In videos, the fresh version of ChatGPT was capable of rapid- fireplace, natural tone conversations, picked up on personal cues, and displayed simulated personal reactions of its own.
During a video from the bank’s headquarters in San Francisco on Monday, Mira Murati, OpenAI’s chief technology officer, announced that ChatGPT may be powered by a novel, more powerful AI type called GPT- 4o. Both the existing mobile software and website version of ChatGPT may have access to the model, which will be accessible to both free and paid ChatGPT users.
Murati said the GPT- 4o design allows ChatGPT to respond more quickly to words, image, and video input than OpenAI’s past technology. She and other OpenAI workers had quick-paced conversations with ChatGPT in demos, which responded with a lively, expressive feminine voice and rapidly followed when interrupted.
ChatGPT sporadically spoke in different emotional accents during the exchange, and occasionally even seemed to be feeling something. When an OpenAI individual said he had been talking about how “useful and incredible” the robot is, it responded teasingly, gushing” Oh prevent it, you’re making me blush”.
” This really feels so beautiful, and that’s wonderful”, Murati said, adding, “over the next few months we’ll be rolling out these skills to all”.
Barret Zoph, a researcher for OpenAI, responded to ChatGPT’s question,” How can I brighten your day today?” at another point in the demo. When Zoph asked the chatbot to look at a selfie of him and describe his feelings, ChatGPT responded with” I’ll put on my emotional detective hat” and warmly stated,” It seems like you’re feeling pretty happy and cheerful… whatever’s going on, it looks like you’re in a great mood.”
In a blog post Monday, OpenAI’s CEO, Sam Altman, highlighted the significance of the new interface. ” It feels like AI from the movies, and it’s still a bit surprising to me that it’s real”, Altman wrote. ” Getting to human-level response times and expressiveness turns out to be a significant change.”
OpenAI’s tweaks may make ChatGPT more engaging and popular, but they may also change the way users think about the program.
ChatGPT gained notoriety last year as a result of its remarkable ability to generate text that frequently resembles human-like comprehension and respond to questions. However, Google and other major players like OpenAI have historically attempted to limit the anthropomorphism of their chatbots by requiring users to remember that they are just computer programs and refusing to answer specific questions.
Despite that caution, the occasionally lifelike banter of the newest chatbots can understand users ‘ emotions. Some businesses have made the technology more effective by providing AI girlfriends and boyfriends as well as role-playing various human emotions. Some AI researchers have warned that the long-term effects of developing programs that mimic and regulate emotions might be unwelcome.
A research paper from a Google DeepMind team last month examined the ethical concerns that could arise as AI assistants gain more experience. The researchers warn against using such assistants as a form of extreme persuasion and addiction, among other things. A British man admitted to breaking the law last year by breaking the security at Windsor Castle, claiming that a chatbot he had communicated with over 5, 000 times had prompted him to do so.
In order to provide a more personalized experience, OpenAI claims its GPT- 4o model improves its ability to make sense of images, including photos and charts, and can store more user data in its “memory” to provide a more personalized experience.
Users of the free version of ChatGPT will now be able to access GPT- 4o, according to OpenAI, essentially bringing all users to the most advanced AI model available through the free version.
At the search company’s I/O developer conference on Tuesday, Google is anticipated to introduce its own AI technologies. It seems unlikely that the AI industry’s competition will slow down. At the end of today’s OpenAI event, Murati promised that the company would soon be offering more announcements, “on our progress towards the next big thing”.