5 new GPT-4o features making ChatGPT better than ever

Here’s what GPT-5 could mean for the future of AI PCs

The model can pick up on the tone in your voice, and will try to respond in an appropriate tone of its own. In some circumstances you can even ask it to add more or less drama to its response, or use a different voice — like a robotic one for a story being told by a robot, or singing for the end of a fairytale. It’ll be interesting to see how the new GPT-5 model performs with copyrighted material restrictions.

Its advanced reasoning will allow it to suggest treatment options based on patient symptoms and medical history. GPT-5 could revolutionize healthcare by addressing challenges that require speed, precision, and adaptability. AI-powered systems are already assisting doctors, researchers, and patients, but GPT-5’s improved features will take this to a new level.

ChatGPT-maker is set to launch GPT-5 with new and enhanced capabilities compared to its predecessor during the summer. OpenAI is committed to addressing the limitations of previous models,such as hallucinations and inconsistencies. ChatGPT-5 will undergo rigorous testing to ensure it meets the highest standards of quality. This groundbreaking collaboration has changed the game for OpenAI by creating a way for privacy-minded users to access ChatGPT without sharing their data.

Meta’s Llama upgrade

Level 3 is when the AI models begin to develop the ability to create content or perform actions without human input, or at least at the general direction of humans. Sam Altman, OpenAI CEO has previously hinted that GPT-5 might be an agent-based AI system. In the same way that GPT-3.5 was at the start of level 1, the start of level 2 could be achieved this year with the mid-tier models. OpenAI is expected to release GPT-4.5 (or something along those lines) by the end of the year and with it improvements in reasoning. Many of the frontier models have human-level problem-solving on specific tasks, but none have achieved that on a general, broad level without very specific prompting and data input.

According to Bloomberg’s unnamed sources, OpenAI has 5 steps to reach AGI and we’re only just moving towards step two — the creation of “reasoners”. These are models capable of performing problem-solving tasks as well as a human with a PhD and no access to a textbook. One of the most anticipated features in GPT-4 is visual input, which allows ChatGPT Plus to interact with images not just text, making the model truly multimodal.

OpenAI CTO Mira Murati opens the event with a discussion of making a product that is more easy to use “wherever you are”. Also launching a new model called GPT-4o that brings GPT-4-level intelligence to all users including those on the free version of ChatGPT. There is no specific timeframe when safety testing needs to be completed, one of the people familiar noted, so that process could delay any release date. It’s not clear when we’ll see GPT-4o migrate outside of ChatGPT, for example to Microsoft Copilot.

GPT-5 could be released sooner than you think

Altman says they have a number of exciting models and products to release this year including Sora, possibly the AI voice product Voice Engine and some form of next-gen AI language model. Essentially we’re starting to get to a point — as Meta’s chief AI scientist Yann LeCun predicts —where our entire digital lives go through an AI filter. Agents and multimodality in GPT-5 mean these AI models can perform tasks on our behalf, and robots put AI in the real world.

Before ChatGPT’s popularity skyrocketed, I was already testing the chatbot and other models. As a result, in the past two years, I have developed a sense of what makes a model great, including speed, reliability, accessibility, cost, features, and more. Since Copilot launched in February 2023, it has been at the top of my list — until now. OpenAI is launching GPT-4o, an iteration of the GPT-4 model that powers its hallmark product, ChatGPT.

Better Advanced Voice Mode chats

Rivals like Google’s Gemini and other emerging models such as Perplexity are rapidly gaining attention. Each company is racing to develop smarter, more versatile, and reliable AI systems. The scientific community relies heavily on data analysis and hypothesis generation, both of which could be transformed by GPT-5.

• In this scenario, you—the web developer—are the human agent responsible for coordinating and prompting the AI models one task at a time until you complete an entire set of related tasks.
• The other primary limitation is that the GPT-4 model was trained on internet data up until December 2023 (GPT-4o and 4o mini cut off at October of that year).
• GPT-3, the third iteration of OpenAI’s groundbreaking language model, was officially released in June 2020.As one of the most advanced AI language models, it garnered significant attention from the tech world.
• Beyond its text-based capabilities, it will likely be able to process and generate images, audio, and potentially even video.
• It can understand and respond to more inputs, it has more safeguards in place, provides more concise answers, and is 60% less expensive to operate.

The forthcoming months are expected to reveal the full capabilities of this advanced model. GPT-5 hype surges with impending release date and feature descriptions. ChatGPT-5 is expected to offer enhanced natural language processing capabilities. It may demonstrate improved understanding of context and nuance in conversations. The AI could potentially handle more complex tasks and provide more accurate responses.

What Has OpenAI Said?

For example, reportshave suggested that GPT-3.5 was trained on 175 billion parameters, while GPT-4 was trained on 1 trillion. OpenAI has just released the text-to-video generator, but some users have already requested improvements. That’s coming next year, and I’ll remind you that we saw a Sora 2 demo leak around when Sora came out.

