We got a first look at the much-anticipated big new language model from OpenAI. But this time how it works is even more deeply under wraps.
More powerful and unpredicatable than before
OpenAI has finally unveiled GPT-4, a next-generation large language model that was rumored to be in development for much of last year. The San Francisco-based company’s last surprise hit, ChatGPT, was always going to be a hard act to follow, but OpenAI has made GPT-4 even bigger and better.
Too fast and too furious
Never before has powerful new AI gone from lab to consumer-facing products so fast.
And yet large language models remain fundamentally flawed. GPT-4 can still generate biased, false, and hateful text; it can also still be hacked to bypass its guardrails. Though OpenAI has improved this technology, it has not fixed it by a long shot. The company claims that its safety testing has been sufficient for GPT-4 to be used in third-party apps.
But it is also braced for surprises.
GPT v4 outperforms in many ways
OpenAI’s new model appears to be better at some basic reasoning than ChatGPT, solving simple puzzles such as summarizing blocks of text in words that start with the same letter.
GPT-4 took in a document about taxes and answered questions about it, citing reasons for its responses.
According to OpenAI, GPT-4 performs better than ChatGPT—which is based on GPT-3.5, a version of the firm’s previous technology—because it is a larger model with more parameters (the values in a neural network that get tweaked during training). This follows an important trend that the company discovered with its previous models. GPT-3 outperformed GPT-2 because it was more than 100 times larger, with 175 billion parameters compared to GPT-2’s 1.5 billion.
Other AI models
GPT-4 may be the best multimodal large language model yet built. But it is not in a league of its own, as GPT-3 was when it first appeared in 2020. A lot has happened in the last three years. Today GPT-4 sits alongside other multimodal models, including Flamingo from DeepMind. And Hugging Face is working on an open-source multimodal model that will be free for others to use and adapt, says Wolf.
