Copilot is the worst AI agent on the market. Over 50% of people I've spoken to that say AI is overhyped, when pressed, admit they were only using Copilot.
This would be profitable if they could ship garbage for cheap, a la Microsoft Teams or Internet Explorer. But Copilot is worse at integrating with Office than Claude!
This is because Copilot has aggressive context pruning to meet its price point of $20/month. That prevents the AI from meaningfully using tools or being multimodal or anything else their competitors have.
If they added a $200/month tier many of their issues would go away.
Their target is not coders, it is the professional world who do 90% of their work in Office applications, like me. A $200/m model absolutely does not fly when rolled out to entire corporations. It needs to be a $20/user/month product.
But I agree, it sucks. It is the only AI we are able to use at work and for tasks that it should be good at (compare comment sheets against a deliverable register and assign to specific packages) and it just can’t do it. It can read the spreadsheet and understand them just fine but outputs are garbled nonsense.
“Copilot” is not one product, it’s around 15 different products, seriously.
I think that people often compare apples to oranges by comparing the “copilot” they have in Windows/Office/Teams etc to Claude Code which is ridiculous.
A better product to compare Claude Code to would be “Github Copilot CLI”, but I haven’t seen the two seriously compared anywhere.
In the context of knowledge workers, It is really about Claude Cowork against Microsoft Copilot suite for all their applications, which is what the OP is referencing ?
Github Copilot can use Claude APIs and has its own problems and challenges.
Microsoft AI performance is primarily not being affected by Github - while significant is much much smaller part of the enterprise revenue stream and their DAU compared to their Office suite apps.
Same for their PR exposure. It is lot more likely to here about Copilot in the office context than Github outside of small niche's like this forum.
Absolutely. They've made the same mess with "Copilot" as with ".NET" in the early 00's [0]. Everything was ".NET" from consumer oriented services (".NET Passport"), to "Visual Studio .NET" without anyone understanding what ".NET" was.
Now it's "Microsoft Copilot" which is different from "Microsoft 365 Copilot", which is different from "Copilot Chat" and from "GitHub Copilot", and the many other flavors.
It's a mess.
Still, their developer-focused offering seems to be "GitHub Copilot", which among other things includes "GitHub Copilot CLI" [1], their terminal-based agent. It's not bad.
We have copilot at work integrated very deeply in our E5 landscape. It def sucks in Office, and I can break Copilot very easily when building small notebooks. It often crashes and the next page build in a notebook doesn’t come close to the previous iteration. That’s maddening.
But, the Teams integration for meeting summaries and in-meeting “what did Bob just say about the data center project?” prompts is magical and very useful. If you live in meetings or are trying not to. They need to put that team on rescue duty.
I have no interest in Copilot from Microsoft in general but I do like GitHub Copilot overall. That said, I’m _very_ interested in a viable alternative. I only use it for “fancy autocomplete” and have zero use for the agent/chat capabilities (I use Claude Code for that). It’s been a year or so since I looked at and tried alternatives but when I last did, copilot was the best IMHO.
It’s odd to me how much Microsoft over committed to Copilot. They added unwanted Copilot buttons to laptops. They renamed Office to Copilot. And in the end, it’s a terrible product anyways. They can only get away with this because of the control they have over OEMs, distribution channels, and the inability for consumers to opt out of all of thus.
Meanwhile startups can’t compete fairly because they don’t have the same channels to flood with their own branding.
Copilot was by far the worst for coding. Not that the code snippets it would generate were not good, but due to the insane number of bugs in its UI. It would just spit out blank blocks thinking they contained code. When I asked it to repeat the steps which were empty, it would generate the same empty blocks like "here's your code" lol
Kind of ongoing theme in tech industry and beyond. "Investors" will punish every stock unless company announces mass layoffs. My company is posting 10% year over year growth yet stocks keep falling.
If one were to think a major stock market crash was coming up, led by the AI bubble bursting, but reinforced by the major self-own that is the Iran war, how would one best prepare ones investments?
Either you need the cash now and are already in short-term treasuries or you're it it for the long-term and you'll be laughing at this question when the Dow hits 100k.
One should either weather out the storm or if one wants to cash out soon or manage their portfolio more closely they would pick the defensive assets they trust the most and hold until they stop thinking the stock market crash is coming up or stop trusting those assets. If they really think the crash is imminent, maybe investing some excess money into shorting the market while setting trailing stop loss would be a fun activity that might turn profitable
The advice I have heard is if you think there will be a significant drop in the market you liquidate all your holdings while they are still high and then rebuy when the price is low. Granted this is a gamble though, if you’re wrong then you just sold all your stock and are no longer participating in the market plus you need to pay capital gains tax
If you genuinely think these two events will shake the foundations of the US economy, then no investment is really going to protect you; you can't expect a system to shield you from the very thing that's breaking it.
