ChatGPT-4 has officially been announced, confirming the longtime rumors around its improvements to the already incredibly impressive language skills of OpenAI’s ChatGPT. (ChatGPT is an artificial intelligence chatbot developed by OpenAI and launched in November 2022. It is built on top of OpenAI’s GPT-3.5 and GPT-4 families of large language models and has been fine-tuned using both supervised and reinforcement learning techniques.)
OpenAI has called it the company’s “most advanced system, producing safer and more useful responses.”
When can you use it?
GPT-4 was officially announced on March 13 2023 as was confirmed ahead of time by Microsoft even though it is only available in the ChatGPT Plus paid subscription. The free offering still runs ChatGPT-3.5.
GPT-4 will also be accessible through an API “for developers to build applications and services.”
How to use it?
The easiest way to get started with GPT-4 today is to sign into Bing Chat in Microsoft’s Edge browser. Some features are still missing in Bing Chat such as visual input but you’ll still have access to the expanded LLM (large language model) and the advanced intelligence that comes with it. GPT-4 in Bing Chat is still limited to 15 chats per session and 150 sessions per day.
The paid subscription of GPT-4 is a $20 paid subscription. Select the upgrade option in Bing Chat and once you’ve entered your payment information you’ll be able to toggle between GPT-4 and older versions. You will know that you are accessing the latest version GPT-4 as they use a black logo instead of the older green logo.
What’s new in GPT-4?
According to OpenAI, this next-generation language model is more advanced in three key areas:
Creativity
In terms of creativity, OpenAI says GPT-4 is much better at both creating and collaborating with users on creative projects. Examples of these include music, screenplays, technical writing, and even “learning a user’s writing style.”
Longer Context
GPT-4 can now process up to 25,000 words of text from the user. You can even just send GPT-4 a web link and ask it to interact with the text from that page. OpenAI says this can be helpful for the creation of long-form content, as well as “extended conversations.”
Visual Input
GPT-4 can also now receive images as a basis for interaction. In the example provided on the GPT-4 website, the chatbot is given an image of a few baking ingredients and is asked what can be made with them. It is not currently known if video can also be used in this same way.
GPT-4 is also apparently significantly safer to use than the previous generation with 40% more factual responses in OpenAI’s own internal testing.
Limitations
OpenAI also notes some of the limitations of the new language model, it says the latest model still has problems with “social biases, hallucinations, and adversarial prompts.”
Microsoft originally states that Bing Chat was more powerful than ChatGPT but they have now confirmed that Bing Chat is built on GPT-4. Features like the visual input aren’t available on Bing Chat, so it’s not yet clear what exact features have been integrated and which have not.
Bing Chat clearly has been upgraded with the ability to access current information via the internet which is a vast difference from it’s predecessor that only had information up until 2021.
In addition to internet access, the AI model used for Bing Chat is much faster, something that is extremely important when taken out of the lab and added to a search engine. We’d be interested in knowing which safeguards are in place.
An evolution, not a revolution?
If you’ve tried out the new Bing Chat you’ve likely had a taste of GPT-4.
Sam Altman has acknowledged the potential of AI to get out of hand considering there are other companies running iterations of the AI without limitations. We look forward to seeing what the future holds for A.I and how it will affect entire economies and the workforce.