![]() “Getting it out in the world is how we learn. “We spent a lot of time trying to understand what GPT-4 is capable of,” Brockman said. When asked about GPT-4’s robustness, Brockman stressed that the model has gone through six months of safety training and that, in internal tests, it was 82% less likely to respond to requests for content disallowed by OpenAI’s usage policy and 40% more likely to produce “factual” responses than GPT-3.5. But many had hoped, this reporter included, that GPT-4 might deliver significant improvements on the moderation front. Meta’s BlenderBot and OpenAI’s ChatGPT, too, have been prompted to say wildly offensive things, and even reveal sensitive details about their inner workings. It’s not a new phenomenon in the language model domain. Hours after the model’s release, Israeli cybersecurity startup Adversa AI published a blog post demonstrating methods to bypass OpenAI’s content filters and get GPT-4 to generate phishing emails, offensive descriptions of gay people and other highly objectionable text. At the time, OpenAI claimed that upgrades to its safety system made the face-editing feature possible by “minimizing the potential of harm” from deepfakes as well as attempts to create sexual, political and violent content.Īnother perennial is preventing GPT-4 from being used in unintended ways that might inflict harm - psychological, monetary or otherwise. After initially disabling the capability, OpenAI allowed customers to upload people’s faces to edit them using the AI-powered image-generating system. OpenAI dealt with similar ethical dilemmas around DALL-E 2, its text-to-image system. “We need to figure out, like, where the sort of danger zones are - where the red lines are - and then clarify that over time.” “There’s policy issues like facial recognition and how to treat images of people that we need to address and work through,” Brockman said. Brockman says that the wider rollout, whenever it happens, will be “slow and intentional” as OpenAI evaluates the risks and benefits. Only a single launch partner has access to GPT-4’s image analysis capabilities at the moment - an assistive app for the visually impaired called Be My Eyes. For example, fed the prompt “What’s funny about this image? Describe it panel by panel” plus a three-paneled image showing a fake VGA cable being plugged into an iPhone, GPT-4 gives a breakdown of each image panel and correctly explains the joke (“The humor in this image comes from the absurdity of plugging a large, outdated VGA connector into a small, modern smartphone charging port”). GPT-4’s image understanding abilities are quite impressive. (Training data has gotten OpenAI into legal trouble before.) OpenAI says that the training data came from “a variety of licensed, created, and publicly available data sources, which may include publicly available personal information,” but Brockman demurred when I asked for specifics. That’s because GPT-4 was trained on image and text data while its predecessors were only trained on text. ![]() an image of giraffes in the Serengeti with the prompt “How many giraffes are shown here?”). “Write an essay about giraffes”), GPT-4 can take a prompt of both images and text to perform some action (e.g. ![]() Unlike GPT-3 and GPT-3.5, which could only accept text prompts (e.g. Shifting gears, one of GPT-4’s more intriguing aspects is the above-mentioned multimodality. ( GPT-3.5, the intermediate model between GPT-3 and GPT-4, also scores a 4.) And in a simulated bar exam, GPT-4 passes with a score around the top 10% of test takers GPT-3.5’s score hovered around the bottom 10%. On the AP Calculus BC exam, GPT-4 scores a 4 out of 5 while GPT-3 scores a 1. “There’s still a lot of problems and mistakes that makes … but you can really see the jump in skill in things like calculus or law, where it went from being really bad at certain domains to actually quite good relative to humans.” “It’s just different,” he told TechCrunch. To get a better handle on GPT-4’s development cycle and its capabilities, as well as its limitations, TechCrunch spoke with Greg Brockman, one of the co-founders of OpenAI and its president, via a video call on Tuesday.Īsked to compare GPT-4 to GPT-3, Brockman had one word: Different. In one example on OpenAI’s own blog, GPT-4 describes Elvis Presley as the “son of an actor.” (Neither of his parents were actors.) Like GPT-3, the model “hallucinates” facts and makes basic reasoning errors. It’s also multimodal in the sense that it can understand images, allowing it to caption and even explain in detail the contents of a photo.īut GPT-4 has serious shortcomings. GPT-4 improves upon its predecessor, GPT-3, in key ways, for example giving more factually true statements and allowing developers to prescribe its style and behavior more easily. OpenAI shipped GPT-4 yesterday, the much-anticipated text-generating AI model, and it’s a curious piece of work.
0 Comments
Leave a Reply. |