There is crypto but even that got infiltrated by institutional wall street money. There are off-shore jurisdictions but the recent Iran war has showed these can be very vulnerable at a moment notice. There is China, but a Taiwan invasion could reduce your assets there to zero.
Honestly, I think the best bet is crypto/Bitcoin, by far. It operates across borders and still relatively insulated from government reach. Unlike gold, oil, or anything physical, it can be moved without physical visibility.
I don' think that logic is intended for the top... It is what you should do when you are closing to bottom or are recovering already but most of the market does not yet see it.
Being greedy at the top will take longest time to recover. Catching the falling knife.
Not investing advice, I’ve reallocated away from US domestic equities to international equities (VXUS) as a majority of a portfolio. This hedges against a correction from overweight Mag 7 exposure and US economic growth impairment from current policies (imho).
You can go all cash so not risking leverage via shorts, etc. But a lot of folks think that's dumb. I feel better being in cash right now as the mental stress of big loses in middle age is not worth the missed gains if I'm wrong. To each their own.
If your debts are also denominated in USD, their value will be fixed relative to your cash assets. This assumes a fixed rate, of course, but a 30 year fixed is common in the US and makes up a substantial portion of most folks’ debt.
What percent of dollars are tied up in in-flight oil transactions? And I suppose also in accounts that will be used for oil transactions in the planned future? That’s the mechanism for that supporting the value of the dollar, right, like, increased dollar demand via being used for oil market transactions?
The point is that the Petrodollar system requires countries to buy US treasuries to be able to buy oil. That is what makes borrowing cheap for the US, and what keeps the dollar demand up.
I'm not clear on how the Petrodollar system actually works.
If oil is sold in dollars, that only has to affect dollar demand for the time it takes to transit through some other currency to dollars to oil. So however long that takes to settle.
Where do the additional demand components come from? Why do countries have to buy US treasuries to be able to buy oil? Don't they just have to use dollars for the transaction?
Microsoft is one of the least likely large companies to benefit from an AI boom. They don’t have the capacity to support OpenAI and their own foundational models, they aren’t providing a compelling story for wrapping OpenAI, windows continues to suck…
OpenAI signed an agreement with GCP , that should say a lot.
OpenAI also signed an agreement with AWS. And Anthropic has signed on with Microsoft as well as GCP.
The underlying dynamic is that no single cloud provider has the capacity required to host all the demand, so the frontier AI labs have no choice but to diversify for their infrastructure needs.
can you elaborate on this? diversifying compute doesn't create more compute - is it that the different LLM vendors have different peak times and so spreading themselves over more compute vendors spreads peak load?
Yes but if there are three banana shops around and there are five banana addicted people living nearby the number of bananas available on average for every person is not 15.
In other words, if all ai companies need more compute that a single provider can provide, then there's just not enough of it. So the question "why everyone partners with everyone" must have a different answer.
It's not really "creating more compute" it's just a natural outcome of everyone desperately grabbing whatever becomes available. The dynamics make sense for all parties involved.
Firstly, it's very clear now that everyone is seriously crunched for capacity (like, each of the hyperscalers' backlogs -- i.e. capacity for which payment is committed, but as yet unsatisfied -- are in the double-digit billions.)
So as the compute providers bring more capacity online, everyone with demand wants to get a slice of that. Like, why would anyone NOT dive in and try to secure some capacity for themselves? Especially when the rate of capacity growth is constrained by the availability of GPUs and energy and data center buildouts, which is measured in years.
On the flip side, why would the compute providers NOT want multiple customers? It creates competition and drives prices up.
There are likely other forces at play too. For one, none of the parties - the model providers and the compute providers, with some of them like Google being both -- wants to get too dependent on any of the other parties, but they also want to secure a slice of each others' future growth, so they're all partnering with each other. Obviously, Google wants Gemini to win and Microsoft wants Copilot to win, but as a hedge, they'll be happy hosting their competitors' products and taking a cut.
This is partly the origin of the "circular investments" concerns. The scale at which this industry is growing, all these players have enormous mountains of money that they must invest to secure their future, but they are also the only players that can operate at this scale, and so the only place they can invest that money in is each other.
Once we are able to run language models on any consumer hardware with good T/s Windows will become an absolute powerhouse just like it did in gaming with DirectX. Any application will be able to be AI infused and the API to do so will be consistent and free to the business offering it.
spacex is particular is desperately searching for a bagholder. Leveraging the obvious synergies between rockets (a mediocre business), xai (a horrid business), and social media that lost more than half it's already modest revenues...
