Skip to Content
Artificial intelligence

OpenAI’s new language generator GPT-3 is shockingly good—and completely mindless

The AI is the largest language model ever created and can generate amazing human-like text on demand but won't bring us closer to true intelligence.

falling letters
Getty

“Playing with GPT-3 feels like seeing the future,” Arram Sabeti, a San Francisco–based developer and artist, tweeted last week. That pretty much sums up the response on social media in the last few days to OpenAI’s latest language-generating AI.  

OpenAI first described GPT-3 in a research paper published in May. But last week it began drip-feeding the software to selected people who requested access to a private beta. For now, OpenAI wants outside developers to help it explore what GPT-3 can do, but it plans to turn the tool into a commercial product later this year, offering businesses a paid-for subscription to the AI via the cloud.

GPT-3 is the most powerful language model ever. Its predecessor, GPT-2, released last year, was already able to spit out convincing streams of text in a range of different styles when prompted with an opening sentence. But GPT-3 is a big leap forward. The model has 175 billion parameters (the values that a neural network tries to optimize during training), compared with GPT-2’s already vast 1.5 billion. And with language models, size really does matter.

Sabeti linked to a blog post where he showed off short stories, songs, press releases, technical manuals, and more that he had used the AI to generate. GPT-3 can also produce pastiches of particular writers. Mario Klingemann, an artist who works with machine learning, shared a short story called “The importance of being on Twitter,” written in the style of Jerome K. Jerome, which starts: “It is a curious fact that the last remaining form of social life in which the people of London are still interested is Twitter. I was struck with this curious fact when I went on one of my periodical holidays to the sea-side, and found the whole place twittering like a starling-cage.” Klingemann says all he gave the AI was the title, the author’s name and the initial “It.” There is even a reasonably informative article about GPT-3 written entirely by GPT-3.

Others have found that GPT-3 can generate any kind of text, including guitar tabs or computer code. For example, by tweaking GPT-3 so that it produced HTML rather than natural language, web developer Sharif Shameem showed that he could make it create web-page layouts by giving it prompts like “a button that looks like a watermelon” or “large text in red that says WELCOME TO MY NEWSLETTER and a blue button that says Subscribe.” Even legendary coder John Carmack, who pioneered 3D computer graphics in early video games like Doom and is now consulting CTO at Oculus VR, was unnerved: “The recent, almost accidental, discovery that GPT-3 can sort of write code does generate a slight shiver.”

Yet despite its new tricks, GPT-3 is still prone to spewing hateful sexist and racist language. Fine-tuning the model helped limit this kind of output in GPT-2.

It’s also no surprise that many have been quick to start talking about intelligence. But GPT-3’s human-like output and striking versatility are the results of excellent engineering, not genuine smarts. For one thing, the AI still makes ridiculous howlers that reveal a total lack of common sense. But even its successes have a lack of depth to them, reading more like cut-and-paste jobs than original compositions.

Exactly what’s going on inside GPT-3 isn’t clear. But what it seems to be good at is synthesizing text it has found elsewhere on the internet, making it a kind of vast, eclectic scrapbook created from millions and millions of snippets of text that it then glues together in weird and wonderful ways on demand.   

That’s not to downplay OpenAI’s achievement. And a tool like this has many new uses, both good (from powering better chatbots to helping people code) and bad (from powering better misinformation bots to helping kids cheat on their homework).

But when a new AI milestone comes along it too often gets buried in hype. Even Sam Altman, who co-founded OpenAI with Elon Musk, tried to tone things down: “The GPT-3 hype is way too much. It’s impressive (thanks for the nice compliments!) but it still has serious weaknesses and sometimes makes very silly mistakes. AI is going to change the world, but GPT-3 is just a very early glimpse. We have a lot still to figure out.”

We have a low bar when it comes to spotting intelligence. If something looks smart, it’s easy to kid ourselves that it is. The greatest trick AI ever pulled was convincing the world it exists. GPT-3 is a huge leap forward—but it is still a tool made by humans, with all the flaws and limitations that implies.

Deep Dive

Artificial intelligence

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

Providing the right products at the right time with machine learning

Amid shifting customer needs, CPG enterprises look to machine learning to bolster their data strategy, says global head of MLOps and platforms at Kraft Heinz Company, Jorge Balestra.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.