The new Bing is now available to all (no more waiting list). And, it turns out that it’s been using GPT-4 all along!
We compared Bing to ChatGPT two issues ago and saw that they generated similar outputs from the same
prompt. That means we were inadvertently comparing GPT-4 to GPT-3. So, does that mean GPT-4 isn’t that big of a change from GPT-3? Yes and no.
As far as generating text, you likely won’t see a noticeable difference between the two models—especially if you know how to use temperature settings.
But besides having a larger parameter set, GPT-4 has some other nifty abilities, like being able to interpret and respond to images. Take a snapshot of your fridge, and it’ll give you recipes for what you can cook. It’s also better at passing exams and can handle longer prompts. This means we can input more contextual information, which can help generate better outputs.
So, where’s this all heading? 🤷
If you’d like to explore AI writing tools in-depth, check out the presale for our upcoming new course: Generative AI for UX Writers.