GPT-4 is the latest milestone in OpenAI's effort to scale up deep learning

GPT-4 is a large multimodal model that, while less capable than humans in many real-world scenarios, exhibits human-level performance on various professional and academic benchmarks.

For example, it passes a simulated bar exam with a score around the top 10% of test takers; in contrast, GPT 3.5âs score was around the bottom 10%.

We spent 6 months iteratively aligning Gpt-4 using lessons from our adversarial testing program as well as ChatGPT.

GPT-4 can accept a prompt of text and images, which lets the user specify any vision or language task.

It can be augmented with test-time techniques that were developed for text-only language models, including few-shot and chain-of-thought prompting.

It still is not fully reliable (it âhallucinates facts and makes reasoning errors).

We aim to make AI systems we build have reasonable default behaviors that reflect a wide swathe of users' values.

GPT-4 generally lacks knowledge of events that have occurred after the vast majority of its data cuts off (September 2021) The model can also be confidently wrong in its predictions, not taking care to double-check work when itâs likely to make a mistake.