I Don't See a Future for MS Office
A while ago, I wrote on LinkedIn that I don’t see a future for MS Office. It sparked quite a debate – some people agreed, others shook their heads in disbelief. So I decided to gather my arguments in one place and develop them more thoroughly than a social media post allows.
Before I get into the details, let me state my thesis plainly: Microsoft Office in its current form is destined to die out. Not in twenty years. Not in ten. The process has already begun and is accelerating with each passing month.
Three Phases of Transition
To understand why Office is dying, you need to see the bigger picture of how our interaction with computers is changing. We’re passing through three fundamental phases:
Phase one: clicking buttons and verifying results. This is the world we’ve lived in for decades. You open Word, click “Insert Table,” format manually, check if it looks right. You open Excel, type a formula, verify the result. You open PowerPoint, drag elements around a slide, adjust fonts. Every operation requires a graphical user interface and your constant, undivided attention.
Phase two: issuing commands and verifying results. This is the phase we’re in right now. You tell the AI: “reformat this document,” “merge these two spreadsheets by the ID column,” “extract data from this PDF and put it in a table” – and you get your result. You still need to check whether everything is correct, but you no longer need to know where to click. You don’t need to know the interface. You don’t need to remember which menu hides that option.
Phase three: issuing commands without needing to verify. This is the phase we’re heading toward. When AI model error rates drop low enough – and they’re dropping with every new release – verification will only be necessary for the most critical matters. And that’s precisely when the entire Office interface – all those ribbons, context menus, dialog boxes, toolbars – will cease to have any relevance whatsoever.
Think about it for a moment. Why would you need a program’s interface if you never open it? Why would you need to know VLOOKUP when the AI executes the task on your command? Why would you need to know how to format a table in Word when all you have to say is “make it look nice”?
Of all the companies pushing us through these phases, the fastest is Anthropic with its Claude model. I’m not saying this as a fanboy – I’m saying this as someone who works daily with Claude, GPT, Gemini, and other models, and can see how Claude consistently outpaces the competition in office tasks.
What Claude Can Already Do Today
Instead of theorizing, let me show you what Claude can do in practice. Not next year, not in beta – right now, in production, every single day:
- Fills out forms – give it a form and source data, and it fills in all fields correctly, with proper formatting and validation.
- Edits PDFs – extracts data from PDF documents, modifies content, creates new documents from existing ones. Things that used to require Adobe Acrobat Pro and an hour of clicking.
- Copies document fragments and creates new ones from them – no manual copy-paste, no formatting that falls apart, no “why does this heading have a different font.”
- Modifies Excel files – changes data, recalculates, formats, creates new worksheets, merges files. More on this in a moment.
- Creates presentations – just recently, a dedicated PowerPoint plugin appeared that lets Claude directly manipulate slides.
These aren’t experimental lab features. This is the daily reality of thousands of people who, instead of clicking through the Office interface, simply type a command in a chat window.
More importantly, the quality of these operations is improving at a pace that’s hard to imagine. Just a year ago, Claude struggled with more complex spreadsheets. Today, it handles them flawlessly. Each new model version is noticeably better than the last at these kinds of tasks. And there’s no indication this trend is slowing down – quite the opposite.
Excel – The Best Proof
Of all the tools in the Office suite, Excel is perhaps the best example of what’s happening. Why Excel specifically? Because it was always considered “that advanced tool” that requires specialist knowledge to master. Companies spend fortunes on Excel training. Recruiters ask for “advanced Excel skills” in job postings. Microsoft Excel Specialist certificates are treated as serious qualifications.
Let me put it bluntly: there is no task involving data processing in xlsx or csv files that today’s AI cannot execute flawlessly.
And you don’t need to:
- open Excel, or
- know Excel.
All you need is a data file and access to AI. Nothing more.
By processing, I mean the full range of operations that used to require hours of work and advanced expertise:
- Merging and splitting data – combining worksheets from different files, breaking large datasets into smaller ones, joining data from various sources by a common key. What used to require VLOOKUP, INDEX/MATCH, or Power Query in Excel, AI does with a single command.
- Moving data – between worksheets, files, cells, in any configuration. No manual copying, no reference errors.
- Editing data – bulk changes, deduplication, format normalization, data type conversion. Tasks that required complicated formulas or macros in Excel.
- Statistical analysis – from simple averages and medians to advanced regression models, correlation analyses, statistical tests. No need to install additional plugins.
- Charts and visualizations – generated automatically, with appropriate chart type selection, axis formatting, legends. What takes a dozen clicks and constant tweaking in Excel.
Think about how many people in companies spend hours learning Excel. How many courses, trainings, certifications, how much time wasted debugging formulas and macros. And all you need to say is: “merge these two files by the ID column and show me the differences” – and it’s done. No formulas, no macros, no VBA, no knowledge of the interface required.
This isn’t a vision of the future. This is the present.
