GPT 6.2 Astra beats Sol and Opus 5.5 in my latest Cowork evaluation

Three marbled-ink circular swirls in indigo, vermilion and gold on cream, in the suminagashi style

On 13 November I called a winner in Copilot Cowork: Opus 5.5 edges Fable 5.2. That head-to-head was Opus 5.5 at 8.9 against Fable 5.2 at 8.6. I said then that I had not re-run Astra, so it was not a new overall series winner, and that I would test the next model when it turned up.

The next round is in. I have now put two new GPT 6.2 reports through the same kind of evaluation, alongside the Opus 5.5 report from September. The full evaluation is here on GitHub. The winner is not Opus 5.5 this time.

What was compared

The evaluation assessed three reports, all responding to the same underlying prompt: is the traditional MSP and TSP business model becoming obsolete between 2029 and 2036? The three documents are:

  • Document A: the GPT 6.2 Astra report from Cowork, dated 18 November 2026, written as an Australian owner and investor decision report for CIAOPS.
  • Document B: the Opus 5.5 report from Cowork, dated 23 September 2026, a strategic research report for MSP owners, Microsoft partners, boards and private equity investors.
  • Document C: the GPT 6.2 Sol report from Cowork, dated 18 November 2026, a board and investment briefing for CIAOPS.

Scoring uses four weighted criteria: Argument 35%, Evidence 25%, Detail 20% and Presentation 20%. The evaluation is relative and substance-focused.

The result

RankDocumentOverallArgumentEvidenceDetailPresentation
1A: Astra / GPT 6.28.939.29.08.58.8
2C: Sol / GPT 6.28.789.08.98.78.5
3B: Opus 5.58.698.58.79.28.5

Document A, the GPT 6.2 Astra report, comes first. The evaluation calls it the tightest on epistemic discipline, with a fully reconciled decision model and the most practical capital gates for an owner or investor audience. The Sol report is a very close second. Opus 5.5 is third, but it still wins on Detail, 9.2 against 8.5 and 8.7, because it has the broadest coverage of market, security and history.

The evaluation says A and C are both clearly ahead of B on the two heaviest criteria, Argument and Evidence. Its stated reason is that Opus 5.5 was more assertive on magnitude and probability claims than its evidence strictly supports, including the 30 to 60 per cent labour-removal figure. It also says the earlier research cut-off meant Opus 5.5 missed some later November Microsoft detail that A and C used.

Where Astra and Sol differ

The gap between first and second is small. According to the evaluation, A is tighter on capital gates, inference language and the executive reading path. C is stronger on cash-trough quantification, pricing-headroom sensitivity and residual multi-vendor service, and uses a larger reference firm. The file notes that C’s title says the traditional model is becoming obsolete, while its body concludes it remains viable. It lists that as the only notable polish issue, and says C would pass A with a title that matches the body and tighter tranche limits.

How this compares with 13 November

In Opus 5.5 edges Fable 5.2, Opus 5.5 beat Fable 5.2 on evidence (9.2 against 8.3) and detail (9.3 against 8.4), while Fable 5.2 was the more polished read. The same pattern shows up again here. Opus 5.5 is still the most detailed report and still loses points for how tightly it handles claims. What has changed is the competition: two GPT 6.2 reports, Astra and Sol, are now ahead of it on overall score.

A few cautions. This is a different evaluation from the one in the 13 November post, so the Opus 5.5 score of 8.69 here should not be set against the 8.9 there as if it were a drop. Fable 5.2 was not part of this evaluation, so I cannot tell you where it would land among these three. The file does not include credit costs, so there is no price comparison here. And this is one prompt on one topic, so it is a data point, not a verdict on every task.

I have not checked which of these models are available, or in which Microsoft 365 Copilot surfaces, for your tenant. This post says nothing about availability. Check Microsoft’s official documentation for that.

What it means for choosing models in Copilot

For an MSP, I think the useful lesson is about what you are choosing for. The evaluation rewarded the reports that were careful about what they could and could not claim, and that turned analysis into reversible, bounded decisions. That is the same quality I want in a client-facing report. A longer, broader report lost to tighter ones on the criteria that carried the most weight.