SpaceX has reasonable business in it. Not a hypergrowth one, but one which should be solid in long term.
To get that to work they just would need to discard Musk and most things with him. Stop trying to make starship a thing, dump everything attached to it. Make a long term plan to improve the core lift capacity with actually achievable improvements.
> To get that to work they just would need to discard Musk and most things with him
I'll have a pet unicorn shitting rainbows before Musk leaves one of his toys or we see in-orbit leading-edge (or anything close to it) processor production. SpaceX is a decent albeit capital intensive business if it's valued at $100-200B. At the proposed $1.5T+ valuation for this dog... the bagholder search is on.
Copilot is the worst AI agent on the market. Over 50% of people I've spoken to that say AI is overhyped, when pressed, admit they were only using Copilot.
This would be profitable if they could ship garbage for cheap, a la Microsoft Teams or Internet Explorer. But Copilot is worse at integrating with Office than Claude!
This is because Copilot has aggressive context pruning to meet its price point of $20/month. That prevents the AI from meaningfully using tools or being multimodal or anything else their competitors have.
If they added a $200/month tier many of their issues would go away.
Their target is not coders, it is the professional world who do 90% of their work in Office applications, like me. A $200/m model absolutely does not fly when rolled out to entire corporations. It needs to be a $20/user/month product.
But I agree, it sucks. It is the only AI we are able to use at work and for tasks that it should be good at (compare comment sheets against a deliverable register and assign to specific packages) and it just can’t do it. It can read the spreadsheet and understand them just fine but outputs are garbled nonsense.
Copilot is actually significantly more reliable at technical tasks with SQL or C# than others, i've found. Do we have different use cases?
Copilot seems to hit the technical level I'm asking about much more reliably. It keeps a more grounded general semantic model.
You might be using different copilots since there are approximately three different Microsoft copilot products
Are you using the one that's part of Microsoft 365?
“Copilot” is not one product, it’s around 15 different products, seriously.
I think that people often compare apples to oranges by comparing the “copilot” they have in Windows/Office/Teams etc to Claude Code which is ridiculous.
A better product to compare Claude Code to would be “Github Copilot CLI”, but I haven’t seen the two seriously compared anywhere.
In the context of knowledge workers, It is really about Claude Cowork against Microsoft Copilot suite for all their applications, which is what the OP is referencing ?
Github Copilot can use Claude APIs and has its own problems and challenges.
Microsoft AI performance is primarily not being affected by Github - while significant is much much smaller part of the enterprise revenue stream and their DAU compared to their Office suite apps.
Same for their PR exposure. It is lot more likely to here about Copilot in the office context than Github outside of small niche's like this forum.
That's on Microsoft for their choice of naming/branding.
Absolutely. They've made the same mess with "Copilot" as with ".NET" in the early 00's [0]. Everything was ".NET" from consumer oriented services (".NET Passport"), to "Visual Studio .NET" without anyone understanding what ".NET" was.
Now it's "Microsoft Copilot" which is different from "Microsoft 365 Copilot", which is different from "Copilot Chat" and from "GitHub Copilot", and the many other flavors.
It's a mess.
Still, their developer-focused offering seems to be "GitHub Copilot", which among other things includes "GitHub Copilot CLI" [1], their terminal-based agent. It's not bad.
[0]: https://en.wikipedia.org/wiki/Microsoft_.NET_strategy
[1]: https://github.com/features/copilot/cli/
I've been a C# developer on and off since the dotnet core/standard split. I still have no clue what .NET was ever supposed to mean.
We have copilot at work integrated very deeply in our E5 landscape. It def sucks in Office, and I can break Copilot very easily when building small notebooks. It often crashes and the next page build in a notebook doesn’t come close to the previous iteration. That’s maddening.
But, the Teams integration for meeting summaries and in-meeting “what did Bob just say about the data center project?” prompts is magical and very useful. If you live in meetings or are trying not to. They need to put that team on rescue duty.
I have no interest in Copilot from Microsoft in general but I do like GitHub Copilot overall. That said, I’m _very_ interested in a viable alternative. I only use it for “fancy autocomplete” and have zero use for the agent/chat capabilities (I use Claude Code for that). It’s been a year or so since I looked at and tried alternatives but when I last did, copilot was the best IMHO.
What are people using as an alternative?
