menu

menu

menu

AI and the future of documentation

AI and the future of documentation

How is AI changing how we create, deliver and consume documentation?

How is AI changing how we create, deliver and consume documentation?

AI is changing how teams create and deliver documentation. In this section, we look at how teams are using AI today, what they expect in the future, and how AI is changing the way people interact with docs.

60%

60%

are currently using generative AI as part of their docs workflows

42%

42%

THINk ai will let us build docs that intelligently adapt to user needs

“Over the past few years, our documentation has evolved into dynamic, user-centric knowledge systems — driven by a powerful combination of AI, automation, real-time feedback loops, and the irreplaceable insight of human expertise.”

Mira Balani

Staff Technical Writer

Miro

“Over the past few years, our documentation has evolved into dynamic, user-centric knowledge systems — driven by a powerful combination of AI, automation, real-time feedback loops, and the irreplaceable insight of human expertise.”

Mira Balani

Staff Technical Writer

Miro

“Over the past few years, our documentation has evolved into dynamic, user-centric knowledge systems — driven by a powerful combination of AI, automation, real-time feedback loops, and the irreplaceable insight of human expertise.”

Mira Balani

Staff Technical Writer

Miro

AI’s impact docs workflows and consumption

spotlight

Lee Robinson: Vercel's approach to AI-powered documentation

spotlight

Lee Robinson: Vercel's approach to AI-powered documentation

spotlight

Lee Robinson: Vercel's approach to AI-powered documentation

AI adoption is growing fast — but not everywhere

60% of companies said they’re already using generative AI in their documentation workflows, at least sometimes. About 31% use it often. But 25% said they don’t use AI at all.

Do you use generative AI in your documentation workflow?

AI isn’t just for big enterprises or tiny startups — usage is pretty even across company sizes. Interestingly, the percentage of mid-sized teams using AI is a little higher than small or large teams.

Size of companies that use generative AI in documentation workflows at least occasionally

How teams are using AI in their docs processes

When we asked how teams are using AI, most said it’s for writing and editing content. But 30% still aren’t using any AI tools in their docs process.

AI tools integrated into documentation processes

[Select all that apply]

So while AI is making a difference for many teams, it’s clear that human oversight is still essential to keep docs accurate, clear, and useful.

Teams expect AI to completely change how docs are made and used

Nearly half of the people we surveyed think AI will have a huge impact on documentation. 87% said it will be at least somewhat impactful.

How big do you think the impact of AI will be on the future of user-facing documentation?

When asked how they expect documentation to change, 42% said docs will start adapting automatically to what users need. Another 25% believe docs will be written mainly for AI and large language models (LLMs) to read and process.

How do you think AI will change the way we approach and format documentation in the future?

How docs show up when and where users need them

We’re already seeing this shift — with docs becoming smarter and showing up exactly when users need help. Tooltips, embedded guides, onboarding flows, and live chat are common now, and they’re just the start.

How do you currently provide documentation at point of need within your product?

[Select all that apply]

Looking ahead, teams expect more AI-powered assistants, chat interfaces, personalized help, and interactive guides. But no matter how advanced AI gets, it works best when paired with human oversight.

What new contextual formats do you think will play a larger role in documentation in the future?

[Select all that apply]

The companies seeing the best results with AI are the ones that set clear rules — knowing when to use AI, and when humans need to step in for quality and strategy.

Strategic implications

“Over the past few years, our documentation has evolved into dynamic, user-centric knowledge systems — driven by a powerful combination of AI, automation, real-time feedback loops, and the irreplaceable insight of human expertise.”

Mira Balani

Staff Technical Writer

Miro

“Over the past few years, our documentation has evolved into dynamic, user-centric knowledge systems — driven by a powerful combination of AI, automation, real-time feedback loops, and the irreplaceable insight of human expertise.”

Mira Balani

Staff Technical Writer

Miro

“Over the past few years, our documentation has evolved into dynamic, user-centric knowledge systems — driven by a powerful combination of AI, automation, real-time feedback loops, and the irreplaceable insight of human expertise.”

Mira Balani

Staff Technical Writer

Miro

AI is transforming docs — but writers’ roles are expanding, not disappearing

AI isn’t replacing documentation teams — it’s changing what they focus on. AI can speed up documentation work, but it still needs human guidance to get things right. The best teams use AI to handle repetitive tasks, while writers focus on strategy, quality, and complex content.