So I would not pick a model on the scoreboard alone. Run your own representative prompt, such as a client QBR, a security review summary or a pricing paper. Then judge the outputs on argument and evidence first and detail second. If you need breadth, Opus 5.5 still earned the best Detail score here. If you need something a board or an owner can act on, the GPT 6.2 reports did better in this test.

Model names matter here: Astra and Sol are both GPT 6.2, and Opus 5.5 is a different model again. Keep a record of exactly which model produced each output when you compare, because they change quickly and the winner changes with them.

As soon as the next model shows up in Cowork, I will run it through the same test.

We have a new LLM in Copilot winner: Opus 5.5 edges Fable 5.2

Two colleagues comparing printed reports at a desk at dusk, teal and amber light on blank pages

A while back I compared the different LLMs in Copilot, and Fable 5.1 took the honours in Cowork:

https://blog.ciaops.com/2026/09/02/we-have-a-new-llm-in-copilot-winner/

GPT 6 Astra later took the trophy, and I then ran Opus 5.5 against it:

https://blog.ciaops.com/2026/09/17/astra-takes-the-trophy/

https://blog.ciaops.com/2026/09/23/i-think-we-need-a-count-back/

I have now put the latest LLM in Copilot through the same prompt in Cowork, judged again by Grok. First I compared Fable 5.2 with the September Fable 5.1 report. The results are here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20261113-Cowork-Fable-51-52-Grok-eval.md

and Fable 5.2 (Cowork) is ahead, 8.7 to 8.1. It was stronger on argument, evidence and presentation. Depth was roughly even.

I then put that November Fable 5.2 report against Opus 5.5. That evaluation is here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20261113-Cowork-Fable52-Opus55-Grok-eval.md

and the winner of that head-to-head is now:

Opus 5.5 (Cowork)

The scores are close: 8.9 against 8.6. Argument was a tie at 9.0 each. Opus 5.5 was stronger on evidence (9.2 vs 8.3) and detail (9.3 vs 8.4). Fable 5.2 was the more polished, executive-friendly read (9.0 vs 8.0). Both reports reach much the same strategic conclusions. The gap is in how tightly the evidence is handled, not in a different industry story. Opus 5.5’s research cut-off was earlier (23 September against 11 November), so Fable 5.2 had some later data and still came second.

You can see the Fable 5.2 output here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260816/20261113-Cowork-Fable52.docx

and the Opus 5.5 output here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260816/20260923-Cowork-Opus55.docx

along with the other outputs:

https://github.com/directorcia/general/tree/master/Copilot/Comparisons/20260816

These two evaluations don’t include credit costs, so I am not going to invent a new price list. They also don’t re-run Astra, so this is not a new overall series winner. For this latest LLM in Copilot pairing, though, Opus 5.5 edges Fable 5.2, after Fable 5.2 had already improved on Fable 5.1.

As soon as the next model turns up in Cowork, I’ll test it and update this information.

GPT 6.1 Sol Copilot battle

If you have been following along you know that I have been comparing LLMs inside the different services in Copilot. Unsurprisingly, the most current and powerful models tend to be the winner, but what about the same LLM between different services?

Screenshot 2026-10-10 080747

Recently, GPT 6.1 Sol appeared in my Copilot Chat as you can see above. Thus, I put it head to head with GPT 6.1 Sol in Cowork and the results are here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20261009-GPT61-Grok-Eval.md

The news is that Copilot Chat won!

Now, I think the real take away here is no so much that Copilot Chat is ‘better’ than Cowork, it is that the same LLM in any Copilot service is equivalent. That leads to the following rule of thumb:

‘You DON’T need to use Cowork for everything! Copilot Chat is the best place to start’

For validation I repeated the test but with Opus 5.5 in Chat and Cowork here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20261009-Opus55-Grok-Eval.md

and again Chat came out on top. So, I am pretty confident that my rule of thumb above is valid.

The big different is that Cowork require PAYG billing. Thus, for these tests:

Cowork GPT Sol 6.1 = US$101

Cowork Claude Opus 5.5 = US$234

Not cheap at all! But if you use these same LLMs, in Copilot Chat instead you pay NO MORE than the flat monthly M365 Copilot license.

