This looks bad for Microsoft. They added a Copilot button to all their products but it doesn't do much more than open a chat side panel.
I recently tried Claude Cowork for PowerPoint and I was stunned by the content as well as design quality of the deck it produced. That's a threat for Microsoft because now you don't need the editing tools of PowerPoint, AI replaces it, so all you need is the presentation mode of PowerPoint.
Copilot for Excel is useless. Ask it what is in cell A1 and it can't answer. I am looking forward to trying ChatGPT for Excel.
Claude for excel is already amazing. Fully capable of doing junior work. Formatting is great. Can refactor large multi-tab spreadsheets. It just burns tokens. If OpenAI is going to subsidize this on the monthly enterprise plans for a while then it's a game changer.
Claude for Excel (I work in finance) was one of the absolutely critical reasons we added Anthropic enterprise licenses. But they've turned out to be quite expensive ($100/day for heavy users). We'll see what OpenAI's quotas are.
Chatgpt for Excel is still an office add-in running in the same sandbox though. strongpigeon described the exact bottleneck upthread, process boundary crossings, context.sync() roundtrips that take seconds on web. That's a platform limitation, not a model limitation.
Swapping AI behind the add-in doesn't fix the fundamental constraint that third-party add-ins can't deeply integrate with Excel's runtime the way a native feature can. If copilot is bad despite having more access to excel internals(I don't like how Copilot is designed or implemented tho), an add-in with less access is likely not be better.
There is a significant difference in experience between Copilot Basic for a M365 user whose IT admins have blocked integration capabilities with Sharepoint content vs Copilot Premium for a M365 user whose IT admins have allowed integration capabilities with Sharepoint content.
We have many people in my wider team (Finance) that are AI skeptics purely because of their experience with Copilot. Like they don't know what AI is actually capable of when outside of the shackles of Copilot.
Oh wow, I used to work on Excel Add-Ins about 10 years ago. Even got a patent for it. I'd be curious to see how they implemented the calls.
We came up with what I still consider a pretty cool batch-rpc mechanism under the hood so that you wouldn't have to cross the process boundary on every OM calls (which is especially costly on Excel Web). I remember fighting so hard to have it be called `context.sync()` instead of `context.executeAsync()`...
That being said, done poorly it can be slow as the round-trip time on web can be on the order of seconds (at least back then).
Do you mean that you worked on the Excel Add-Ins platform in Excel (and not on a specific Add-In)?
If you were working on the platform itself, then I would be interested in hearing your more detailed thoughts on the matters you mentioned (especially since I am developing an open source Excel Add-In Webcellar (https://github.com/Acmeon/Webcellar)).
What do you mean with a "OM" call? And why are they especially costly on Excel web (currently my add-in is only developed for desktop Excel, but I might consider adding support for Excel web in the future)?
In any case, `context.sync()` is much better than `context.executeAsync()`.
I worked on the Excel Add-Ins platform at Microsoft, yes. By OM call I mean "Object Model" call, basically interacting with the Excel document.
The reason those calls are expensive on Excel Web is that you're running your add-in in the browser, so every `.sync()` call has to go all the way to the server and back in order to see any changes. If you're doing those calls in a loop, you're looking at 500ms to 2-3s latency for every call (that was back then, it might be better now). On the desktop app it's not as bad since the add-in and the Excel process are on the same machine so what you're paying is mostly serialization costs.
Happy to answer more questions, though I left MSFT in 2017 so some things might have changed since.
In principle, I find it valuable to integrate tools. However, in this case I would be somewhat cautious, especially as "your chats, attachments, and workbook content — may be shared with OpenAI" (as per the Microsoft Marketplace description: https://marketplace.microsoft.com/en-us/product/WA200010215?...).
This seems like a security nightmare, which is especially relevant because sensitive data is often stored in Excel files.
Even accidentally. So you want it for the spreadsheet you made to organize your giant collection of vintage disposable straw wrappers… ok sure. Whoops— I forgoti had installed it when I just opened up the test results for the propriety work thing, or, the spreadsheet I keep all my passwords in… (yes people do that.) Hopefully there’s at least a confirmation dialog?
That's the nature of these add-ins. Modern Add-ins are all little XML frames with some JS or whatever. All processing occurs server-side, hosted by the add-in publisher.
This is counter to the old (security nightmare) COM model where processing could be local.
