Fei-Fei Li, a pioneering AI scientist, called this an inflection point. 2023 will ideally be remembered in history for the enormous changes in technology as well as the public enlightenment. It also demonstrates how clumsy this technology is.
"It took a year for people to figure out what this is, how to use it, what the impact is, all the good, bad, and ugly," she said.
ANXIETY ABOUT AI The first AI panic of 2023 began soon after New Year's Day, when schools from Seattle to Paris began censoring ChatGPT. Teenagers were already requesting essays and take-home tests from the chatbot, which was introduced in late 2022.
AI big language models, such as ChatGPT, work by guessing the next word in a sentence repeatedly after learning the patterns of a massive trove of human-written works. They frequently get facts wrong. However, the results appeared so natural that it prompted interest in the next AI breakthroughs and their potential use for deception and cunning.
Concerns grew as a new generation of generative AI tools churning forth not only words but also novel art, music, and synthetic voices threatened the livelihoods of everyone who writes, draws, strums, or codes for a career.
It fueled Hollywood writers' and actors' strikes, as well as legal challenges from visual artists and best-selling authors.
Some of the AI field's most eminent scientists cautioned that the technology's unfettered advancement was advancing toward outsmarting humans and potentially jeopardizing their existence, while others dismissed their fears as exaggerated or focused on more immediate threats.
AI-generated deepfakes, some more convincing than others, have sprung into U.S. presidential campaigns by spring, with one falsely depicting Donald Trump embracing the country's former top infectious disease specialist. The technology makes distinguishing between actual and manufactured conflict images in Ukraine and Gaza exceedingly difficult.
By the end of the year, the focus of the AI crisis had shifted to ChatGPT's creator, the San Francisco startup OpenAI, which had been nearly destroyed by corporate turmoil over its charismatic CEO, and to a government meeting room in Belgium, where exhausted political leaders from across the European Union emerged after days of intense talks with a deal for the world's first major AI legal safeguards.
The new AI law will not go into effect until 2025, and other legislative bodies, including the United States Congress, are still a long way from enacting their own.
IS THERE TOO MUCH HYPE? There is no doubt that commercial AI devices introduced in 2023 featured technology advances not conceivable in earlier stages of AI research dating back to the mid-20th century.
However, the newest generative AI movement has reached peak excitement, according to Gartner, a market research group that has been tracking what it calls the hype cycle of developing technologies since the 1990s. Consider a wooden rollercoaster approaching its highest point, about to plummet into what Gartner refers to as a "trough of disillusionment" before coasting back to reality.
According to Gartner researcher Dave Micko, generative AI is at the pinnacle of exaggerated expectations. Vendors and creators of generative AI make numerous promises about its capabilities and capacity to deliver such capabilities.
This month, Google was chastised for manipulating a video demonstration of their most powerful AI model, Gemini, to make it appear more impressive and human-like.
According to Micko, leading AI developers are pushing specific applications of the latest technology, the majority of which correspond to their current product lines, whether they be search engines or workplace productivity software. That doesn't mean the rest of the world will utilize it that way.
As much as Google, Microsoft, Amazon, and Apple would like us to adopt their method of thinking about and delivering technology, he believes that adoption occurs from the bottom up.
IS THIS TIME DIFFERENT? It's easy to forget that this isn't the first wave of artificial intelligence commercialization. Li and other scientists used computer vision techniques to search through a massive library of photographs in order to detect things and individual faces.
When we introduced Siri in 2011, it was the fastest-growing consumer app at the time and the only big mainstream application of AI that consumers had ever seen, according to Tom Gruber, co-founder of Siri Inc., which Apple purchased and integrated into the iPhone.
But, according to Gruber, what is happening now is the biggest wave in AI history, unleashing new opportunities as well as risks.
"We were surprised to discover this astonishing ability with language by training a machine to play solitaire on the entire internet," Gruber added. "It's quite amazing."
The hazards may arrive quickly in 2024, when key national elections in the United States, India, and elsewhere may be overwhelmed with AI-generated deepfakes.
Longer ahead, AI technology's rapidly developing language, visual perception, and step-by-step planning capabilities might accelerate a digital assistant's vision, but only if given access to the inner loop of our digital life stream, according to Gruber.
They have the ability to manipulate your attention, as in, You should watch this video. This is a book you should read. 'You should respond to this individual's communication,' Gruber said. That is the job of a true executive assistant. And we could have it, but at a high cost in terms of personal information and privacy.
You must be logged in to post a comment.