That’s the real message here I feel, start with the included services BEFORE moving to the paid option because from what I see they produce the same result.

I think we need a count back

image

I have been comparing all the LLMs as they become available in Copilot inside the different services in Microsoft 365. The previous winner was GPT 6 Astra, with the details here:

https://blog.ciaops.com/2026/09/17/astra-takes-the-trophy/

As is the world AI, it isn’t long before another new model becomes available, this time Opus 5.5, so I put it through the same test as all the other models and it produced this out which you can download yourself:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260816/20260923-Cowork-Opus55.docx

As always I pitted the newcomer against the incumbent with the analysis report here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260923-Cowork-Opus55-Grok-eval.md

and is where the controversy is going to start. According to the ‘standard’ overall rating Astra scored 8.65 overall beating the newcomer Opus 5.5 with a score of 8.40. Close. However, if you dig a little further you see that Opus 5.5 score all 9’s for

– Evidence

– Detail

– Presentation

and was let down by

– argument

The overall score a weighted average in favour of argument. This helped Astra to pip Opus 5.5, but honestly I would suggest on review that Opus 5.5 is pretty much the equal of Astra 6 but a win is a win.

Ok, next point of amazement is that cost of Opus 5.5

Opus 5.5 = 23,494 credits

Astra 6 = 6,067 credits

Fable 5.1 = 3,453 credits

Fable 5 (Preview) = 3,387 credits [Model no longer shown]
Fable 5 (Copilot)(Preview) = 3,373.8 credits [Model no longer shown]

GPT 5.6 Sol = 2,800.50 credits
Sonnet 5 = 2,509.3 credits
Opus 4.8 = 2,487 credits [Model no longer shown]
Opus 5 = 2,240 credits

GPT 5.5 = 1,200 credits
GPT 5.6 Terra = 260 credits

So, the same prompt with Opus 5.5 cost a whopping US$235! That is roughly 5 x the cost of Astra 6 and almost 8 x the price of Fable 5.1, even though Opus 5.5 is supposed to be more ‘cost effective’ than Fable 5.1 according to Claude.

I’ll have to run a report and compare Opus 5.5 to Fable 5.1 and see what differences are evident but at 8 x the price they’d wanna be MASSIVE!

I have also updated the summary report for all the documents created by the different models here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260816-File-paramters.md

GPT 6 Sol has also just been made available in Copilot so I’ll be testing that next.

We have a new LLM in Copilot winner

image

A while back I compared all the different LLM’s (i.e. GPT, Claude) in various Copilot services (i.e. Chat, Cowork) here:

https://blog.ciaops.com/2026/08/17/comparing-llms-in-copilot-services-round-7-final/

The ultimate winner was:

Opus 5 (Cowork)

which cost US$22 to complete the work.

Microsoft has just announced that Fable 5.1 is available in Cowork so I took it for a spin through the same prompt I used in the recent testing. The results are here (as judged by Grok):

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260902-Cowork-Fable51-Grok-eval.md

and the new winner is now:

Fable 5.1 (Cowork)

you can see the result for yourself here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260816/20260902-Cowork-Fable-51.docx

along with all the others outputs:

https://github.com/directorcia/general/tree/master/Copilot/Comparisons/20260816

Here is an update on the costs in Cowork for each model:

Fable 5.1 = 3,453 credits

Fable 5 (Preview) = 3,387 credits [Model no longer shown]
Fable 5 (Copilot)(Preview) = 3,373.8 credits [Model no longer shown]

GPT 5.6 Sol = 2,800.50 credits
Sonnet 5 = 2,509.3 credits
Opus 4.8 = 2,487 credits [Model no longer shown]
Opus 5 = 2,240 credits

GPT 5.5 = 1,200 credits
GPT 5.6 Terra = 260 credits

In summary then, the new best model, Fable 5.1, produces the best results from my standard prompt and is also the most expensive to use using that prompt.

The comparison report for all document paramteres has been updated here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260816-File-paramters.md

The next updated model that we expect to see will probably be GPT 6.0 which is expected in the October timeframe if not sooner. As soon as it becomes available I’ll test it and update this information.

