OpenAI's ChatGPT-5 has been long-awaited, with speculation about its capabilities, new features, and release date steadily intensifying.
A wealth of information about ChatGPT-5 has already circulated, and this article will synthesize and predict the release date, features, and pricing of ChatGPT-5 based on market rumors.
What is ChatGPT-5?
ChatGPT-5 is the latest version of ChatGPT. ChatGPT is a large language model (LLM) that captured global attention when it was launched in November 2022, bringing artificial intelligence (AI) into the mainstream.
It has brought indescribable changes to businesses, small enterprises, and personal lives—always-on chatbots that can intelligently and quickly generate text, answer questions, and even produce code in various programming languages.
ChatGPT-5 is based on the GPT-5 AI model and is expected to improve upon the capabilities of GPT-4, acting as a bridge between human and machine communication.
As a generative pre-trained transformer, it aims to enhance interactions by providing more personalized and accurate responses and is expected to handle a broader range of content, possibly including video. This indicates that ChatGPT-5's capabilities will far exceed those of GPT-4.
There are also speculations that an OpenAI project named "Strawberry" will play a crucial role in achieving GPT-5. According to Reuters, Strawberry will be able to browse the internet autonomously and conduct "in-depth research," providing improved reasoning capabilities and the ability to independently manage complex tasks.
When will ChatGPT-5 be released?
Expected release date: Fall 2024.
Although OpenAI has not yet determined the final release date for ChatGPT-5, more and more speculation suggests it might launch in the fall of 2024. However, as of now, there is very little information about ChatGPT-5.
Initially, there were rumors that the model would be released in the summer of 2024, following updates from OpenAI executives about the continuous development and improvement of their AI models.
At various tech and AI conferences, OpenAI CEO Sam Altman hinted that the new model would bring substantial improvements, reinforcing expectations for a significant advancement in AI capabilities that could redefine user interaction with machine learning (ML) technologies.
For example, unverified reports on Reddit claim that a small number of people are already using ChatGPT-5—which sounds like fine-tuning before the full release is the name of the game.
What new features will ChatGPT-5 have?
Autonomous Agents, Multimodal, "Substantially Better"
Sam Altman, in a candid admission on the Lex Fridman podcast, acknowledged that compared to the new AI, the engine behind their advanced ChatGPT product, GPT-4, is "a bit lousy."
In the interview, Altman also revealed plans to launch an "amazing new model" this year, hinting at a version of OpenAI's ChatGPT-5 or GPT-5.
According to Business Insider, several enterprise customers had the opportunity to preview the latest enhancements to ChatGPT, along with a glimpse of the new GPT-5 model.
A CEO described the GPT-5 model as a "substantial improvement" and noted that OpenAI tailored the demonstration to include data and use cases specific to his company.
He also hinted at other unannounced features of ChatGPT-5, such as the model's ability to deploy AI agents for autonomous task execution and its capability to handle a broader range of content, possibly including video.
OpenAI stated that its ChatGPT-5 updates would "bridge the gap between human and machine communication" by providing a more seamless and intuitive conversational experience.
However, since OpenAI has not released any information about ChatGPT-5's parameters, it is difficult to judge how it compares to existing products like GPT-4, Grok, or Gemini.
Autonomous AI Agents
Common tasks without human supervision The next version of ChatGPT (GPT-5) is expected to include over 1.5 trillion parameters, surpassing its predecessor GPT-4 in scale and capability.
The increase in complexity aims to enhance the model's reasoning abilities and improve the naturalness of conversations. The goal is to make interactions feel more like conversing with a human rather than a machine.
Although we can only speculate about ChatGPT-5's new features, the development of GPT-5 also hints at the introduction of AI agents capable of autonomous operation. These agents are expected to perform real-world tasks, such as scheduling appointments, managing emails, or online shopping—all without human supervision.
Such features suggest that the system will be able to manage daily activities, potentially making ChatGPT-5 a personal assistant.
These advancements are not just about handling daily tasks; they aim to enhance the role of AI in more complex decision-making processes by providing more personalized and context-aware services, potentially transforming areas such as customer service, healthcare, and education.
