In March 2023, Microsoft announced Microsoft 365 Copilot. Copilot will leverage the power of large language models alongside your Microsoft Graph data to reinvent productivity.
Copilot will be integrated into the Microsoft 365 apps we use every day (Outlook, Word, Excel, PowerPoint, Teams, Viva Engage, etc.). You’ll be able to use natural language commands to prompt Copilot (e.g., draft a 2-page proposal in Word based on the contents of this Excel spreadsheet and PowerPoint presentation). You’ll then be able to work with Copilot to refine the text, change its voice, make it more concise, etc.
Microsoft is testing Copilot in a small private preview with 20 customers today. Microsoft says they’ll be expanding the pilot, with more release details to come in the Microsoft Roadmap. Expect Copilot to be a key part of Microsoft’s announcements at upcoming conferences. It’s going to be a huge year!
How AI (Artificial Intelligence) will continue to change the way we work
Examples of the productivity enhancements Copilot will bring to Word, Excel, PowerPoint, and Outlook
Ways in which we believe Copilot can help us manage our Teams meetings (e.g., summarizing meeting topics discussed, catching users up to speed on topics discussed if they arrive late, etc.)
How Copilot can prompt leaders with suggested verbiage for Viva Engage posts
A quick overview of the newly announced Microsoft Security Copilot
On March 16, Microsoft launched its latest AI-powered innovation in Microsoft 365 CoPilot, aimed at using AI to remove the drudgery of many Office tasks. We don't know when the new capabilities will be available or what they will cost. The demos looked wonderful, but how things work out when exposed to real users is an entirely different manner. Time will tell!
Merci à vous de suivre le flux Rss de www.sospc.name. ;o)<
Dans la plupart des articles que l'on peut lire actuellement sur la toile on a du mal à comprendre ce qu'est en fait ChatGPT. Je suis ChatGPT, un modèle de langage développé par OpenAI. Mon objectif est de répondre à une grande variété de questions, de fournir des informations et d'engager des conversations avec les […]
Chatbot Delivers Answers, But Usable Responses Are a Different Matter
The launch of the free research version OpenAI’s ChatGPT project generated lots of reactions, with some journalists predicting that using AI in this manner could mark the end of Google search results. According to Chris Johns, an economist whose podcast I subscribe to, the chatbot is capable of producing answers that meet the standard of first year university exams. Closer to home, MVP Doug Finke (author of the ImportExcel PowerShell module) thought the results generated for PowerShell questions were impressive (here’s his YouTube video).
Given the opinions voiced, I decided to sign up to test ChatGPT. My conclusion is that the chatbot is an idiot savant when it comes to technology. The answers generated by ChatGPT are plausible and cogent in some areas, but once it goes outside its area of comfort, the answers become weaker and weaker.
The Need for Good Source Material
By its very nature, AI depends on the source material used to train models. Inside Microsoft 365, a trainable classifier doesn’t work in scenarios like auto-label policies unless the set of source documents used to create the model underpinning the classifier are good enough. In the case of ChatGPT, OpenAI admit that the material used to build the model comes from 2021 or earlier. Given the nature of technology, especially cloud services, out-of-date information leads to bad answers.
A problem also arises when source material is wrong or contains information that might be accurate at a point in time but will be superseded by developments. This happens all the time in blog posts. For example, if you search for something like “How to update Azure AD accounts with PowerShell,” you’ll get a bunch of responses describing how to perform the task using cmdlets from the Azure AD or Microsoft Online Services (MSOL) modules. Posts published last week that I know of still reference these cmdlets, but people working in this space know that Microsoft plans to deprecate both modules in June 2023. The upshot is that the answer is right, works today, but is flawed because the code will stop working in six months. The lack of awareness of context is a flaw of AI and that shows through in its answers.
Asking About Azure AD Accounts
Take the example shown in Figure 1. The chatbot response to the question is inaccurate for two reasons: I asked about finding Azure AD accounts with the Microsoft Graph. The response is to use the soon-to-be-deprecated Azure AD module. There’s no trace of a Graph API request or the Microsoft Graph PowerShell SDK cmdlets.
Figure 1: Asking ChatGPT about finding Azure AD accounts
I have no idea why my question might have violated OpenAI’s content policy. That’s just a glitch. The important thing is that the code generated by ChatGPT works. Even though I wouldn’t use the Azure AD module now, the code runs perfectly and is a valid answer to the question
The Microsoft Graph PowerShell SDK existed in 2021, so I decided to check what the chatbot knew about the SDK. Figure 2 is the result. I think this is a good example of the ability of ChatGPT to generate a reasonably cogent (if wordy) answer in response to a question. The text is rather like the response you’d get from a Microsoft marketing person, but that’s another story.