Comparing LLMs in Copilot services–Round 3 – Cowork (Claude)

image

Round 1 – https://blog.ciaops.com/2026/07/30/comparing-llms-in-copilot-services-round-1-chat/

Winner – Opus

Round 2 – https://blog.ciaops.com/2026/08/08/comparing-llms-in-copilot-services-round-2-researcher/

Winner – Critique

I’ve been pitting different LLMs inside M365 Copilot against each other ina world cup style elimination to see which comes out on top. The process involves taking a standard prompt and running it against all options. This prompt creates a multi page document requiring deep research and is quite involved. The results are then compared against each other using SharePoint Copilot and Gemini. Conclusions are then drawn.

This round the battle is between all the models available in Copilot Cowork Claude. Namely, these:

Screenshot 2026-08-11 074851

The other interesting factor here is that all these models as PAYG, so I’ll also give the costs for the same prompt.

To assess all these results together I found Gemini and SharePoint Copilot to both fail completely. Gemini keeps saying that it can’t find all the document, even though I have uploaded and also tried linking. SharePoint Copilot on the other hand starts processing but never gives me a result, no matter how long I wait. I will admit that these files are quite long (20+ pages) and quite complex, so there is a lot to digest. That said I threw them into Grok and those  results are here:

https://github.com/directorcia/general/blob/master/Copilot/Comparisons/20260811-Cowork-Claude-Grok-eval.md

In summary, the ranking where:

1. Opus 5

2. Fable 5 (Preview)

3. Fable 5 (Copilot) (Preview)

4. Opus 4.8

5. Sonnet 5

which all kind of makes empirical sense but still quite subjective I feel. However, for now I’ll swap to using Grok to evaluate these documents as a standard approach.

Now for the costs which don’t factor into the results:

Claude Fable 5 (Preview) = 3,387 credits
Claude Fable 5 (Copilot)(Preview) = 3,373.8 credits
Sonnet 5 = 2,509.3 credits
Opus 4.8 = 2,487 credits
Opus 5 = 2,240 credits


Which, on initial analysis, indicates that costs for all models are in the same kind of range (i.e. at least US$20 per prompt), with Fable 5 models being about 50% more expensive.

Given the analysis and costs, it seems Opus 5 with Cowork, if you are using Claude, is the most cost effective for the best result in Cowork.

So:,

Round 3 winner – Copilot Cowork Claude = Opus 5

Onto the next round.

Has Microsoft Made Copilot Cowork Too Expensive and Too Hard to Use for SMB?

image

A lot of SMBs think their AI journey starts by buying more AI.

I think that’s the wrong place to start.

Over the last few months I’ve spoken to plenty of SMBs and MSPs who are excited about Microsoft Copilot, curious about Copilot Cowork, and interested in agents. Then they see the pricing model, hear the term pay-as-you-go, start looking at consumption costs, and immediately become nervous.

I don’t blame them.

The real question isn’t whether Copilot Cowork is good. The question is whether Microsoft has made it too expensive and too complicated for the average SMB to realise value from it.

And right now, I think there’s a genuine argument that they have.

Technology should remove friction, not introduce it.

What is Copilot Cowork, really?

Forget the product names for a moment.

Copilot Cowork is essentially Microsoft’s way of allowing AI experiences to go beyond the capabilities included in your existing Microsoft 365 subscription.

That can mean accessing additional AI services, consuming AI-powered workflows, running agents, or using capabilities that incur usage-based costs rather than being covered under a fixed licence.

The attraction is obvious.

You only pay for what you use.

The problem is that most SMBs have spent the last twenty years moving away from unpredictable IT costs. Fixed monthly subscriptions made budgeting easier. Consumption billing moves the conversation in the opposite direction.

That’s where things start getting uncomfortable.

“Isn’t this just cloud pricing all over again?”

In many ways, yes.

The difference is that AI usage can be harder for business owners to understand than storage, mailboxes or virtual machines.

Step-by-Step: Getting Value Before Paying More

Review the Copilot Features You Already Have

Open the Microsoft 365 apps your users work with every day.

Look at the built-in Copilot experiences already available in products such as Edge, Outlook, Teams and Microsoft 365.

Microsoft continues to expand these capabilities, often without organisations fully realising what’s already included.