I've experimented with ChatGPT for spreadsheets the past 6 months, and while the results look nice now it has been excruciatingly slow for even the simplest spreadsheet. I'm talking 15-20 minutes to make some pretty basic calculator with graphs. IIRC, it used a lot of time purely on the styling.
I asked GPT-5.4 High to draw up an architecture diagram in SVG and left it running. It took over an hour to generate something and had some spacing wrong, things overlapping, etc. I thought it was stuck, but it actually came back with the output.
Then I asked it to make it with HTML and CSS instead, and it made a better output in five seconds (no arrows/lines though).
SVG looks similar to the XML format of spreadsheets. I wonder if LLMs struggle with that?
The LLMs seem to struggle at anything that isn't relatively well anchored in whatever space. HTML documents have a lot of foundation to them in the training data, so they seem to perform well by comparison to other things.
I just spent a few hours trying to get GPT5.4 to write strict, git compatible patches and concluded this is a huge waste of time. It's a lot easier and more stable to do simple find/replace or overwrite the whole file each time. Same story in places like Unity or Blender. The ability to coordinate things in 3d is really bad still. You can get clean output using parametric scenes, but that's about it.
Claude's diagramming tool that they have built into their web UI is my goto for this task. It's reliable enough that I often will delegate to it first with what I need written in prose instead of using mermaid/lucid diagram
Microsoft has this built-in using Claude models (for M365 Copilot licensed users). I don't know why you'd use this as an M365 subscriber in an enterprise. I'm sure there's some edge cases, but MSFT has been moving away from OAI. Even Copilot Studio agents now default to Sonnet 4.6 and not GPT 5.
Based on my discussion with DSEs, enterprises have not been impressed in the results of "Copilot", i.e. OAI models. MSFT has been replacing (or changing the default) to Claude across a variety of Copilot endpoints.
Several months ago, ChatGPT swore to me it had interoperability with both excel and Google Sheets. I spent 90 minutes thinking I was an idiot, trying to follow its guidance before asking the internet.
It was partially a joke but someone posted a image of Co-pilot in Excel to demonstrate the limits of these things. Three cells with three numbers (1, 2, 3) and co-pilot asked to sum these three up.
Instead of answering with 6, it came up with 15. The comment was "If AI is doing this, a global financial crash is inevitable."
Might not be real but it is something to keep an eye on. Hopefully, they are a bit more cautious on how this is implemented.
I wonder why it's so bad. Do they just paste a CSV into the raw model? Because in my experience, even small local models can handle it reasonably well if the harness forces them to write & run a Python script that parses the table and performs the calculations, instead of relying solely on next-token prediction.
Your comment is recognized as low effort, but Copilot has been OAI models behind the scenes. For enterprise customers, quickly being replaced by Sonnet as a default.
As someone that knows a high-flying portfolio manager who works at a very well known firm that I wont name... I can confidently state these tools are DOA. Ive spoken to them at length about the nature of what these people actually do day-to-day. If you think its just about using excel then you're already way off.
They (OAI+Anthropic) very much do not get exactly what these people are doing in the job (accounting+corporate finance+valuation+asset management) and what the actual production process is. These tools are irrelevant, disrupt flow and if anything just add noise to what one is doing.
Remember when Steve said 'The computers for the rest of us'?
I suppose it isn't a surprise. Are researchers/generally geeky people meant to be able to relate to the average person's day-to-day beyond their sphere? Lmao.
You can't produce stuff for people you don't understand. Understand being a very key term.
This looks bad for Microsoft. They added a Copilot button to all their products but it doesn't do much more than open a chat side panel.
I recently tried Claude Cowork for PowerPoint and I was stunned by the content as well as design quality of the deck it produced. That's a threat for Microsoft because now you don't need the editing tools of PowerPoint, AI replaces it, so all you need is the presentation mode of PowerPoint.
Copilot for Excel is useless. Ask it what is in cell A1 and it can't answer. I am looking forward to trying ChatGPT for Excel.
Claude for excel is already amazing. Fully capable of doing junior work. Formatting is great. Can refactor large multi-tab spreadsheets. It just burns tokens. If OpenAI is going to subsidize this on the monthly enterprise plans for a while then it's a game changer.
Claude for Excel (I work in finance) was one of the absolutely critical reasons we added Anthropic enterprise licenses. But they've turned out to be quite expensive ($100/day for heavy users). We'll see what OpenAI's quotas are.
