In this article, I will focus on new features in ChatGPT, and how they might effect you as a Project Manager or part of a PMO. And it’s Halloween today, so here’s a spooky Project Manager:
3 new features in ChatGPT
Browsing is now out of preview and even though you probably already use it (maybe through Bing or even Bing Enterprise), this is a great feature to have. With the option to retrieve relevant and up to date information you can now prompt the AI and brainstorm on the latest information.
I used this recently in grabbing relevant external risks for a project I’m working on. I simply asked ChatGPT to look for recent news in the country regarding the topic and things I needed to be aware of.
The mobile app has voice activated prompting now! I’ve tried it out, and it is just like talking to a well informed co-worker (if you want this). The downside I found is that the GPT voice only starts responding when the response is complete. That means, that you will need to wait a moment before you get a response if you for instance ask it to tell you a step by step guide on how to make a scary pumpkin with the face of that SCREAM mask.
Added downside that might only be my paranoia: my voice is now recorded on OpenAI servers. And it has been proven time and time again that only a clip of 15 seconds is enough to mimic your voice almost perfectly.
Dall-E 3 image generation. Do you remember the early days? And I don’t mean 1980, I mean March 2023! Back in the days, we got lovely images like this one:
Now, if I ask the same prompt “The Project Management Office and Artificial Intelligence” to Dall-E 3 I get this response instead:
Prompt of PMO interest
No responses yet to my call for collaboration. But here’s something I add to my prompts recently to get better responses:
Instead of writing Act as a Project Manager fluent in PMBOK I now write Act as a Project Manager, fluent in PMBOK, with 10 years of experience working with [type of project]. Your answers should be short but complete, use references where possible.
This improved prompt statement gives the LLM more guidance in how to respond. Hope you like it!
Final notes
This is part of an ongoing string of articles I publish each week (or I try at least). Make sure you come back frequently and also take a look at the AI archive.
Apart from that, I’d of course like to welcome you to the YouTube channel.