You can learn more from https://learn.microsoft.com/microsoft-365/copilot/microsoft-365-copilot

Enable Microsoft 365 Copilot Chat

Navigate to:

Microsoft 365 App > Copilot

Many organisations still aren’t getting consistent value from Copilot Chat, despite it being available to users.

If your staff aren’t regularly using Copilot Chat, adding more AI services is unlikely to solve the problem.

Identify Repetitive Business Processes

Look for activities that happen repeatedly:

  • Status reporting

  • Meeting preparation

  • Policy summarisation

  • Document review

  • Knowledge retrieval

These are usually the areas where AI can provide immediate value.

Microsoft’s adoption guidance is available through https://learn.microsoft.com/microsoft-365/copilot/adoption/ adoption resources.

Measure Usage Before Expanding

Use the reporting and administrative tools available within Microsoft 365 to understand whether users are actually engaging with Copilot.

Many organisations skip this step entirely.

They purchase AI licences first and measure value later.

That’s backwards.

Introduce Consumption-Based Services Carefully

Only after people are regularly using the AI capabilities already available should you consider broader AI consumption models.

If you do move forward, make sure you understand how billing controls and cost management work.

Microsoft provides additional information through https://learn.microsoft.com/microsoft-365/copilot/microsoft-365-copilot-page.

Why this actually changes behaviour

The biggest issue isn’t price.

It’s uncertainty.

SMBs don’t generally object to paying for technology that creates value.

They object to paying for technology when they can’t predict what the bill will be or explain what they’re getting in return.

That’s an adoption problem, not a technical problem.

Copilot Cowork asks organisations to think differently about AI consumption. That’s understandable from Microsoft’s perspective, but it introduces complexity at a time when many SMBs are still trying to establish basic AI habits.

Notice what’s missing?

People.

If users aren’t actively incorporating AI into their daily workflows, no pricing model will save the deployment.

My recommendation?

Use every Copilot capability already included in Microsoft 365 first.

Build habits.

Measure outcomes.

Identify repeatable wins.

Only then move toward consumption-based AI services.

Copilot doesn’t get tired. Use that.

But don’t pay for more AI until you’ve proven the value of the AI you already have.

The real opportunity isn’t buying more AI.

It’s getting more value from the AI you’ve already paid for. If you’re not showing clients this, you’re leaving value on the table.

The Browser Is Becoming Part of the Conversation

image

The more I use Copilot Cowork, the more I realise that the interesting part isn’t the AI itself. It’s what happens when the AI can finally interact with the same systems I use every day.

For years we’ve talked about AI helping us write documents, summarise meetings and draft emails. Useful? Absolutely. But most work doesn’t stop there. Sooner or later you need to log into a website, update a portal, retrieve information from a dashboard, or complete a process that lives inside a browser window.

That’s why the addition of local browser use in Copilot Cowork has caught my attention.

The Gap Between Advice and Action

One of the frustrations I see with AI tools is that they often stop at recommendations. They’ll tell you what to do, but they won’t actually do it.

Think about a typical day. You receive an email asking for an update. You find the relevant Teams conversation. You look through a SharePoint document. You then log into a service portal to check the current status before replying.

The work isn’t creating the email. The work is gathering the information across multiple locations and then taking action.

What browser use starts to do is close some of that gap.

Instead of simply suggesting a next step, Cowork can potentially navigate the website on your behalf while remaining inside the flow of the conversation. That changes the relationship between the user and the tool. You’re no longer just asking questions. You’re delegating pieces of work.

The Real Opportunity for SMBs

When I talk with SMBs and MSPs, one common theme emerges. Many important business systems don’t have polished integrations or APIs.

There are customer portals, supplier websites, expense systems, training platforms and countless line-of-business applications that still require someone to open a browser and manually complete a task.

That’s the reality of most small businesses.

Imagine asking Cowork to collect information from a supplier portal, compare it against a spreadsheet in Excel, create a summary in Word and then draft an Outlook email ready for review.

The value isn’t that any individual step is difficult. The value comes from reducing all the switching between systems that quietly consumes large portions of the day.

Every context switch has a cost. Most people simply don’t notice it because they’re so used to doing it.

Human Control Still Matters

Of course, this isn’t a licence to let AI loose across every system without oversight.