Figure 2: ChatCPT discusses the Microsoft Graph PowerShell SDK
Testing a Real-Life Question
As a test of a real-life question, I took one about mailbox archiving from Practical365.com and input it to ChatGPT. The answer (Figure 3) is just plain wrong. First, only Exchange Online mailbox retention policies operate against archive mailboxes. Second, neither Microsoft 365 nor Exchange Online retention policies (there is no such thing as an online archiving policy) operate on the basis of mailbox size. Retention, including move to archive, is driven by item age. Like any assertion from a consultant, the confident nature of the response means that it might be accepted by someone who doesn’t know the technology. It seems like the text might be influenced by the way that Exchange Online expandable archives work, but the context is all wrong and the answer isn’t at all helpful.
Figure 3: ChatGPT gets mailbox archiving wrong
Finally, I asked about the world’s best Office 365 book. I was amused that ChatGPT recommended Office 365 for IT Pros but got the authors wrong. I have never met Ben Curry and he’s never been involved with the book, but hey, it’s still a highly plausible answer.
Figure 4: Who’s the Ben Curry guy that ChatGPT thinks wrote the Office 365 for IT Pros book?
Interesting but Flawed
The bottom line is captured in OpenAI’s admission that “ChatGPT sometimes writes plausible-sounding but incorrect or nonsensical answers.” This, allied to the other flaw that “The model is often excessively verbose and overuses certain phrases, such as restating that it’s a language model trained by OpenAI” means that you can’t trust the chat bot’s responses to any question about technology that evolves quickly. Answering some basic PowerShell questions is fine. Seeking help to administer Office 365 is quite another matter.
ChatGPT is interesting and worthwhile technology that points to the way we might seek information in the future. Based on a $1 billion investment, Microsoft and OpenAI have been working since 2019 and OpenAI trained the ChatGPT model on Azure. With that kind of backing, I’m sure that OpenAI will improve the model and increase the accuracy of the answers that it generates. But for now, I think I shall stick with querying Google and sorting the wheat out of whatever chaff Google replies with.
Stay updated with developments across the Microsoft 365 ecosystem by subscribing to the Office 365 for IT Pros eBook. We do the research to make sure that our readers understand the technology.
Last week, we had a client wondering why they saw this in the Search & Intelligence Admin Center.
Not only was Top queries showing “No data”, but there was “No data” for No results queries and Abandoned queries, too. This was the case even when we changed the filter from the last 31 days to the last 12 months. Surely people were searching for something!
We decided to try putting in a ticket with Microsoft to find out why it wasn’t working. At least it wasn’t working in such a way where we could get any “intelligence” from it.
After about a week, I finally got past the first line of defense at Microsoft (who told me to try a different account, etc. – nothing useful) to find out why we weren’t seeing much in the Search & intelligence stats. Unfortunately, the answer isn’t a good one.
It seems like the search queries are generated only when the searches are made on the Sharepoint Landing Page and not within the Sharepoint site search. Here is the information for your reference.
“You can get the tenant level usage analytics reports in Microsoft Admin center under Settings> Search and Intelligence> Insights to access the 5 usage analytics reports (top queries, abandoned queries, no result queries, query volume, and Impression distribution) aggregated over SharePoint home (This only covers searches made from the SharePoint landing page) and office.com workloads.”
(And yes, he spelled SharePoint wrong.) I vaguely remember learning this from Mikael Svenson (@mikaelsvenson) long ago. I probably blocked it out of my mind.
Rereading the articles in the References section below – with a fine-toothed comb- I see the answer was there, too.
With data only coming from “SharePoint Home (the site with URL ending in /SharePoint.aspx), Office.com, and Microsoft Search in Bing work tab search boxes”, we get barely a slice of the searches people actually do, and there’s very little intelligence we can get from such narrowly scoped analytics. Now that the Microsoft 365 search box is at the top of almost every page in Microsoft 365, I’d expect we could see all the search queries people do. I’m hoping this isn’t a permanent state.
As an Intranet manager, one of the most valuable things you can know is what people are searching for. It tells you what content is missing or where you haven’t done a good job in building the navigation or other parts of the information architecture. Unfulfilled searches mean content is missing entirely or maybe the permissions aren’t set right on content which is available. Setting up search isn’t a “one and done” thing. Monitoring and mining search should be a role at every organization with electronic content. Without good tools to do so, we’re driving blind.
Connexion