Multimodal
Will ChatGPT-5 support new content types? GPT-5, along with ChatGPT-5, could also be multimodal. This would enable it to handle and respond to various inputs beyond text, including images, videos, and possibly other data types.
While the advanced model of ChatGPT-4 has encountered some bumps in achieving image results through Dall-E, further expansion into multimodality would allow GPT-5 to perform more complex tasks and make more integrated responses across different media types. It is currently unclear whether Sora's video skills will be incorporated into ChatGPT-5.
Voice Features
Advancements in voice generation technology herald a conversational future On March 29, 2024, OpenAI announced its progress in AI voice generation—being able to use a 15-second voice clip to generate a voice model that sounds like the original speaker.
OpenAI stated:
"Today, we are sharing preliminary insights and results from a small-scale preview of a model called Voice Engine, which uses text input and a single 15-second audio sample to generate natural speech very similar to the original speaker."
"Notably, a small model with only a 15-second sample can create emotionally rich and realistic voices."
"At the same time, due to the potential for abuse of synthetic voices, we are cautious and informed about broader releases."
"We hope to engage in a dialogue about the responsible deployment of synthetic voices and how society can adapt to these new features."
"Based on these dialogues and the results of small-scale testing, we will make more informed decisions about whether and how to deploy this technology on a large scale."
Training Data Will Increase
Closer to general artificial intelligence? Improvements in GPT-5 are expected to address some limitations of previous models, such as the reliability of responses and the model's ability to handle diverse and dynamic data sets.
OpenAI plans to enrich GPT-5 with broader training data, including proprietary data sets that can cover more specialized knowledge domains and complex scenarios.
This approach can enhance the accuracy and practicality of AI in professional environments where specialized knowledge is crucial.
As GPT-5 becomes more reliable and powerful, it is expected to contribute to the slow and steady process of achieving general artificial intelligence (AGI). Although, as mentioned in the tweet above, ChatGPT-5 is not expected to achieve this significant leap.
Even without AGI, ChatGPT-5 should bring us closer to AI systems that can understand and perform a wide range of human activities without specific programming for each task.
What is the price of ChatGPT-5?
Estimated price: $20 per month
If OpenAI follows this strategy, the expected price range for ChatGPT-5 will be from a free usage model to a $20 monthly subscription model with extended features. Therefore, the cost of ChatGPT-5 may be the same as the ChatGPT Plus subscription fee.
What are the differences between ChatGPT-4 and ChatGPT-5?
Before paying the $20 monthly subscription fee, many will immediately try to determine who will win in the battle between ChatGPT-5 and ChatGPT-4.
The goal of GPT-5 is to surpass GPT-4 and its Turbo variant by significantly addressing and expanding its capabilities.
GPT-4 Turbo extended the model's knowledge timeline to April 2023 and greatly increased the length of prompts it can handle. GPT-5 is expected to further push these boundaries.
Its advancements may include deeper integration with tools like Dall-E3 and enhanced search capabilities, providing a more seamless and intuitive user experience.
Additionally, GPT-5 is expected to exhibit superior performance in understanding and executing complex tasks, demonstrating significantly stronger capabilities than GPT-4 in academic and professional testing environments.
It is speculated that the release of ChatGPT-5 will prompt OpenAI to release the current advanced features of ChatGPT-4 to the public for free. Meanwhile, the Pro model of ChatGPT-5 may allow users to go further than the free version of ChatGPT-5.
ChatGPT-5 will address the flaws of ChatGPT-4
It is worth emphasizing that the Reddit user community has sometimes been frustrated with the performance issues of GPT-4.
Many users expressed great dissatisfaction, pointing out that the AI has become noticeably slower and even needs help understanding basic commands.
Complaints also included broken code outputs, sudden restarts during interactions, and inexplicable switching between languages, all of which led to a decrease rather than an increase in productivity.
Reddit users called on OpenAI to pay more attention to the stability and reliability of the model rather than introducing new features to impress content creators.
In these discussions, there is growing concern that future AI models may become exclusive to high-paying customers, while the basic or free versions will be inefficient and filled with poor-quality content.
Therefore, users have high expectations for ChatGPT-5, so OpenAI needs to make significant improvements to maintain its market-leading position.