that “the era of AI has begun”. And the Biden administration last month began explorer new measures to hold artificial intelligence systems accountable for their impact.

But for many people, it remains a blurry concept that doesn’t affect their daily lives.

So this might be a good time to take a step back and review the basics. Here’s a guide to help you understand what all the fuss is about.

Why is everyone suddenly talking about AI?

You can thank (or blame) one specific company: OpenAI, a San Francisco-based tech startup with a few hundred employees. In November, OpenAI released the ChatGPT chatbot to the public, and it quickly became clear that it was leaps and bounds from previous chatbots. It was like talking to someone who knew everything.

The tool, which the company says is just one step in a long AI development process, it quickly went viral. Other tech companies, such as Google and Meta, had been testing similar chatbots behind closed doors, but OpenAI made it widely available, a decision that was controversial due to unknown risks.

What’s good about a chatbot?

Mediocre chatbots have been around for a long time. Think of the customer service chat windows that appear on some websites. In 2016, Microsoft even released an AI chatbot called Tay, but quickly canceled it after people taught him to use racist language.

ChatGPT came on the scene as something different. Not only could he answer a seemingly unlimited number of questions, but he could also write scripts, summarize large amounts of information, and imitate a human in conversation somewhat convincingly. It immediately seemed, at the very least, that one day it might make everyday life more efficient.

And chatbots are just one part of the AI, along with images, animated videos, facial recognition technology, and more.

Let’s go back. What is AI?

In its simplest form, AI can be summed up in a few words: thinking machines. Or, better yet, machines that can mock thought.

The term has its origin among scientists after World War II. ALAN TURING British mathematician in 1950 anything but predicted the development of «digital computers» that could persuasively mimic 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 newer term, refers to software like ChatGPT that gives rise to new material. You can find a more extensive glossary of AI terms here.

Is it really possible for computers to «think»?

We could write an entire book on this, but here’s a short answer: no, they can’t. While some people believe that AI is already coming to life, they are a small group and the idea is really a distraction from what goes on inside computers.

If you want a longer answer, NBC News spoke to a number of philosophers about how they approach the question.

So what’s really going on inside computers?

AI software can imitate humans so convincingly because it’s good at prediction: Guess the word, sentence, or picture you want to see next. (Some detractors have called this «glorified autocomplete.”)

And the systems are so good at predicting because their human creators have given them many past human-created examples, including large parts of the internet. The raw material used in AI models is called training data, and while some companies keep secrets about what they use, well-known data sources include Reddit and Wikipedia.

Well, AI extracts information from a lot of data. As?

AI learns by example. By looking at us, language models identify patterns in the way we write and speak, distilling concepts like tone, word placement, and even idioms. Those patterns are then translated into mathematics in a process called «model training.» Like children learning new words and grammar, the AI ​​must understand the rules of engagement.

When large language models like ChatGPT receive prompts, that knowledge allows them to understand what we’re asking and construct answers.

ChatGPT takes training a step further with its secret ingredient: Reinforcement Learning from Human Feedback, or RLHF. This fine tuning technique does the heavy lifting. At this stage, human raters rate the model output, heavily penalizing responses that are wild, inappropriate, or downright nonsense, while rewarding those that are informative and human-like. That allows fluid conversational exchanges.

While other fine tuning techniques exist, RLHF has been considered innovative in language modeling and is used by companies such as open AI either hug facea startup that offers tools to programmers building your own AI models.

Is AI just another Silicon Valley fad?

The tech industry has been going through one fad after another lately, ever since driverless cars and the metaverse to NFTs and web3.

On one level, AI chatbots may bear some resemblance to those underwhelming ideas: do we all really want to spend all day talking to a computer? — but there are reasons to think that AI is more than just another passing trend.

For one thing, money is pouring into the sector, with $1.7 billion in seed investment in the first three months of 2023 alone, according to the research firm. tone book. In addition, tangible uses are already emerging, from popular song to help the blind.

Why is all this happening now?

Twenty-six years have passed since the triumph of IBM’s Deep Blue computer program over chess champion Garry Kasparov, a milestone in AI research and development. Since then, computer chips have gotten much faster and can handle the vast amounts of data required for modern AI, and new ways of writing software have also made the process more efficient.

Chip manufacturers like nvidia and tech companies like Google, Meta, and OpenAI have poured resources into those two areas, as well as nurturing talented computer scientists under their respective roofs.

When can I expect this to start affecting my life?

Don’t expect to wake up one morning and suddenly live in a world of AI. Instead, expect the changes to come little by little: a hit song Created with AI, a new test in the doctor’s office to detect cancer or a little better Customer service. OpenAI has licensed its technology to Morgan Stanley so its investment advisers can provide better advice and to Khan Academy so its students have access to a chatbot tutor.

Think of all the businesses or products you deal with every day, and chances are good that one is using similar technology or will be using similar technology in the near future, even if the only immediate impact is a little more efficiency.

Can we expect big changes?

It’s hard to know what to count on, but yes, there are many dreams in AI startups. If AI software can make both human labor and computers more efficient, could all that brainpower be put to major breakthroughs in other new areas?

Two areas where there seems to be a lot of optimism: drugstore shelves full of new AI-Designed Pharmaceuticals and artificial intelligence software that could enable new power plants based on cleaner fusion energy.

Will AI make many jobs irrelevant?

Predictions run the gamut, so if you’re confused, you’re not alone. OpenAI CEO Sam Altman has suggested that AI will lead to a utopia where people don’t need to work, while others warn of mass unemployment. among computer programmers.

Even economists who specialize in labor are perplexed, advising that AI it will change people’s work and complement existing work but avoid specific predictions.

Recently, a group of researchers tried to classify jobs by risk of AI altering what people do. In trouble, according to them: telemarketers, humanities professors and credit authorizers. Hardest to replace: dancers, stonemasons, and steel workers.

And who is going to make money from this?

Once again, the predictions are everywhere, from a more equal society to a less equal one. Much depends 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 early leaders are big tech companies, like Google, Meta, and Amazon; OpenAI, which in 2019 transitioned from a non-profit to a for-profit company; and who survives among the dozens of AI startups that collectively are raising billions of dollars of the first investors.

What could go wrong?

If you are based on sci-fi movies or the nightmares of some researchers, there is a possibility that killer robots – AI become sentient beings with their own motivations.

Driven by that scenario, thousands of people, including Elon Musk and some AI researchers, signed a petition calling for a pause of at least six months in the training of new AI systems. However, some top executives and technology researchers did not sign it. And at least so far, there’s no overwhelming data to suggest that humans are in any immediate danger from AI.

So how worried should you be?

depends on who you ask.

Most of the immediate risks have to do with short-term abuse by humans, not robots. There is ongoing research on the use of AI to crack people’s passwordsand The Washington Post discovered someone using an AI-generated photo as a thirst trappossibly for cash.

One thing to watch: how quickly we see progress in physical robots. Hardware hasn’t advanced as much as software, and two years ago, OpenAI dissolved his robotics team even after getting a robot to solve a Rubik’s cube. But now OpenAI is invest at a Norwegian robotics company.