6 realistic use-cases for AI on your intranet
AI is impacting every aspect of work and every corner of the digital workplace, including the humble intranet. Intranet software providers are selling AI-powered platforms. Microsoft Copilot is hard baked into SharePoint and the Microsoft 365 digital workplace, meaning that conversational interfaces and AI agents are already potentially part of your intranet. And many employees are using AI to generate content, including intranet pages and news items.
Going forward, some even believe that intranets might not even be needed with the advent of AI. While we definitely don’t agree, we do think that the nature of intranets and how people interact with them will change.
Six great use cases for AI on your intranet
AI is already here and is waiting to improve your intranet. But what are the best ways to use it? Where will it make the most difference? Let’s explore six realistic use cases for AI which can help improve user experience, save time and drive intranet adoption today.
Search and findability
A critical job for any intranet is to help employees find the information they need to complete tasks and get stuff done. AI has obvious power to radically improve search relevance and accuracy, while also evolving the search experience for the user:
- Offering the opportunity for employees to use a conversational interface or natural language for queries.
- Delivering semantic search with greater understanding of search intent through natural language processing.
- Demonstrating greater capacity to understand the context of searches, for example based on someone’s profile or previous search behaviour and then delivering personalisation at scale.
- Reciting, summarizing or guiding a user towards the part of a document or page that best satisfies the search query.
- Summarising individual search results to give users a better understanding of the value of each item.
- Working more efficiently and effectively over a number of different sources.
- And more!
Answering employee questions
ChatGPT, Claude, Gemini and other popular AI services are changing the way that some users interact with technology, especially with greater use of natural language. Using AI to answer employee questions using natural language queries through the intranet is a natural evolution for many of us.
Chatbots have been part of the intranet landscape for quite a long time now. Bots, copilots and agents have been embedded into the intranet experience and answer particular questions on topics such as corporate policies, HR questions or IT support, for example. Up to now the results have been a mixed bag, but generative and agentic AI has the potential to make bots far more effective. If you are using Microsoft Copilot, it is already very likely being used for employee queries.
There is also some overlap with AI-powered search here, as employee questions are sometimes answered by finding the right section of a document, summarising answers from content, or linking to the content – or a combination of all three.
Content generation
Creating content remains a core use for all generative AI and it is an obvious use case for intranets. Most intranets are built on a decentralised publishing model which relies on the active contribution from content owners, publishers and authors cross a business. Most of these people will not be trained communicators and will have a very busy “day job” so producing purposeful and valuable content for the intranet is not necessarily top of mind.
Generative AI can create new content for the intranet, particularly as a starting point, and lower the barrier for busy content owners. It not only speeds up the time involved but could also generate a relevant page from say a policy or instructional document. It can also help to explore content designs by creating mock-ups for what a page might look like, or act as a critique to help guide creation.
While there is little risk involved using AI to support content generation for communicators and experienced intranet teams who can ensure brand alignment and publishing standards are met, there is more risk with content owners through the business. Using AI makes it all too easy to create intranet content that is bland, has little value or doesn’t meet required standards. To mitigate for this, it helps to:
- Involve review of any AI-generated content in your publishing workflow.
- Train content owners on how to use AI and ensure they are aware of publishing standards.
- Set rules within the AI to reference publishing standards and guidelines to influence the output – see more on content optimisation and governance below!
Content optimisation
Intranet content management also needs to be optimised to meet publishing standards, make it more findable and more. Generative AI and agentic AI has an obvious role to play in doing much of the heavy lifting in this area, reducing the effort for both busy content publishers and even busier central intranet teams.
Theare are multiple ways AI can help make intranet content better:
- Automatically adding tags and other metadata to make content more findable, particularly if tags are derived from enterprise taxonomies
- Providing automatic summaries that could appear in other places in the intranet such as search or content spotlights
- Making intelligent suggestions in the editing experience to guide content owners and editors to improve content
- Helping to automatically apply content governance and formatting rules to existing content
- Helping content owners to add or even create the right image for an item
- And more.
Translation
Machine translation has been used in intranets for a while to support multi-language content within global companies or where a section of the workforce has a significant proportion of non-native speakers. The quality of translation was already pretty good before the “generative AI” era, but AI has helped to broaden the coverage of languages that have seen improvement.
In an intranet, machine translation can work in different ways:
- It can automatically translate all content into a preferred language in a user’s settings.
- It can translate content on the fly based on an action / request by the user.
- It can work as part of a translation workflow which also involves some human oversight and potentially manual translation, with the machine translation as the starting point.
Content governance
Successful intranets rely on content that is useful, up to date, accurate, engaging, findable, compliant and on brand. The “secret sauce” behind keeping intranet content valuable is governance, usually to ensure content is regularly reviewed and meets certain standards. AI can support intranet content governance in a number of ways:
- Checking to ensure content meets defined brand and publishing standards.
- Spotting if there are any potential security or data privacy issues.
- Making suggestions for content that may have expired based on defined or suggested criteria
- Identifying patterns of which content is regularly updated and when.
- Intelligently using analytics to support content governance processes.
- And more!
Birth of the AI-powered intranets
The era of the AI-powered intranet is already here and we’re already seeing that rather than AI replacing your intranet, it is actually going to enhance it, particularly across the six use cases we’ve featured in this article. If you’d like to see how you can extend the power of your AI-powered SharePoint intranet with Lightspeed365 features, then why not arrange a free demo?Frequently asked questions
Will AI replace the intranet?
While it may change the way people interact with intranets, AI is making intranets better rather than reducing the need for them. Intranets also often act as “the one source of truth” for businesses with reliable and up-to-date information, which is exactly the content AI needs to be valuable.
What is the best use case for using AI on the intranet?
There are currently several strong use cases for using AI on the intranet. These include improving search and findability, replying to employee questions, generating new intranet content, optimising content for intranets, using machine translation to support multi-language intranets and also supporting content governance.
About the author
John Scott
