Jaw-Dropping Features Of GPT-4 That GPT-3.5 Couldn’t Do
OpenAI announced the next generation of AI technology on Tuesday, they introduced this version of chatGPT by showing us all the different features that were improved and all the features that were added separately such as,
Analyzing Images:
The biggest change to GPT-4 is its ability to work with photos that users upload, one of the most jaw-dropping features so far showed us how the demonstrator uploaded the image into GPT-4 and then pasted the resulting code into a preview that showed how it could be a working website.
Open AI also showed how GPT-4 was asked to explain a joke from a series of images, which featured a mobile phone with the wrong charger and then to describe why it was funny. While it could sound a bit straightforward, dissecting a joke is more complicated for AI tools to learn because of the required context.
Coding Has Been Made Easier Than Before:
Some early GPT-4 users with little to no prior coding knowledge have also used it to recreate iconic games like Tetris or snake. Others have made their own original games, “The powerful language capabilities of GPT-4 will be used for everything from storyboarding, and character creation to gaming content creation,” said Arun Chandrasekaran.
“This could give rise to more independent gaming providers in the future. But beyond the game itself, GPT-4 and similar models can be used for creating marketing content around game previews, generating news articles, and even moderating gaming discussion boards.”
Just like gaming, GPT-4 could change the way people develop applications. One user on Twitter said that they made a simple drawing app in minutes while another claimed to have coded an application that recommends five new movies to watch every day, along with providing trailers and details on where to watch them.
This is a game changer and an absolute want of people who like to binge-watch movie series and don’t know any other good ones to watch which is why this is a creative attempt made possible only through ChatGPT’s advanced coding capabilities.
Responding With More Accuracy:
Compared to the previous version, GPT-4 is able to produce longer and more detailed answers and more reliable responses, the latest version can now give responses of up to 25,000 words, up from about 4,000 previously, and can provide detailed instructions for different scenarios, as the intro showed us, ChatGPT gave complete and efficient instructions on how to clean a piranha’s fish tank, which immediately makes it seem like more reliable as the company is flexing its reliability by asking the bot for instructions in a somewhat dangerous scenario.
Streamlining Work Across Various Industries:
Joshua Browder, CEO of legal service chatbot, DoNotPlay, said his company is already working on using the tool to generate “one-click-lawsuits” to sue robo-callers, in an early indication of the vast potential for GPT-4 to change how people work across industries.
“Imagine receiving a call, clicking a button, [the] call is transcribed, and the 1,000-word lawsuit is generated. GPT-3.5 was not good enough, but GPT-4 handles the job extremely well,” Browder said in a tweet.
Meanwhile, Jake Kozloski, the CEO of dating site keeper, said his company is using the tool to better match its users.
And according to Su at ABI research, it’s possible we will also see major advancements in “connected car [dashboards], remote diagnosis in healthcare, and other AI applications that were previously not possible.”
Passing Tests With Ease:
OpenAI said that the update is “less capable” than humans in many real-world scenarios, it exhibits “human-level performance” on various professional and academic tests. The company further said that GPT-4 recently passed a simulated law school bar exam with a score around the top 10%. The latest version performed the same test and it performed strongly on the LSAT, GRE, SATs, and many AP exams, according to OpenAI.
ChatGPT made headlines for its capabilities of passing prestigious graduate-level examinations in January, such as one from the University of Pennsylvania’s Wharton School of Business, but not with uniquely high marks. The company said that it spent months using lessons from its testing program and ChatGPT to improve the system’s accuracy and ability to stay on top.