Chatgpt for Excel is still an office add-in running in the same sandbox though. strongpigeon described the exact bottleneck upthread, process boundary crossings, context.sync() roundtrips that take seconds on web. That's a platform limitation, not a model limitation. Swapping AI behind the add-in doesn't fix the fundamental constraint that third-party add-ins can't deeply integrate with Excel's runtime the way a native feature can. If copilot is bad despite having more access to excel internals(I don't like how Copilot is designed or implemented tho), an add-in with less access is likely not be better.
There is a significant difference in experience between Copilot Basic for a M365 user whose IT admins have blocked integration capabilities with Sharepoint content vs Copilot Premium for a M365 user whose IT admins have allowed integration capabilities with Sharepoint content.
Microsoft is better off not allowing copilot basic because of the reputational harm it will do. Not that they are thinking through copilot rationally
A recent funny story on this topic: https://idiallo.com/blog/what-is-copilot-exactly
HN discussion: https://news.ycombinator.com/item?id=47603231
If AI winning means that data center companies win out, then the wins for Azure will more than make up for the death of Office.
I am surprised that Microsoft's own copilot product is so far behind though.
We have many people in my wider team (Finance) that are AI skeptics purely because of their experience with Copilot. Like they don't know what AI is actually capable of when outside of the shackles of Copilot.
Microsoft fumbled so badly here.
its baffling how badly microsoft has handled copilot, this is exactly what copilot in office should have been
It's called Microslop for a reason.
You have to use the "agent" toggle for Copilot to behave the same way lol. Otherwise its pretty simple chat interface with the context, that's all.
Oh wow, I used to work on Excel Add-Ins about 10 years ago. Even got a patent for it. I'd be curious to see how they implemented the calls.
We came up with what I still consider a pretty cool batch-rpc mechanism under the hood so that you wouldn't have to cross the process boundary on every OM calls (which is especially costly on Excel Web). I remember fighting so hard to have it be called `context.sync()` instead of `context.executeAsync()`...
That being said, done poorly it can be slow as the round-trip time on web can be on the order of seconds (at least back then).
Do you mean that you worked on the Excel Add-Ins platform in Excel (and not on a specific Add-In)?
If you were working on the platform itself, then I would be interested in hearing your more detailed thoughts on the matters you mentioned (especially since I am developing an open source Excel Add-In Webcellar (https://github.com/Acmeon/Webcellar)).
What do you mean with a "OM" call? And why are they especially costly on Excel web (currently my add-in is only developed for desktop Excel, but I might consider adding support for Excel web in the future)?
In any case, `context.sync()` is much better than `context.executeAsync()`.
I worked on the Excel Add-Ins platform at Microsoft, yes. By OM call I mean "Object Model" call, basically interacting with the Excel document.
The reason those calls are expensive on Excel Web is that you're running your add-in in the browser, so every `.sync()` call has to go all the way to the server and back in order to see any changes. If you're doing those calls in a loop, you're looking at 500ms to 2-3s latency for every call (that was back then, it might be better now). On the desktop app it's not as bad since the add-in and the Excel process are on the same machine so what you're paying is mostly serialization costs.
Happy to answer more questions, though I left MSFT in 2017 so some things might have changed since.
In principle, I find it valuable to integrate tools. However, in this case I would be somewhat cautious, especially as "your chats, attachments, and workbook content — may be shared with OpenAI" (as per the Microsoft Marketplace description: https://marketplace.microsoft.com/en-us/product/WA200010215?...).
This seems like a security nightmare, which is especially relevant because sensitive data is often stored in Excel files.
Even accidentally. So you want it for the spreadsheet you made to organize your giant collection of vintage disposable straw wrappers… ok sure. Whoops— I forgoti had installed it when I just opened up the test results for the propriety work thing, or, the spreadsheet I keep all my passwords in… (yes people do that.) Hopefully there’s at least a confirmation dialog?
That's the nature of these add-ins. Modern Add-ins are all little XML frames with some JS or whatever. All processing occurs server-side, hosted by the add-in publisher.
This is counter to the old (security nightmare) COM model where processing could be local.
I've experimented with ChatGPT for spreadsheets the past 6 months, and while the results look nice now it has been excruciatingly slow for even the simplest spreadsheet. I'm talking 15-20 minutes to make some pretty basic calculator with graphs. IIRC, it used a lot of time purely on the styling.
Adding a tangential anecdote.
I asked GPT-5.4 High to draw up an architecture diagram in SVG and left it running. It took over an hour to generate something and had some spacing wrong, things overlapping, etc. I thought it was stuck, but it actually came back with the output.