It’s odd to me how much Microsoft over committed to Copilot. They added unwanted Copilot buttons to laptops. They renamed Office to Copilot. And in the end, it’s a terrible product anyways. They can only get away with this because of the control they have over OEMs, distribution channels, and the inability for consumers to opt out of all of thus.
Meanwhile startups can’t compete fairly because they don’t have the same channels to flood with their own branding.
Copilot was by far the worst for coding. Not that the code snippets it would generate were not good, but due to the insane number of bugs in its UI. It would just spit out blank blocks thinking they contained code. When I asked it to repeat the steps which were empty, it would generate the same empty blocks like "here's your code" lol
How TF can you go to market with such bugs.
i can't even upload co e properly because it gets mangled
In stock price, no the earnings
Kind of ongoing theme in tech industry and beyond. "Investors" will punish every stock unless company announces mass layoffs. My company is posting 10% year over year growth yet stocks keep falling.
If one were to think a major stock market crash was coming up, led by the AI bubble bursting, but reinforced by the major self-own that is the Iran war, how would one best prepare ones investments?
Either you need the cash now and are already in short-term treasuries or you're it it for the long-term and you'll be laughing at this question when the Dow hits 100k.
One should either weather out the storm or if one wants to cash out soon or manage their portfolio more closely they would pick the defensive assets they trust the most and hold until they stop thinking the stock market crash is coming up or stop trusting those assets. If they really think the crash is imminent, maybe investing some excess money into shorting the market while setting trailing stop loss would be a fun activity that might turn profitable
The advice I have heard is if you think there will be a significant drop in the market you liquidate all your holdings while they are still high and then rebuy when the price is low. Granted this is a gamble though, if you’re wrong then you just sold all your stock and are no longer participating in the market plus you need to pay capital gains tax
Did you hear this advice from 100 broke people or one lucky schmuck?
This is significantly safer than shorting that some people here suggest.
You can't lose money sitting on cash. While when shorting your potential loss is infinite.
Mistiming the market is losing out on potential gains, which is losing money compared to sitting on cash. Cash doesn't grow.
Most people are best off investing in index funds and forgetting about it for 10+ years.
If you genuinely think these two events will shake the foundations of the US economy, then no investment is really going to protect you; you can't expect a system to shield you from the very thing that's breaking it.
There is crypto but even that got infiltrated by institutional wall street money. There are off-shore jurisdictions but the recent Iran war has showed these can be very vulnerable at a moment notice. There is China, but a Taiwan invasion could reduce your assets there to zero.
Honestly, I think the best bet is crypto/Bitcoin, by far. It operates across borders and still relatively insulated from government reach. Unlike gold, oil, or anything physical, it can be moved without physical visibility.
Dunno, but yesterday was the first time ever I felt confident shorting nearly across the board. Nearly.
If everyone (most people) think the same, shouldn't you do the opposite?
Be fearful when others are greedy, and greedy when others are fearful. Etc.
I don' think that logic is intended for the top... It is what you should do when you are closing to bottom or are recovering already but most of the market does not yet see it.
Being greedy at the top will take longest time to recover. Catching the falling knife.
That might be the case if the market was completely abstract and removed from ground truths.
My feelings about these things don't come from markets.
I think a crash is coming and I do nothing. I rebalance my stock/bond split and keep a large emergency fund as per usual.
https://www.lynalden.com/march-2026-newsletter/ (control-f “The Investment Implications of Chaos”)
Not investing advice, I’ve reallocated away from US domestic equities to international equities (VXUS) as a majority of a portfolio. This hedges against a correction from overweight Mag 7 exposure and US economic growth impairment from current policies (imho).
https://www.axios.com/2026/03/27/stocks-trump-iran-nasdaq
https://totalrealreturns.com/n/VTI,VXUS?start=2025-01-20
https://www.apolloacademy.com/sp-500-concentration-approachi...
Other than buying oil futures, probably selling to cash and getting ready to buy when it hits the fan.
You can go all cash so not risking leverage via shorts, etc. But a lot of folks think that's dumb. I feel better being in cash right now as the mental stress of big loses in middle age is not worth the missed gains if I'm wrong. To each their own.
You don’t think the US dollar will take a dive along with US stocks?
If your debts are also denominated in USD, their value will be fixed relative to your cash assets. This assumes a fixed rate, of course, but a 30 year fixed is common in the US and makes up a substantial portion of most folks’ debt.
I have a big chunk in FXE (like owning Euros) precisely for this reason.
What specific thing(s) are you worried that USD will take a dive relative to?
Then once you have an answer to that question, that might point you towards what you want to be long.