One thing I appreciate about Microsoft’s approach is that the browser session remains connected to the user’s identity and permissions. The AI is working with the same access the user already has rather than operating as some separate privileged account.

That’s important.

The future of workplace AI isn’t about removing people from the process. It’s about putting people higher in the process.

I see the role shifting from operator to supervisor. Instead of spending ten minutes navigating through screens and forms, you’re reviewing work, making decisions and approving outcomes.

The keyboard becomes a little less important. Judgement becomes more important.

What I’m Watching Next

I suspect many people will initially view browser use as a novelty. They’ll try a few demonstrations and move on.

I think that misses the bigger story.

The real significance is that Copilot is gradually moving beyond content creation and into work execution. The ability to interact with websites, applications and business processes starts to make the assistant feel less like a chatbot and more like a digital team member.

We’re still in the early stages, and I’d encourage organisations to test it carefully and deliberately.

But every major technology shift starts with a small behavioural change.

For me, that change is simple.

When I open a browser now, I’m increasingly asking not “What can I do here?”

I’m asking, “What can I hand off instead?”

Screenshot 2026-07-08 055301

How to Enable Browser Use in Copilot Cowork

Before users can take advantage of Browser Use, there are several prerequisites that must be met. According to Microsoft’s documentation, Browser Use is disabled by default and requires administrator configuration before users can access it. [learn.microsoft.com], [learn.microsoft.com]

Step 1: Enable Copilot Cowork

Browser Use relies on Copilot Cowork being available to users. As an administrator:

  1. Sign in to the Microsoft 365 Admin Center.

  2. Navigate to Copilot.

  3. Open Cost Management.

  4. Configure usage-based billing.

  5. Assign access to the users who will use Copilot Cowork.

If usage-based billing hasn’t been configured, users won’t be able to access Cowork features. [learn.microsoft.com]

Step 2: Enable Browser Access

Once Cowork is available:

  1. In the Microsoft 365 Admin Center, go to Copilot.

  2. Select Settings.

  3. Choose View all.

  4. Locate the Cowork settings.

  5. Turn on Allow browser access.

  6. Save the changes.

Microsoft notes that Browser Use is turned off by default and must be explicitly enabled by an administrator.

Step 3: Verify User Requirements

Each user must meet the following requirements before Browser Use will work:

  • Have a Microsoft 365 Copilot licence assigned.

  • Have access to Copilot Cowork.

  • Have Microsoft Edge installed.

  • Be running a current version of Edge.

  • Be signed into Edge using the same work or school account used for Copilot.

  • Use Cowork in a supported browser experience rather than an unsupported client. [learn.microsoft.com]
Step 4: Launch Copilot Cowork

The user can then:

  1. Open Microsoft 365 Copilot.

  2. Launch Copilot Cowork.

  3. Enter a request that requires interaction with a website.

For supported tasks, Cowork can use a hidden Microsoft Edge session running on the user’s device to interact with websites and web applications whilst keeping the user within the Copilot conversation experience. [learn.microsoft.com]

Screenshot 2026-07-08 060426

Step 5: Accept First-Run Consent

The first time Browser Use is invoked, users are presented with a consent prompt explaining:

  • Browser Use is currently in preview.

  • The feature can perform actions on the user’s behalf.

  • AI-generated actions may not always be perfect.

  • Sensitive actions may still require user confirmation.

Users must accept this prompt before Browser Use becomes available.

Step 6: Start with Low-Risk Scenarios

My recommendation is to begin with simple information-gathering tasks rather than business-critical transactions. Good examples include:

  • Looking up information in customer portals.

  • Checking ticket or order status.

  • Gathering information from vendor websites.

  • Researching data across multiple web-based systems.

As confidence grows, organisations can gradually expand usage to more complex workflows. This allows users to understand where the technology adds value while maintaining appropriate oversight.

If the process needs assistance while browsing it will pause and you will see something like:

Screenshot 2026-07-08 060945


Practical Tip: If Browser Use isn’t working, check the basics first. In my experience the most common causes are that Cowork billing hasn’t been enabled, Browser Access hasn’t been switched on by an administrator, or the user isn’t signed into Microsoft Edge with the same account they use for Copilot. [learn.microsoft.com]