Two headlines:
Microsoft bundles its AI-powered Office features into Microsoft 365 Personal and Home subscriptions for an extra $3 per month
— Tom Warren/The Verge
Google makes all of its Gemini features for Workspace apps free, after previously charging $20+ per user per month, but raises the price of all Workspace plans
— Abner Li/9to5Google
Opinion: AI does not sell.
Meanwhile Microsoft renames the Microsoft 365 app, formerly known as Microsoft Office, as Copilot M365.
Small print regarding Microsoft 365 Family: “AI features only available to subscription owner and cannot be shared; usage limits apply.“
This clearly is a developing story.

Reminds me of IBM desperately naming everything “Watson.”
Exactly
While initially sounding exciting, if you think about it, AI has almost no real use case. Reason is in my opinion, that even if you find something productive it can do, you have to doublecheck everything.
My imagination might be limited, but the tasks I imagine AI to be good in, are only with it, if you don’t have to do that.
That leaves the „private sector“ for summaries and genmoji, which might be nice but not productive.
My experience is a bit more positive about AI tools in the workspace and yesterday’s announcement on pricing is a big win for my company.
We started using Gemini AI with Google Workspace back in October (500 licenses) and the feedback is extremely positive. These are some of the use cases that I know about:
– NotebookLM. People have used this in multiple ways, in some cases just to do research on specific documents but one use case I particularly likes is when someone talked about how they use it to create podcast summaries of large documents that they can then listen to on their commute.
– Stack ranking large numbers of documents based on a requirements doc
– converting requirements into user stories
– Improving quality of content in docs and slides
– Critiquing slides/decks before big meetings. Ask Gemini to review the deck and anticipate questions that may be asked by CFO, CEO, etc.
– Code assist (this was my one, I use it to help me with Apps Script)
– Document classification and labelling for DLP
– Bringing customer personas to “life” – we have used Gems in gemini to create bots that are pre-configured to have the attributes of different customer segments. Marketers can then test different brand messages with these personas and get feedback on what appeals and what doesn’t.
Yes, it also does summaries and help me write but that’s less useful for us as most conversations are in Slack.
Thank you for your insights, Kieren.
An addition, where LLMs proved to be helpful for us in a business setting:
We trained a LLM on our vast documentation. With it our customers are a lot more successful in finding helpful pages than with traditional searches. They also get summaries, but crucial to me always seemed that they get links to handwritten documents and code.
And, no, the LLM does not replace any of our excellent support engineers.