Instead of spending time on repetitive writing tasks, writers are now:

– Reviewing and editing AI-generated content

– Organizing and structuring information

– Making sure docs stay accurate and useful

– Thinking strategically about how users find and use content

As Larry Ullman, Stripe’s first Technical Writer, explains:

“I suspect writers will spend more time in two areas. First, as editors, reviewing generated content, managing the AI, verifying technical accuracy, updating, etc. Second, I suspect writers can do more predictive and visionary docs product work.”

Larry Ullman

Technical Writer

Formerly at Stripe

“I suspect writers will spend more time in two areas. First, as editors, reviewing generated content, managing the AI, verifying technical accuracy, updating, etc. Second, I suspect writers can do more predictive and visionary docs product work.”

Larry Ullman

Technical Writer

Formerly at Stripe

“I suspect writers will spend more time in two areas. First, as editors, reviewing generated content, managing the AI, verifying technical accuracy, updating, etc. Second, I suspect writers can do more predictive and visionary docs product work.”

Larry Ullman

Technical Writer

Formerly at Stripe

The companies seeing real success with AI are the ones that:

– Set clear rules for when to use AI

– Keep writers in charge of final reviews

– Use AI to assist, not to replace

AI is a powerful tool, but without human input, it can’t deliver the clarity, accuracy, and user focus that great documentation needs.

Docs must shift from static pages to dynamic conversations

AI is pushing documentation beyond static pages. More companies are starting to think of docs as conversations — where users ask questions and get answers, instead of digging through long manuals.

This shift is already happening with AI-powered chatbots, smart search, and tools that guide users based on what they’re trying to do.

But for this to work well, teams need a strong base of clear, structured content. AI can only deliver good answers if it has high-quality docs to pull from.

“My very subjective hunch is that users are going to consume the docs mainly through chatbots. While the docs must remain accessible in a traditional way — through a website, search, or navigation — the main way to look up something in the docs will be by asking a chatbot to produce an answer based on those docs.”

Marco Spinello

Senior Technical Writer

Booking.com

“My very subjective hunch is that users are going to consume the docs mainly through chatbots. While the docs must remain accessible in a traditional way — through a website, search, or navigation — the main way to look up something in the docs will be by asking a chatbot to produce an answer based on those docs.”

Marco Spinello

Senior Technical Writer

Booking.com

“My very subjective hunch is that users are going to consume the docs mainly through chatbots. While the docs must remain accessible in a traditional way — through a website, search, or navigation — the main way to look up something in the docs will be by asking a chatbot to produce an answer based on those docs.”

Marco Spinello

Senior Technical Writer

Booking.com

“Our users are using [an AI chatbot pointing to our docs] to try and unblock themselves, and I think that’s the main way this is providing value for us.”

Danny Neira

Support Engineer

Warp

“Our users are using [an AI chatbot pointing to our docs] to try and unblock themselves, and I think that’s the main way this is providing value for us.”

Danny Neira

Support Engineer

Warp

“Our users are using [an AI chatbot pointing to our docs] to try and unblock themselves, and I think that’s the main way this is providing value for us.”

Danny Neira

Support Engineer

Warp

The future of docs isn’t just about writing pages — it’s about creating systems where users can interact, explore, and get exactly the help they need, when they need it.

Clear, well-structured docs are the foundation for successful AI

AI is only as good as the content it works with. If your documentation is messy, outdated, or unclear, AI will just spread that confusion faster.

That’s why teams need to focus on getting the basics right — clear writing, good structure, and up-to-date content — before leaning too hard on AI tools.

“AI is going to be looking at your docs. So if you’re using AI to generate your docs, they’re probably not going to be that great. Instead, it should be training on your documentation to be able to provide good and clear answers. And so this is why I think, now more than ever, you should really be focusing on providing clarity, being comprehensive, and helping your user.”

Bekah Hawrot Weigel

Senior Developer Experience and Content Manager

The Linux Foundation

“AI is going to be looking at your docs. So if you’re using AI to generate your docs, they’re probably not going to be that great. Instead, it should be training on your documentation to be able to provide good and clear answers. And so this is why I think, now more than ever, you should really be focusing on providing clarity, being comprehensive, and helping your user.”

