Microsoft 365 and the apps that go with it such as Outlook and Calendar are down. Check Microsoft’s status dashboard for the fiasco. The company said:

“Current status: We've validated an alternative health check method on a subset of affected infrastructure, which allowed the servers in that subset to successfully return to service, and we're now applying this change more broadly. We're also preparing an automated recovery process to continually detect and restart unresponsive servers, to help keep them in service while this change progresses. Additionally, our review of recent changes has identified a small number of configuration updates that we're investigating as potential contributors to the issue.”

Google launched Gemini 4 Hydrogen, a fast model designed for multiple workflows. The model will run $1.50 per million input tokens and $7.50 per million output tokens. Gemini 4 Hydrogen also serves as an update across Google services including search, Gmail, Drive and more.

ServiceNow announced Autonomous CRM Engine in an effort to migrate enterprises from legacy CRM vendors quickly and then continuously watch for configuration drift. ServiceNow has been a bit obsessed about the CRM industry and gaining wallet share from incumbents.

Nebius reported a third quarter net loss of $37.5 million on revenue of $980.8 million, up 571% from a year ago. On a non-GAAP basis, Nebius reported earnings of $90.9 million. Nebius said it has funded its capex largely by prepayments.

In a shareholder letter, Nebius said:

"We expect prepayments and financing against existing contracts with investment-grade customers to cover 75%-85% of our 2027 capex, and have a wide range of other funding options available."

AWS launched AWS Emulator, which is a local emulator of AWS services that was totally written by autonomous AI agents. According to Amazon, thousands of autonomous agents write tests, find issues and fix code without human intervention. AWS' Kiro built AWS Emulator over a weekend. Here's a look at the three teams of agents that build it. Here's a look at the docs.

CoreWeave Database, a fully managed Postgre-SQL compatible database, is generally available. CoreWeave Database is designed to run withing CoreWeave Kubernetes Service clusters so data is on the same platform as GPUs.

• CoreWeave's Q3 shows revenue surge, CEO aims to allay debt worries

There's a major perception gap about AI automation, according to a survey from Latch.

A few findings worth noting from Latch's State of AI Automation at Work report include:

• Executives believe AI handles nearly a third of their team's recurring work. Employees say it's 19% of work.
• 8% of respondents said AI projects failed due to incapable models and 57% blame systems that miss how work actually gets done.
• More than half of companies have rehired a former employee because the know-how was never captured.

The survey covered 742 working professionals and 311 of them were executives.

In a blog post, the AI Futures Project riffed on Anthropic's AI progress. The article is based on a METR report, which estimates that 2027 will be the year where AI won't need humans. This stuff is worth a read but be sure to bring your salt.

ServiceNow has splurged on acquisitions in the last year and its balance sheet is showing the strain. In its quarterly SEC filing, ServiceNow outlined the following:

• Short-term debt for the third quarter was $2.09 billion, up from nothing at the end of 2025. See: ServiceNow Q3 mixed, but strong Control Tower demand
• Long-term debt at the end of the third quarter was $5.44 billion, up from $1.5 billion at the end of 2025.
• Goodwill on the balance sheet was $10 billion in the third quarter, up from $3.58 billion at the end of 2025.

Recent big deals:

• ServiceNow acquires Moveworks for $2.85 billion as it eyes CRM for agentic AI expansion
• ServiceNow acquires Logik.ai, as it steps up CRM efforts
• ServiceNow makes its cybersecurity move, acquires Armis for $7.75 billion
• ServiceNow acquires Veza, will integrate into AI Control Tower
• ServiceNow acquires Pyramid Analytics

Quantum computing company D-Wave reported a third quarter net loss of $48.2 million, or 13 cents a share, on revenue of $4.5 million, up 21% from a year ago. About half of that revenue was from professional services and quantum computing as a service. Earnings missed estimates slightly and revenue was a bit better than expected.

The biggest takeaway from quantum computing earnings is how immature these pure play vendors remain.

Rocket Lab's third quarter earnings call included the now obligatory question about space data centers. An analyst noted Rocket Lab CEO Peter Beck's skepticism relative to SpaceX and Elon Musk. Beck said:

"We've always retained a healthy level of kind of skepticism, but I would say that we've always been bullish certainly on defense-based orbital data centers. And also, clearly, I've been wrong plenty of times. So, we certainly weren't going to cut off any opportunities for ourselves, hence, the reasons why we developed and brought to market the new silicon cell for orbital data centers. Obviously, we have an orbital data center customer for a Neutron vehicle. And I would say that, like everybody else, we are doing our fair share of work around that opportunity. So we certainly haven't written it off, and we are monitoring all of the events. And like I said, doing our work to make sure that if we are able to gain conviction, then we're in a really good place to be able to capitalize on that."

"For me, it's economics. There's a number of kind of key things that have to happen for those economics to close and those key things are extremely difficult."

"I think there needs to be some major technological advances to unlock the economics. I think if you want to put smaller data centers or data centers of a small scale in orbit, I think that's all completely doable. Hence, the reasons why we are very bullish on orbital data centers for secure comms or defense purposes. I think that it will definitely happen. But the moving all data centers off Earth into space, I think there's a few challenges to solve there."

Cognizant said it has launched the Claude Frontier Capability Group that'll have more than 25,000 people dedicated to the effort. Cognizant said its managed services model will give customers to run Claude in production with more predictable AI spend.

• Why Cognizant Stands Out In AI-Powered Transformation | ShortList Spotlight
• Cognizant aims to underwrite AI enterprise results