From Documents to Protocols
What I’ve described above is only the first act. The real change will be even deeper and more fundamental.
Act one: Office is beginning to morph into a prompt-driven tool. Microsoft is already doing this by introducing Copilot across all Office applications. But this is merely a transitional phase – something like a smartphone with a physical keyboard at the dawn of the iPhone era. Modern on the surface, but still desperately clinging to the old paradigm.
Act two: the document workflow itself will begin to systematically lose its relevance.
Why? Because documents are fundamentally information carriers designed for humans. A report in Word exists so that a human can read it. A spreadsheet in Excel exists so that a human can see data in tabular form. A PowerPoint presentation exists so that a human can view slides.
But if AI agents stand on both sides of a process – one creating information, the other consuming it and making decisions based on it – then why have a document at all? An agent doesn’t need pretty formatting in Word. It doesn’t need colorful charts in Excel. It doesn’t need transition animations in PowerPoint. It needs data in a structured format it can efficiently process.
And this is where agent communication protocols come in: APIs, MCP (Model Context Protocol), WebMPC, and other standards that the entire AI industry is actively developing. Instead of generating a Word report and sending it by email, an agent will simply expose data through an API. Instead of creating a quarterly summary presentation, an agent will pass structured data to another agent that will make an operational decision.
This is a world where documents in the traditional sense become artifacts of a bygone era. And if documents are unnecessary, then the tool for creating them – Office – is unnecessary too.
But People Will Always Want to Read Reports...
I can already hear the counterarguments. “People will always want to read reports.” “Presentations are needed for meetings.” “Excel is a standard that nobody will replace.”
My answer is an analogy: people also wanted a physical keyboard on their phones. Studies were conducted that confirmed this. BlackBerry was the king of the market. Then the iPhone appeared with its touchscreen, and it turned out people wanted something completely different – they just didn’t know it existed until someone showed them.
The same will happen with office documents. When AI agents start getting things done faster, cheaper, and better than traditional document workflows, people will stop missing Word. Just as they stopped missing physical keyboards on their phones.
We can already see this process in business process automation. ATS systems in recruitment are a perfect example – instead of manually reviewing CVs in Word, algorithms process application documents automatically. Element takes it even further, eliminating successive manual steps from the process. Every workflow that relies on creating, sending, and analyzing documents is a candidate for replacement by direct system-to-system communication.
What This Means in Practice
If you work at a company that relies heavily on the Office suite – and most companies do – it’s worth starting to think about how to prepare for the coming change.
First: stop investing in Excel training. Seriously. Invest that money and time in training people to use AI effectively. The ability to write a good prompt is worth more today than the ability to write a VBA macro. And this gap will only widen.
Second: start automating document workflows. Every process where someone creates a document in Office, emails it, waits for a response, opens it, edits it, and sends it back – that’s a process AI can speed up by an order of magnitude. And within a few years – eliminate entirely.
Third: watch the development of agent protocols. MCP, APIs, WebMPC – these are the technologies that will replace documents as information carriers between systems. The sooner your organization starts understanding and implementing them, the better prepared it will be.
In Closing
You can note all of the above and check in a few years how well it aged. I’m convinced this text will age well. Not because I’m clairvoyant, but because I can see the trends – and their direction is unambiguous.
MS Office was a revolutionary tool that changed how millions of people work. But every revolution eventually gets replaced by the next one. And this next revolution – the AI revolution – doesn’t need an interface with ribbons, toolbars, and dialog boxes.
It needs a single field to type a command. And soon, it won’t even need that.
Clicking is on its way out. Commands are taking over. And then – even commands will become unnecessary, because agents will know what to do.
DISCOVER ELEMENT!
Maciej Michalewski
CEO @ Element. Recruitment Automation Software
Recent posts:
I Don’t See a Future for MS Office
Three phases of transition from clicking buttons to AI commands. Why Microsoft Office in its current form is destined to disappear.
Postings up 11%, Dorsey cuts 40%. What’s going on with dev jobs?
Postings up 11%, Dorsey cuts 40%. What’s going on with dev jobs? The data that surprised me I’ve been writing about work automation for years.
Postings up 11%, Dorsey cuts 40%. What’s going on with dev jobs?
The data that surprised me I’ve been writing about work automation for years. I consider it inevitable and, in the long run, beneficial. So when
Automation: How Capitalism Is Leading Us to Socialism
Paradoxically, leaders of modern capitalism like Elon Musk are paving the way toward a socialist reality — through relentless automation that eliminates the need for human labor.
Keywords Cloud: The Best Way to Prepare Your CV for ATS
Learn how to use a keywords cloud in your CV to boost your chances of being found by recruiters using ATS systems.
Is Artificial Intelligence Truly Intelligent?
Can AI truly be called intelligent? We examine the nature of machine intelligence versus human cognition, from reasoning and knowledge retention to understanding context — with practical ChatGPT experiments.