In contrast, GPT-4 has a relatively smaller context window of 128,000 tokens, with approximately 32,000 tokens or fewer realistically available for use on interfaces like ChatGPT. So, for GPT-5, we expect to be able to play around with videos—upload videos as prompts, create videos on the go, edit videos with text prompts, extract segments from videos, and find specific scenes from large video files. But given how fast AI development is, it’s a very reasonable expectation. I analysed my usage of LLMs, which spans Claude, GPT-4, Perplexity, You.com, Elicit, a bunch of summarisation tools, mobile apps and access to the Gemini, ChatGPT and Claude APIs via various services. Excluding API access, yesterday I launched 23 instances of various AI tools, covering more than 80,000 words.

Users and developers are curious about its capabilities, release timeline, and potential impacts across various industries. Improvements in natural language processing may allow ChatGPT 5 to better understand nuanced queries. This could result in more human-like conversations and more precise answers to user questions. ChatGPT 5 is set to bring significant improvements in language understanding and generation. The upcoming model promises enhanced capabilities across multiple domains. If you use ChatGPT now, you can expect a smoother, more powerful experience with ChatGPT 5.

Chat GPT-5 is very likely going to be multimodal, meaning it can take input from more than just text but to what extent is unclear. Google’s Gemini 1.5 models can understand text, image, video, speech, code, spatial information and even music. ChatGPT 5 is set to revolutionize AI interactions across various sectors. Its advanced capabilities will reshape how businesses and individuals engage with AI technology.

This includes detecting sarcasm and handling complex conversational context. With the new voice mode, ChatGPT can use the context from your environment to provide voice answers, as seen in the demo below in which the chatbot comments on the user’s emotions just by looking at his face. As much as GPT-4 impressed people when it first launched, some users have noticed a degradation in its answers over the following months.

While it’s good news that the model is also rolling out to free ChatGPT users, it’s not the big upgrade we’ve been waiting for. The technology behind these systems is known as a large language model (LLM). These are artificial neural networks, a type of AI designed to mimic the human brain. They can generate general purpose text, for chatbots, and perform language processing tasks such as classifying concepts, analysing data and translating text.

According to one of the unnamed executives who have tested GPT-5, the new language model is “really good” and “materially better” than the current versions of ChatGPT. OpenAI should release it this summer, after it completes the final round of internal testing. That’s according to unnamed execs who have been able to try the new model, and who anonymously detailed some of its improvements. Generating images with legible text has long been a weak point of AI, but GPT-4o appears more capable in this regard. Text can not only be legible, but arranged in creative ways, such as typewriter pages, a movie poster, or using poetic typography.

As for OpenAI, the company is not ready to make any GPT-5 announcements. It was initially expected to drop in 2024, but OpenAI encountered unexpected delays while burning through cash. Training GPT-5 might cost up to $500 million per run, and the results aren’t exciting.

This website is using a security service to protect itself from online attacks. There are several actions that could trigger this block including submitting a certain word or phrase, a SQL command or malformed data. Chris Smith has been covering consumer electronics ever since the iPhone revolutionized the industry in 2007. When he’s not writing about the most recent tech news for BGR, he closely follows the events in Marvel’s Cinematic Universe and other blockbuster franchises. Altman teased “lots of Sora improvements” when another X user asked for specific Sora upgrades. Regarding GPT-4o, someone asked for image generation support, with Altman saying he hoped it’s coming.

🔥 What to expect when you’re expecting GPT-5

Jak joined the TweakTown team in 2017 and has since reviewed 100s of new tech products and kept us informed daily on the latest science, space, and artificial intelligence news. Jak’s love for science, space, and technology, and, more specifically, PC gaming, began at 10 years old. It was the day his dad showed him how to play Age of Empires on an old Compaq PC. Ever since that day, Jak fell in love with games and the progression of the technology industry in all its forms. The sources further disclosed that OpenAI’s GPT-5 model is still under development and is in the training phase. Beyond this point, the ChatGPT-maker will run internal tests on the model.

ChatGPT: Everything you need to know about the AI-powered chatbot – TechCrunch

ChatGPT: Everything you need to know about the AI-powered chatbot.

Posted: Tue, 14 Jan 2025 08:00:00 GMT [source]

At a cost of $200 per month, the Pro tier costs 10 times as much as a standard, single-user Plus account. The other primary limitation is that the GPT-4 model was trained on internet data up until December 2023 (GPT-4o and 4o mini cut off at October of that year). However, since GPT-4 is capable of conducting web searches and not simply relying on its pretrained data set, it can easily search for and track down more recent facts from the internet.

This included the transcript of a four-hour podcast, which I wanted to query, and a bunch of business and research questions. And once you access a GPT-5-class model, you can use dozens or more of those PhD-level software assistants. We should also expect to see these models—while still unreliable—become substantially more reliable than previous versions.

