We hear a lot about AI these days.
From Elon Musk to Joe Biden, everyone worries that AI will take over our jobs, spread misinformation, or, if we’re not careful, one day kill us all. I’m here. On the other hand, rather than sticking to long-term hypothetical apocalyptic scenarios, some AI experts focus on how AI is actively harming us now, and how AI’s development is being controlled. He said there needs to be a focus on the concentration of power in a small number of companies. This error-prone technology is already being used to make defamatory lies, hack bank accounts and falsely arrest criminal suspects.
But short-term and long-term concerns aside, there are key factors that make it feel like AI is suddenly taking over the world. Because it’s fun.
Over the past few weeks, I’ve been playing around with the latest AI tools and talking to people using them. It turns out that the most exciting form of AI today is not the kind of AI that people use to process spreadsheets or write emails to improve their productivity. (Though my boss loves the idea!) They are the kind that are used to entertain us.
In the last six months, AI has come an incredibly long way to help people create all kinds of media. With varying degrees of direction, AI can create photorealistic illustrations, design video games, or come up with catchy songs with Top 40 potential.
So what should we make of the fact that people are eager to use technology with obvious serious flaws and consequences?
Margaret Mitchell, chief ethics scientist at AI platform Hugging Face, said, “I think it’s *perfectly* rational that people get excited and have fun. Mitchell previously founded the Ethical AI team at Google. She was controversially fired after co-authoring a paper pointing out the risks associated with large language models powering many AI apps. A visionary early critic of the shortcomings of AI technology, she also acknowledges its potential.
Pietro Schirano is a design lead at financial services startup Brex. He was also an early adopter of GPT-4, the latest iteration of technology from the company behind the viral ChatGPT app, OpenAI.
When GPT-4 came out in March, Schirano couldn’t wait to use it. He decided to test the ability to create working lines of code from a simple prompt. So Schirano set out to recreate his Pong for video games. Because, in his words, “This is the first video game ever and it would be cool to do that.”
After typing a few sentences into GPT-4 and copying and pasting the code into the code engine, Schirano had a working Pong in less than 60 seconds. he was surprised
“Oh my God,” he said. “This is different.”
his video post tweet The process went viral.
When I asked Schirano if he was worried about AI taking over the jobs of people like him, he said he wasn’t too worried. At work, he uses ChatGPT to help him be more productive and focus on higher-level decision-making, he says.
“These tools don’t necessarily replace us in practice, but they basically make us superhuman in some ways,” Schirano said.
As my colleague Rani Molla reported, many workers are in the same camp as Schirano. They don’t expect AI to completely replace their jobs, and they aren’t particularly afraid of it right now.
We spoke with Professor Ethan Morrick of the Wharton School of Business at the University of Pennsylvania about the wider reaction to new AI tools. According to Mollick, all sorts of new “general-purpose technologies” such as electricity, steam power, computers, etc., can be very disruptive, but they are popular because of unexpected, novel and often interesting use cases. is winning. He added that new advances in AI like GPT-4 fit very well into that general-purpose technology category.
AI “perfectly enhances creativity,” says Mollick. “How can you not spend every minute trying to get this to work? It’s unbelievable. I think it can be incredibly scary.”
These fun and creative ways of using AI also raise questions of authenticity. Will AI replace human creativity, or will it simply help generate it?
Last week, a hip-hop song that looked like a mashup of artists Drake and The Weeknd went viral on social media. Posted by anonymous user ‘ghostwriter’, the song claimed to be made with AI and was played millions of times before being taken down by major platforms.
The proliferation of this kind of AI-generated media has taken record labels by surprise, and Universal Music Group has asked streaming services like Spotify to stop AI companies from using their music to train models, citing intellectual property rights. raised concerns. And last month, a coalition of recording industry unions and trade associations launched a “Human Artistry Campaign” to stop AI from “replacing or eroding” artists.
However, a few artists have embraced the concept of AI. Musician Grimes asked her fans to co-produce music in her likeness and offered to split her 50% of the royalties.
Mollick likened the debate over whether AI will replace artists to the introduction of synthesizers into contemporary music. When the synthesizer first came out, people debated whether it “ruined the music” and whether the person using the instrument was a real musician.
Ultimately, Mitchell said the real danger of AI is not in the technology, but in who controls it and how it is used.
“My problem is less with creative people exploring new technologies than with technology leaders misleading or inappropriately pushing technology.”
A version of this story was first published in the Vox Technology Newsletter. SIGN UP HERE Don’t miss the next one!
