Anji Ismail

Being concerned and skeptical is the best way to understand how things works, so that's a good start. However, don't let it paralyse you. Define what are the outcome you expect from your co-pilots, chat or whatever tools that is going to produce the insights, and go deeper to understand what datasets were used. Depending on the applications either open-source or proprietary and where it's being used (publicly or privately in a company let's say), there are many factors that will affect how you handle the relationship between the outputs generated and the privacy impact. Finally, it also depends on what type of data you are handling (e.g. generic information vs PII).

Like in any project, start by defining urgency. Who said it was urgent? why? AI being a relatively new field, at least in its current form and enterprise vertical applications, going too fast might not be the best idea. You actually want to take time to test, to observe, to iterate etc.

First make sure you are not the one to be wrong. If there is pushback it might come from somewhere. Especially within AI, the field is so new that it's pretty easy to take shortcuts and think everything is going to be magical. Discuss the underlying reasons of the pushback and understand them, and go from there.

If you answer starts with "to ensure a smooth transition to AI roles", it's generated by ChatGPT 😅 Re: on the topic. Your job is to show your team that AI isn’t just for the researchers and engineers —it’s going to be a game-changer for everyone: sales, marketing, design, production etc. If they truly get that AI tools will amplify their skills, just like the internet did for the previous generation, they’ll be way more open to it. Forcing new tools or changes on people who don’t want them never works. As a manager, it’s your job to sell the vision of the future—show them how AI is going to redefine their work. Once they see the value for themselves, the transition won’t be a fight, it’ll be something they’re excited about.

With everything coming from OpenAI, Anthropic, Meta, Nvidia, and thousands of AI startups, predicting the next big trends can feel overwhelming. Every day, there are new models, capabilities, and breakthroughs. Here’s how I do it: Be part of the industry: To know what’s coming, you have to build and be involved in the ecosystem. Simply reading the news or LinkedIn posts won’t cut it Focus on your goals: Predicting trends doesn’t always equal success. Focus on what you want to achieve, and use trends to time your launches, but don’t let the latest buzz dictate your roadmap Discern the noise from real trends: This is the hardest. Stay sharp in distinguishing marketing hype from fundamental shifts that could truly impact your business

The issue of scope creep isn’t unique to AI—this is a basic product management challenge. Every roadmap eventually gets hit with odd features—whether it's competition, customer demands, investor pressure, or a moonshot idea from a PM. To avoid derailing progress, the key is to ship fast. Stick to tight cycles (you know the drill: 2-week sprints, demo, release) and measure everything. Also, resist the temptation of shiny features your competitors are pushing or trends from the media. Your business and product are unique—prioritize based on your own goals, not someone else’s. Focus on delivering value, not distractions.