pixabay cc0 public domain.
OpenAI's artificial intelligence (AI) chatbot, ChatGPT, was released to the unsuspecting public a year ago.
It quickly became the fastest-growing app ever, mastering 100 million users by the end of the second month. Today, more than a billion people can use it through Microsoft's Bing search, Skype, and Snapchat — and OpenAI expects to earn more than $1 billion in annual revenue.
We've never seen technology roll out so quickly before. It took about a decade or so for most people to start using the web. But this time, the pipeline is in place.
As a result, ChatGPT's impact extends far beyond writing poetry about Carol's retirement in the Shakespearean style. It has given many people a glimpse of our AI-driven future. Here are five ways this technology is changing the world.
ChatGPT is forcing countries around the world to be sensible to recognize that AI poses significant challenges – not just economic challenges, but also social and existential challenges.
Joe Biden has put the U.S. at the forefront of AI regulation with an executive order that sets a new standard for AI safety and security. It wants to improve fairness and civil rights while promoting innovation and competition, as well as U.S. leadership in artificial intelligence.
Soon after, the United Kingdom held its first Bletchley Park at Bletchley Park, where computers were born in World War II to crack Germany's Enigma code.
Recently, the EU appears to be sacrificing its early leadership in regulating AI as it struggles to adapt its AI Act to address the potential threat posed by cutting-edge models such as ChatGPT.
While Australia continues to lag behind in regulation and investment, countries around the world are increasingly devoting money, time and attention to tackling this issue that most people didn't think about five years ago.
Before the advent of ChatGPT, it was probably auto workers and other blue-collar workers who were most afraid of the arrival of bots. ChatGPT and other generative AI tools have changed this conversation.
White-collar workers such as graphic designers and lawyers are now also starting to worry about their jobs. A recent study of the job market found that earnings from writing and editing jobs have dropped by more than 10% since the launch of ChatGPT. The gig economy may be the canary in this coal mine.
There is a huge uncertainty as to whether AI will destroy more jobs than it creates. But one thing is certain now: AI will dramatically disrupt the way we work.
The education sector was somewhat hostile to the arrival of ChatGPT, and many schools and education departments immediately banned its use. What would happen to homework if chatgpt could write?
Of course, we don't ask people to write because of shortages, or even because many jobs require it. We asked them to write** because it requires research skills, improved communication skills, critical thinking, and domain knowledge. Whatever ChatGPT offers, these skills are still needed even if we spend less time developing them.
And it's not just schoolchildren who cheat with AI. Earlier this year, a U.S. judge fined two lawyers and a law firm $5,000 for a court document written by ChatGPT that contained fabricated legal citations.
I think these are growing pains. Education is an area where AI has a lot to offer. For example, large language models such as ChatGPT can be fine-tuned to make excellent Socratic tutors. The intelligent tutoring system allows for infinite patience when generating precisely located Xi questions.
This one is personal. Authors around the world are furious to discover that many large language models, like ChatGPT, have taken hundreds of thousands of books from the web without their consent.
AI models are able to speak fluently about everything from AI to zoology because they have been trained in all books from AI to zoology. Books on artificial intelligence include my own copyrighted book on artificial intelligence.
Ironically, an AI professor's book on AI was used to train AI, which caused controversy. Currently, multiple class action lawsuits are being filed in the United States to determine whether this violates copyright law.
Users of ChatGPT have even pointed to examples of chatbots generating entire blocks of text verbatim from copyrighted books.
In the short term, one of the challenges I'm most worried about is using generative AI tools like ChatGPT to create misinformation and disinformation.
This concern goes beyond synthetic text, deepfaked audio and **indistinguishable from real audio and**. A bank has been robbed by an AI-generated clone voice.
The elections now also appear to be under threat. Deepfakes played an unfortunate role in the Slovak parliamentary election campaign in 2023. Two days before the election, a fake audio clip about election fraud allegedly featured a well-known journalist from an independent news platform and chairman of the Slovak Progressive Party, attracting thousands of social media users. Commentators believe that such false content could have a significant impact on the outcome of the election.
According to The Economist, more than 4 billion people will be asked to vote in various elections next year. What happens in an election like this when we combine the power and persuasion of social ** with the power and persuasiveness of AI-generated fake content?Will it bring a wave of misinformation and disinformation to our already fragile democracies?
It's hard to ** what will happen next year. But if there's anything to go in 2023, I suggest we fasten our seatbelts.