Yes, OpenAI and its CEO have confirmed that GPT-5 is in active development. The steady march of AI innovation means that OpenAI hasn’t stopped with GPT-4. That’s especially true now that Google has announced its Gemini language model, the larger variants of which can match GPT-4. In response, OpenAI released a revised GPT-4o model that offers multimodal capabilities and an impressive voice conversation mode.

GPT-5 promises better accuracy and multimodality

So I reckon a ‘generation’ is more likely to look like Schmidt’s four years. OpenAI typically releases its newest models behind a paywall, reserving free access for older versions or limited features. Its advancements in reasoning, context retention, and multimodal features are expected to unlock new possibilities across key sectors.

Although it turns out that nothing was launched on the day itself, it now feels plausible that we’ll get something big announced from the company soon. So, ChatGPT-5 may include more safety and privacy features than previous models. For instance, OpenAI will probably improve the guardrails that prevent people from misusing ChatGPT to create things like inappropriate or potentially dangerous content. Before we see GPT-5 I think OpenAI will release an intermediate version such as GPT-4.5 with more up to date training data, a larger context window and improved performance. GPT-3.5 was a significant step up from the base GPT-3 model and kickstarted ChatGPT. This timeline allows OpenAI to focus on refining and enhancing the capabilities of their AI system.

ChatGPT o3 is coming in January, but there’s still no word on GPT-5 – BGR

ChatGPT o3 is coming in January, but there’s still no word on GPT-5.

Posted: Mon, 23 Dec 2024 08:00:00 GMT [source]

There have been many potential explanations for these occurrences, including GPT-4 becoming smarter and more efficient as it is better trained, and OpenAI working on limited GPU resources. Some have also speculated that OpenAI had been training new, unreleased LLMs alongside the current LLMs, which overwhelmed its systems. At the time, in mid-2023, OpenAI announced that it had no intentions of training a successor to GPT-4. However, that changed by the end of 2023 following a long-drawn battle between CEO Sam Altman and the board over differences in opinion. Altman reportedly pushed for aggressive language model development, while the board had reservations about AI safety. The former eventually prevailed and the majority of the board opted to step down.

The company unveiled the next-gen reasoning model that will power ChatGPT, which is called o3. GPT-5 will have better language comprehension, more accurate responses, and improved handling of complex queries compared to GPT-4. Yes, ChatGPT 5 is expected to be released, continuing the advancements in AI conversational models.

The list of suggested ChatGPT features for 2025 includes other tidbits. However, the “hardware play” suggestion is enough to warrant the response above. Even if the ChatGPT hardware will be unveiled next year, and there’s no indication it will be, Altman would probably avoid addressing it this early.

It should also help support the concept known as industry 5.0, where humans and machines operate interactively within the same workplace. Similar reservations apply to other high-consequence fields, such as aviation, nuclear power, maritime operations, and cybersecurity. We don’t expect GPT-5 to solve the hallucination problem completely, but we expect it to significantly reduce the possibility of such incidents. There is no specific launch date for GPT-5, and most of what we think we know comes from piecing together other information and attempting to connect the dots. Knowing I have access to these tools expands my willingness to use them. We need to move from the technical aspects of these systems to what they actually do.

We’re now into the third year of the AI boom, and industry leaders are showing no signs of slowing down, pushing out newer and (presumably) more capable models on a regular basis. “We rolled it out for paid users about two months ago,” Kevin Weil, OpenAI’s chief product officer, said during Monday’s livestream. “I can’t imagine ChatGPT without Search now. I use it so often. I’m so excited to bring it to all of you for free starting today.” The free version of ChatGPT was originally based on the GPT 3.5 model; however, as of July 2024, ChatGPT now runs on GPT-4o mini. This streamlined version of the larger GPT-4o model is much better than even GPT-3.5 Turbo. It can understand and respond to more inputs, it has more safeguards in place, provides more concise answers, and is 60% less expensive to operate.

It also appears to be adept at emulating handwriting, to the point that some prompts might create images indistinguishable from real human output. Since GPT-4 is already the basis of much of the hype around generative AI, 4o could be poised to send shockwaves throughout the industry. Here’s everything that OpenAI revealed about the new AI technology, and why it’s a big step forward. OpenAI has been the target of scrutiny and dissatisfaction from users amid reports of quality degradation with GPT-4, making this a good time to release a newer and smarter model. Neither Apple nor OpenAI have announced yet how soon Apple Intelligence will receive access to future ChatGPT updates. While Apple Intelligence will launch with ChatGPT-4o, that’s not a guarantee it will immediately get every update to the algorithm.

The first public demonstration of GPT-4 was livestreamed on YouTube, showing off its new capabilities. Many of the displayed voice assistant capabilities were impressive but the live translation tool really seemed take it up a notch. During live demos, OpenAI presenters asked the voice assistant to make up a bed time story. Through the demo they interrupted it and had it demonstrate the ability to sound not just natural but dramatic and emotional. They also had the voice sound robotic, sing and tell the story with more intensity.