Image credits: Didem Minty/Anadolu Agency/Getty Images
ChatGPT, the innovative AI-based chatbot that has taken the world by storm over the past 12 months, was historically limited to data until September 2021 – making it useless as a real-time search engine. However, OpenAI started bringing web services to ChatGPT back in March, a move that always came with inherent risks, given that the live web isn’t regulated in the same way as a static training dataset — and this potentially opens doors to… Abuse by bad actors and good old algorithmic chaos.
Then in May, OpenAI It started rolling Search the web via Bing, Microsoft’s search engine backing OpenAI, before expanding access to the ChatGPT mobile app in late June. However, the new feature was quickly withdrawn after it was discovered that ChatGPT was able to display paywalled content.
Fast forward to late September, and OpenAI started Browse with Bing is rolling out againhaving fine-tuned how ChatGPT follows instructions set by content owners — essentially, it now promises to adhere to everything the site owner says in its Robots.txt file, similar to traditional web crawlers.
Now, browsing with Bing is officially available to all Plus and Enterprise subscribers, without having to toggle their trial switch in settings.
In related news, OpenAI has also moved DALL-E 3 into beta, a month after the latest incarnation of the text-to-image generator debuted.
DALL-E 3 sports integration with ChatGPT, which means users don’t have to think twice about their text prompts when they ask DALL-E to generate an image – ChatGPT can do a lot of the heavy lifting to ensure that the image the user gets is closer to what they want.
But on top of that, with DALL-E 3 integrated directly into ChatGPT, users will now be able to receive images as part of their text queries without having to switch between applications.
DALL-E 3 is available in beta now on web and mobile, with users able to activate the feature by selecting “DALL-E 3 (Beta)” from the GPT-4 tab within ChatGPT.
This all forms part of a broader expansion that is taking ChatGPT away from a purely text-based generator, and onto a path where audio and visuals are a big part of its remit.
Last month, OpenAI ChatGPT gave a mouth and ears to users able to have a verbal conversation with a chatbot, combining the worlds of Alexa-style voice assistants with powerful large language models (LLMs). For example, a user will be able to ask ChatGPT to create a bedtime story for their child and tell it on the spot, though it would probably be helpful to have it on hand to see what they come up with.
Additionally, ChatGPT will also allow users to search for answers using images, meaning anyone can upload a photo of an object and find out what it is or find similar items.
So today’s news fits very well with OpenAI’s endeavors to make ChatGPT a fully integrated, real-time multimedia generating search engine.
“Hipster-friendly explorer. Award-winning coffee fanatic. Analyst. Problem solver. Troublemaker.”