Skip to main content

Copilot: How will AI change Microsoft 365?


 

Microsoft 365 is the world's most popular productivity suite, with more than 300 million active users. It offers a range of apps and services that help people work, learn, and collaborate more effectively. But what if Microsoft 365 could do even more for you? What if it could be your copilot for work?


That's the vision behind Microsoft 365 Copilot, a new feature that leverages the power of large language models (LLMs) and your data in the Microsoft Graph to transform your words into actions. Copilot is integrated into the Microsoft 365 apps you use every day, such as Word, Excel, PowerPoint, Outlook, Teams, and OneNote. It can also work across these apps and your data through a new experience called Business Chat.


Copilot can help you unleash your creativity, unlock your productivity, and uplevel your skills by doing things you've never been able to do before. You can give it natural language prompts and it will generate high-quality content for you, based on your context and preferences. You can also ask it to edit, summarise, rewrite, or format your existing content. You can even get suggestions for different tones, styles, and formats to suit your audience and purpose.


Here are some examples of how Copilot can change the way you work with Microsoft 365:


  • In Word, you can draft a project proposal based on data from a document and a spreadsheet with just a brief prompt. Copilot will create a first draft for you, bringing in information from across your organisation as needed. You can also change the tone of the document to be more casual or professional, depending on who you're writing for.
  • In Excel, you can analyse data and create charts with simple commands. Copilot will understand your intent and generate the best visualisations for your data. You can also ask it to explain the insights of trends in your data in plain language.
  • In PowerPoint, you can create slides from scratch or from existing documents with minimal input. Copilot will design the slides for you, using relevant images, icons, and layouts. You can also ask it to summarise the main points of your presentation or add transitions and animations.
  • In Outlook, you can compose emails faster and better with Copilot's assistance. Copilot will suggest subject lines, greetings, closings, and signatures based on your recipients and message content. You can also ask it to schedule meetings, attach files, or follow up on tasks.
  • In Teams, you can collaborate more effectively with Copilot's help. Copilot will suggest responses, reactions, and emojis based on the conversation context and tone. You can also ask it to create polls, surveys, or quizzes for your team members or students.
  • In OneNote, you can capture and organise information more easily with Copilot's guidance. Copilot will create notes, plans, lists, and outlines for you based on your prompts. You can also ask it to transform existing text by changing the formatting, removing repetitive or unwanted information, or adding visual context.


Business Chat is a new way to interact with Copilot across the Microsoft 365 apps and your data. It works like a chatbot that can do anything you ask it to do related to your work. You can type or speak natural language queries and commands and Copilot will respond with relevant content or actions. For example:


  • Tell my team how we updated the product strategy.
  • Create a budget report for Q1 based on [a spreadsheet].
  • Show me my calendar for next week and book a flight to New York.
  • Find a good restaurant near my hotel and make a reservation for two.
  • Write a thank-you note to [a contact] for their feedback.


Microsoft 365 Copilot is powered by large language models (LLMs), which are AI systems that can produce natural language text from prompts and data. Microsoft has been investing in LLM research and development for years and has recently partnered with OpenAI to use their state-of-the-art GPT-4 model. GPT-4 is trained on billions of words from the internet and can generate coherent and diverse texts on any topic.


Microsoft 365 Copilot also uses your data in the Microsoft Graph, which connects your calendar, emails, chats, documents, meetings, contacts, and more. This allows Copilot to understand your context and preferences better and generate personalised content for you. For example:


- If you ask Copilot to draft an email to a colleague about a project update, it will use your previous emails with that person as well as the project documents


Labels:

Comments

Post a Comment

Popular posts from this blog

The Whispers in the Machine: Why Prompt Injection Remains a Persistent Threat to LLMs

 Large Language Models (LLMs) are rapidly transforming how we interact with technology, offering incredible potential for tasks ranging from content creation to complex analysis. However, as these powerful tools become more integrated into our lives, so too do the novel security challenges they present. Among these, prompt injection attacks stand out as a particularly persistent and evolving threat. These attacks, as one recent paper (Safety at Scale: A Comprehensive Survey of Large Model Safety https://arxiv.org/abs/2502.05206) highlights, involve subtly manipulating LLMs to deviate from their intended purpose, and the methods are becoming increasingly sophisticated. At its core, a prompt injection attack involves embedding a malicious instruction within an otherwise normal request, tricking the LLM into producing unintended – and potentially harmful – outputs. Think of it as slipping a secret, contradictory instruction into a seemingly harmless conversation. What makes prompt inj...
Post a Comment
Read more

AI Agents and the Latest Silicon Valley Hype

In what appears to be yet another grandiose proclamation from the tech industry, Google has released a whitepaper extolling the virtues of what they're calling "Generative AI agents". (https://www.aibase.com/news/14498) Whilst the basic premise—distinguishing between AI models and agents—holds water, one must approach these sweeping claims with considerable caution. Let's begin with the fundamentals. Yes, AI models like Large Language Models do indeed process information and generate outputs. That much isn't controversial. However, the leap from these essentially sophisticated pattern-matching systems to autonomous "agents" requires rather more scrutiny than the tech evangelists would have us believe. The whitepaper's architectural approaches—with their rather grandiose names like "ReAct" and "Tree of Thought"—sound remarkably like repackaged versions of long-standing computer science concepts, dressed up in fashionable AI clot...
Post a Comment
Read more

Podcast Soon Notice

I've been invited to make a podcast around the themes and ideas presented in this blog. More details will be announced soon. This is also your opportunity to be involved in the debate. If you have a response to any of the blog posts posted here, or consider an important issue in the debate around AGI is not being discussed, then please get in touch via the comments.  I look forward to hearing from you.
Post a Comment
Read more