Skip to main content

Bing Chat just got so much better in two important ways

Bing Chat is already one of the fastest and more reliable AI chat tools available, and Microsoft just made it more powerful with several significant upgrades.

Possibly the most valuable changes have to do with Bing Chat’s memory, allowing longer conversations. Previously, Microsoft limited how long you could chat with Bing before a fresh start was required to prevent a frightening AI meltdown. That has been extended, but Microsoft’s blog post didn’t share particular details.

Bing Chat itself told me there was no fixed limit to the length of our chat. The chat box still showed a 2,000-character limit in my testing, but I could paste a link and ask Bing to read it.

Bing Chat no seems to have no limit on chat length.
Bing Chat no seems to have no limit on chat length. Image used with permission by copyright holder

For example, I gave Bing Chat a link to a lengthy Wikipedia entry about the U.S. Constitution, and it was able to quickly summarize the contents. I asked Bing Chat about the length of the Wiki page, and it said it was about 17,000 words.

Microsoft also claimed performance has improved, particularly in the Edge sidebar, but I noticed no significant changes. Bing Chat has always felt fast, considering some of the tasks it’s presented with. The summary of a 17,000-word entry took about 15 seconds.

You can create images in the Edge sidebar with Bing Chat,
You can create images in the Edge sidebar with Bing Chat. Image used with permission by copyright holder

Other upgrades to Bing Chat include the ability to generate images in the sidebar and continue conversations after closing it or navigating to a new page. I asked Bing Chat for a charcoal drawing of a quokka, closed the sidebar, and opened a new tab with an image search. Then I tested its memory by instructing, “now do it as a super sharp color photo with bokeh lights in the background.”

Bing Chat quickly responded in the sidebar with four, adorable critters in front of charming bokeh-rich lights. The accuracy was spot-on, as seen in the image search.

Bing Chat continues to improve. Google is trying to catch up with its own AI release called Bard. So far, OpenAI and Microsoft are still leading the way with refined GPT-4 technology.

Alan Truly
Alan is a Computing Writer living in Nova Scotia, Canada. A tech-enthusiast since his youth, Alan stays current on what is…
A dangerous new jailbreak for AI chatbots was just discovered
the side of a Microsoft building

Microsoft has released more details about a troubling new generative AI jailbreak technique it has discovered, called "Skeleton Key." Using this prompt injection method, malicious users can effectively bypass a chatbot's safety guardrails, the security features that keeps ChatGPT from going full Taye.

Skeleton Key is an example of a prompt injection or prompt engineering attack. It's a multi-turn strategy designed to essentially convince an AI model to ignore its ingrained safety guardrails, "[causing] the system to violate its operators’ policies, make decisions unduly influenced by a user, or execute malicious instructions," Mark Russinovich, CTO of Microsoft Azure, wrote in the announcement.

Read more