Bekah Hawrot Weigel

Senior Developer Experience and Content Manager

The Linux Foundation

“AI is going to be looking at your docs. So if you’re using AI to generate your docs, they’re probably not going to be that great. Instead, it should be training on your documentation to be able to provide good and clear answers. And so this is why I think, now more than ever, you should really be focusing on providing clarity, being comprehensive, and helping your user.”

Bekah Hawrot Weigel

Senior Developer Experience and Content Manager

The Linux Foundation

AI can help scale your docs and deliver answers quickly, but it relies on strong human-created content to do that well. Skipping this step leads to bad user experiences — faster.

The takeaway? Fix your content first, then add AI.

“AI just accentuates all of the existing best practices. Writing good content doesn’t really go anywhere. Actually, it’s even more important because you need to write something that’s accessible to both humans and AIs.”

Lee Robinson

VP of Product

Vercel

“AI just accentuates all of the existing best practices. Writing good content doesn’t really go anywhere. Actually, it’s even more important because you need to write something that’s accessible to both humans and AIs.”

Lee Robinson

VP of Product

Vercel

“AI just accentuates all of the existing best practices. Writing good content doesn’t really go anywhere. Actually, it’s even more important because you need to write something that’s accessible to both humans and AIs.”

Lee Robinson

VP of Product

Vercel

As AI handles more of the routine work, writers are becoming content architects, strategists, and editors — making sure the overall documentation experience actually helps users succeed.

Users expect help exactly when and where they need it

People don’t want to search through long docs when they run into a problem — they expect answers to appear right when they need them. That’s why more teams are focusing on point-of-need documentation, where help is built directly into the product experience.

“I think more interactive content is probably going to be the future. Instead of just a static helper file where you’re going through and reading content, you’re going to be able to ask questions and engage with the help system.”

Rob Gray

Senior Technical Writer

Snowflake

“I think more interactive content is probably going to be the future. Instead of just a static helper file where you’re going through and reading content, you’re going to be able to ask questions and engage with the help system.”

Rob Gray

Senior Technical Writer

Snowflake

“I think more interactive content is probably going to be the future. Instead of just a static helper file where you’re going through and reading content, you’re going to be able to ask questions and engage with the help system.”

Rob Gray

Senior Technical Writer

Snowflake

AI can help deliver that content at the right moment, but it can’t do much if the underlying documentation isn’t well-written and well-organized. Teams that succeed here are the ones that think beyond traditional docs pages and design for when, where, and how users actually need support.

The future of documentation is less about expecting users to come find answers — and more about making sure the answers find them.

New metrics are needed to track docs’ real impact

Traditional metrics like page views and ticket deflection don’t tell the whole story anymore. As documentation becomes more integrated into products — and as AI starts delivering content in new ways — teams need better ways to measure success.

“Since introducing AI-powered features, we have noticed an increase in page views, a decrease in page bounces, and more pages being visited — especially in the case of nested pages. This may indicate that AI-enhanced search provides more accurate and relevant results.”

Marco Spinello

Senior Technical Writer

Booking.com

“Since introducing AI-powered features, we have noticed an increase in page views, a decrease in page bounces, and more pages being visited — especially in the case of nested pages. This may indicate that AI-enhanced search provides more accurate and relevant results.”

Marco Spinello

Senior Technical Writer

Booking.com

“Since introducing AI-powered features, we have noticed an increase in page views, a decrease in page bounces, and more pages being visited — especially in the case of nested pages. This may indicate that AI-enhanced search provides more accurate and relevant results.”

Marco Spinello

Senior Technical Writer

Booking.com

Future-focused teams are starting to track things like:

– How docs improve user onboarding and activation

– How often docs help close deals or drive product adoption

– How documentation supports customer retention and long-term success

These kinds of metrics connect documentation directly to business outcomes, not just support efficiency. To keep up with how documentation is evolving, teams will need to rethink what they measure, and focus on how docs drive real impact across the entire customer journey.

© 2025 Copyright GitBook INC.

440 N Barranca Ave #7171, Covina, CA 91723, USA. EIN: 320502699

© 2025 Copyright GitBook INC.

440 N Barranca Ave #7171, Covina, CA 91723, USA. EIN: 320502699

© 2025 Copyright GitBook INC.

440 N Barranca Ave #7171, Covina, CA 91723, USA. EIN: 320502699