✈️ Microsoft launches Co-pilot for Office

Hey there 👋

Welcome to your 5-min daily briefing where I cover prompt writing tips plus the developments in AI for product people and founders.

In today’s edition:

• How to use “counterfactuals” in your prompts to spark different perspectives

• GPT-4 early impressions

• Adept, the “AI teammate for everyone”, raises $350m

• Microsoft announces 365 Co-pilot

I’m Martin of Product Prompts. I’ve co-founded multiple startups (fintechs) and led corporate ventures, and now offer product strategy consultancy services, including helping forward-thinking clients with the knowledge, tools and training to figure out their AI strategy for their products and internal workflows. Get in touch to learn about how I could help you.

💬 Prompt writing tip: Counterfactuals

When we talk about counterfactuals in the context of prompt writing for product people, it basically means asking "what if" questions to explore different scenarios or outcomes that didn't actually happen in real life 🤔.

For instance, a counterfactual question for a product manager could be like "What if we had chosen a different colour for our product?" or "What if we had launched our product with a totally different marketing campaign?"

Asking these kinds of counterfactual questions can help spark creative thinking and encourage product managers to consider new and different possibilities that they may not have thought of before 🚀.

By exploring counterfactuals, product managers can gain insights that they might not have come across otherwise and use them to make better decisions that lead to a more successful product 🙌.

Couple of example use cases:

Dreaming up New Features: Ask questions like "What if our product had a feature that let users customize their profiles?" or "What if we added voice commands?" to come up with new features and improve user experience 🚀.

Avoiding Potential Problems: Asking "What if we faced a major security breach?" or "What if our primary supplier went out of business?" can help product managers prepare for problems and take preventative measures 🛡️.

Researching the Market: Ask questions like "What if we launched in a different country?" or "What if our competitors launched a similar product?" to understand the market and make informed decisions 🌎.

Testing Hypotheses: Ask "What if we changed the pricing strategy?" or "What if we changed the product's design?" to test hypotheses and evaluate their impact 🤓.

Improving User Experience: Ask questions like "What if our product could learn from user behavior?" or "What if we redesigned the onboarding process?" to identify areas for improving user experience and satisfaction 🤗.

So your prompt might be:

Which might give a response from ChatGPT like:

👀 GPT-4 first impressions

I’ve been using GPT-4, the latest OpenAI model launched a few days ago, which is accessible to everyone with a ChatGPT Plus account. Well, it is, but not the ability to upload images and have GPT reason based on the content of the image. That’s only currently available in the GPT-4 API - and I’ve literally just got my invite through 🤩 - so I’ll feed back on that when I’ve had a chance to play!

👉Here’s my take on GPT-4 so far:

• GPT-4 is launched with improvements such as higher word limit, more languages, image input, improved accuracy, and faster answers.

• GPT-4 has features such as multimodal input, longer text inputs, and advanced reasoning, which will expand the use cases for product people.

• It just feels “smarter”. That’s difficult to quantify, I know. Partly it’s the language used, partly the knowledge itself, plus the much longer memory for context means you can tackle much bigger tasks.

• Possible use cases of GPT-4 in PM work, where it’s a definite improvement over 3.5, include discovery, competitive intelligence, documentation, strategy, research, marketing, and learning.

💰 Adept raises $350m for Actions-based AI

Adept, a startup building AI that enables humans and computers to work together 🤝, just raised a whopping $350 million in a Series B funding round 🤑🤑🤑! With investors like General Catalyst and Spark Capital leading the way, Adept's valuation has reportedly reached a staggering $1 billion 💰💰💰.

So, what's Adept's big idea? They're creating an "AI teammate" that can be trained to use a wide variety of software tools and APIs, turning text instructions into digital actions 🤖. Adept's approach is different from other companies in this space, as they're focused on how people use computers to navigate software and browse the web.

Already, Adept's MVP, called ACT-1, is performing well and can import LinkedIn URLs into recruiting software, displayed as an overlay window on top of existing software like Google Chrome or Salesforce 🤯. And with strategic investors like Microsoft, Nvidia, Atlassian, and Workday, the potential for Adept's AI assistant is huge 💪.

Despite some high-level turnover, Adept's future looks bright. With this funding, they plan to continue productisation, model training, and headcount growth 🚀. And investors seem excited about Adept's potential, with nearly 50% of AI professionals believing that organisation investment in AI development will increase despite the macroeconomic climate, according to a recent survey by Intel-owned Cnvrg.io.

How could Adept be of benefit to product people?

A couple of initial thoughts:

User feedback analysis 🤔💬: With ACT-1, you could analyze user feedback from various sources like app reviews, surveys, or support tickets, and get insights on what users like and dislike about your product. This could help you prioritize your roadmap and build features that your users actually want!

Market research 🌎🔍: ACT-1 could also help you automate market research tasks, like competitive analysis, trend tracking, or customer segmentation. This could save you a ton of time and help you stay on top of industry trends and customer needs.

User onboarding and retention 🤝💼: With ACT-1, you could create and execute personalised onboarding and retention campaigns based on user behaviour data, like app usage patterns or feature adoption rates. This could help you improve user engagement and retention, ultimately driving product growth and success.

Agile project management 🏃‍♂️✅: ACT-1 could also help you automate certain tasks in the agile project management process, like creating and assigning tasks, tracking progress, and sending notifications. This could save you time and improve team efficiency, so you can focus on higher-level tasks like product strategy and roadmap planning.

Product analytics 📈📊: Finally, ACT-1 could help you gather and analyse product usage data, like user flows, conversion rates, or churn rates, and provide insights on product performance and user behaviour. This could help you make data-driven decisions on product optimisation and feature iteration, ultimately improving your product's success in the market.

For now though, all we can do is join the waitlist!

✈️ Microsoft announces Copilot: the AI-powered future of Office documents

So Microsoft just unveiled their new AI-powered Copilot for Microsoft 365 apps and services today! 🚀

This amazing Copilot, powered by GPT-4 from OpenAI, is designed to help users generate documents, emails, presentations, and so much more 👨‍💻👩‍💻

Similar to Clippy (remember that little guy?), the Copilot will sit alongside Microsoft 365 apps like a trusty assistant. It appears in the sidebar as a chatbot that allows Office users to summon it to generate text in documents, create PowerPoint presentations based on Word documents 👀, or even help use features like PivotTables in Excel.

According to Jared Spataro, Microsoft 365 head, Copilot works alongside users, embedded in the apps millions of people use every day: Word, Excel, PowerPoint, Outlook, Teams, and more. This Copilot is a whole new way of working! 😎

Plus every business gets their own version of ChatGPT based on their data.

Huge.

Thanks for reading, and happy prompting,

Martin, Chief Prompt Officer