Then I asked it to make it with HTML and CSS instead, and it made a better output in five seconds (no arrows/lines though).
SVG looks similar to the XML format of spreadsheets. I wonder if LLMs struggle with that?
The LLMs seem to struggle at anything that isn't relatively well anchored in whatever space. HTML documents have a lot of foundation to them in the training data, so they seem to perform well by comparison to other things.
I just spent a few hours trying to get GPT5.4 to write strict, git compatible patches and concluded this is a huge waste of time. It's a lot easier and more stable to do simple find/replace or overwrite the whole file each time. Same story in places like Unity or Blender. The ability to coordinate things in 3d is really bad still. You can get clean output using parametric scenes, but that's about it.
Claude's diagramming tool that they have built into their web UI is my goto for this task. It's reliable enough that I often will delegate to it first with what I need written in prose instead of using mermaid/lucid diagram
I’d try asking it for a mermaid diagram. I think ChatGPT’s web interface will render them.
Gemini is very good with SVG, but I don't really see the similarity to spreadsheets.
This is a drop-in database analysis tool and nobody knows it. Most Excel users are using Excel as a half-baked database instead of as a spreadsheet.
Microsoft has this built-in using Claude models (for M365 Copilot licensed users). I don't know why you'd use this as an M365 subscriber in an enterprise. I'm sure there's some edge cases, but MSFT has been moving away from OAI. Even Copilot Studio agents now default to Sonnet 4.6 and not GPT 5.
> I'm sure there's some edge cases, but MSFT has been moving away from OAI.
You're not wrong, but you'd think that given their 27% stake in OpenAI they'd put more weight behind ChatGPT integration.
Based on my discussion with DSEs, enterprises have not been impressed in the results of "Copilot", i.e. OAI models. MSFT has been replacing (or changing the default) to Claude across a variety of Copilot endpoints.
Several months ago, ChatGPT swore to me it had interoperability with both excel and Google Sheets. I spent 90 minutes thinking I was an idiot, trying to follow its guidance before asking the internet.
It was partially a joke but someone posted a image of Co-pilot in Excel to demonstrate the limits of these things. Three cells with three numbers (1, 2, 3) and co-pilot asked to sum these three up.
Instead of answering with 6, it came up with 15. The comment was "If AI is doing this, a global financial crash is inevitable."
Might not be real but it is something to keep an eye on. Hopefully, they are a bit more cautious on how this is implemented.
I wonder why it's so bad. Do they just paste a CSV into the raw model? Because in my experience, even small local models can handle it reasonably well if the harness forces them to write & run a Python script that parses the table and performs the calculations, instead of relying solely on next-token prediction.
Copilot is so bad that chatGPT is offered to replace it.
hmmYour comment is recognized as low effort, but Copilot has been OAI models behind the scenes. For enterprise customers, quickly being replaced by Sonnet as a default.
Copying Anthropic again lol.
Damn that OAI valuation is like a sore boil that is about to explode.
Also once again, a lack of imagination from OAI. Damn vision really is super scarce huh.
saltman look so desperate.
meanwhile not that ant is genius, except the timing of dow drama right before Iran war.
As someone that knows a high-flying portfolio manager who works at a very well known firm that I wont name... I can confidently state these tools are DOA. Ive spoken to them at length about the nature of what these people actually do day-to-day. If you think its just about using excel then you're already way off.
They (OAI+Anthropic) very much do not get exactly what these people are doing in the job (accounting+corporate finance+valuation+asset management) and what the actual production process is. These tools are irrelevant, disrupt flow and if anything just add noise to what one is doing.
Why are they irrelevant? You do not say anything.
I care not to. I hope Anthropic and OAI keep burning money on stuff that's DOA.
I know there are employees of those firms here that would love to know. But nah lmao.
I know the firm - it's RenTech.
nah the firm in question has much higher AUM.
This might be the first time I've seen a HN comment in a GPT thread that actually reflects what the average business user sees in GPT products.
They don't do the job, reliably or well. No amount of wishful thinking or extra tokens will change that.
No surprise really.
Remember when Steve said 'The computers for the rest of us'?
I suppose it isn't a surprise. Are researchers/generally geeky people meant to be able to relate to the average person's day-to-day beyond their sphere? Lmao.
You can't produce stuff for people you don't understand. Understand being a very key term.
Why though? What is the point of this? I thought they are building towards an AGI.
Next do one for PowerPoint and Outlook
Except for pro and plus users in the EU eh…