Traders could start buying their oil using Yuan, for example. That’s not a theoretical anymore
What percent of dollars are tied up in in-flight oil transactions? And I suppose also in accounts that will be used for oil transactions in the planned future? That’s the mechanism for that supporting the value of the dollar, right, like, increased dollar demand via being used for oil market transactions?
The point is that the Petrodollar system requires countries to buy US treasuries to be able to buy oil. That is what makes borrowing cheap for the US, and what keeps the dollar demand up.
I'm not clear on how the Petrodollar system actually works.
If oil is sold in dollars, that only has to affect dollar demand for the time it takes to transit through some other currency to dollars to oil. So however long that takes to settle.
Where do the additional demand components come from? Why do countries have to buy US treasuries to be able to buy oil? Don't they just have to use dollars for the transaction?
How would buying Euros compare in terms of exposure?
The USD could take a dive against: yuan, eu, gbp, rial, gold, silver, platinum, WTI, SPY, etc.
Only a few of them will matter on a day-to-day basis if you're currently in the US with assets valued in USD.
Couldn't you just exchange your EUR for USD as needed? Use it as your reserve currency?
Microsoft is one of the least likely large companies to benefit from an AI boom. They don’t have the capacity to support OpenAI and their own foundational models, they aren’t providing a compelling story for wrapping OpenAI, windows continues to suck…
OpenAI signed an agreement with GCP , that should say a lot.
OpenAI also signed an agreement with AWS. And Anthropic has signed on with Microsoft as well as GCP.
The underlying dynamic is that no single cloud provider has the capacity required to host all the demand, so the frontier AI labs have no choice but to diversify for their infrastructure needs.
can you elaborate on this? diversifying compute doesn't create more compute - is it that the different LLM vendors have different peak times and so spreading themselves over more compute vendors spreads peak load?
Huh? If I need 10 bananas and my local shop only has 5 bananas available I need to go to multiple stores to satisfy my ravenous banana craving.
Yes but if there are three banana shops around and there are five banana addicted people living nearby the number of bananas available on average for every person is not 15.
In other words, if all ai companies need more compute that a single provider can provide, then there's just not enough of it. So the question "why everyone partners with everyone" must have a different answer.
It's not really "creating more compute" it's just a natural outcome of everyone desperately grabbing whatever becomes available. The dynamics make sense for all parties involved.
Firstly, it's very clear now that everyone is seriously crunched for capacity (like, each of the hyperscalers' backlogs -- i.e. capacity for which payment is committed, but as yet unsatisfied -- are in the double-digit billions.)
So as the compute providers bring more capacity online, everyone with demand wants to get a slice of that. Like, why would anyone NOT dive in and try to secure some capacity for themselves? Especially when the rate of capacity growth is constrained by the availability of GPUs and energy and data center buildouts, which is measured in years.
On the flip side, why would the compute providers NOT want multiple customers? It creates competition and drives prices up.
There are likely other forces at play too. For one, none of the parties - the model providers and the compute providers, with some of them like Google being both -- wants to get too dependent on any of the other parties, but they also want to secure a slice of each others' future growth, so they're all partnering with each other. Obviously, Google wants Gemini to win and Microsoft wants Copilot to win, but as a hedge, they'll be happy hosting their competitors' products and taking a cut.
This is partly the origin of the "circular investments" concerns. The scale at which this industry is growing, all these players have enormous mountains of money that they must invest to secure their future, but they are also the only players that can operate at this scale, and so the only place they can invest that money in is each other.
Once we are able to run language models on any consumer hardware with good T/s Windows will become an absolute powerhouse just like it did in gaming with DirectX. Any application will be able to be AI infused and the API to do so will be consistent and free to the business offering it.
Which is why they now are finally listing to customers.
This is a leading indicator for what is to come after the IPOs of SpaceX, OpenAI and Anthropic this year.
and it is not good.
With any luck they've missed their exit dumping window.
spacex is particular is desperately searching for a bagholder. Leveraging the obvious synergies between rockets (a mediocre business), xai (a horrid business), and social media that lost more than half it's already modest revenues...
SpaceX has reasonable business in it. Not a hypergrowth one, but one which should be solid in long term.
To get that to work they just would need to discard Musk and most things with him. Stop trying to make starship a thing, dump everything attached to it. Make a long term plan to improve the core lift capacity with actually achievable improvements.
> To get that to work they just would need to discard Musk and most things with him
I'll have a pet unicorn shitting rainbows before Musk leaves one of his toys or we see in-orbit leading-edge (or anything close to it) processor production. SpaceX is a decent albeit capital intensive business if it's valued at $100-200B. At the proposed $1.5T+ valuation for this dog... the bagholder search is on.