I'm just about to go on Leo Laporte and team's This Week in Tech and I thought I'd share here how I prepared to be on the show. I primarily used 3 tools: Perplexity, What's Up With That?, and Hawkeye.
I used them for three functions: to see what the most significant parts of each article we're going to discuss are, to discover some overarching themes across the articles, and to learn about the work of the other people who will be participating in the conversation.
Here's what it looks like.
The X-Ray
First, I reviewed the list of all the stories we might discuss that TWiT producer Benito Gonzales shared with me ahead of time. He warned me that new ones could be added to the list right up to show time, and sure enough - we're going on the air in 15 minutes and I just noticed this Simon Willison post appear on the list about the latest version of Google Gemini.
I fire up the WUWT browser extension and:
It tells me what the most important parts are; not just based on the article text, but in context of all the other things I've analyzed recently, other news coverage of the announcement, and other indicators of the state of the art.
Here's what stands out in this article. Awesome.
WUWT picking out what stands out in Simon Willison's Gemini post, in context. Click to enlarge.
The Radar
In addition to analyzing each of the articles individually that way, I also detected some themes from across the 30+ stories we might discuss by using the Hawkeye magazine generation capability.
Usually, the organizations I work with monitor hundreds of organizations in their ecosystem for updates relevant to their organization's current priorities and then get daily and weekly highlight reels that they can editorialize in, use as newsletters to distribute, etc.
But in this case, I had Hawkeye generate two magazines for me. First, based on all the articles on the list of things we might discuss. And then I did a second one, analyzing the transcripts of the last 4 episodes of This Week in Tech, so I could get a refreshed context on the current conversation.
Here's my magazine view of the stories for this week, and here's a magazine overview of the last 4 TWiT discussions.
The Hawkeye magazine view of the last four TWiT episodes. Click to enlarge.
Drawing Connections
Finally, I did some research on the recent work of my two fellow guests, journalists Larry Magid and Jacob Ward. For this, I asked Perplexity to "give me 3 links to the most recent work from each of them." I looked over the headlines and picked some articles that looked interesting, and that might be useful to have in my mind for today's discussion.
Once I was on Jacob's substack, I saw that his most popular post was about AI making us lie to ourselves and one another. That was a very interesting post and video, something I'll be pondering for some time.
Then, Larry's links led me to this recent write up he posted about his use of AI in his own medical treatment. I read that article, and I ran What's Up With That on it. Not just to see what stood out, but then I also ran the "Cutting Edge" function to see what organizations around the world are at the cutting edge of the kind of topics Larry discussed in that post, and I ran a Risk Scan.
WUWT generated hundreds of hypothetical intersections between the themes Larry discussed, the projects it knows I'm working on, and other things I have or could read. It picked just the most impactful and likely ones to bring to my attention.
In this case, it said "you know, the themes brought up here could intersect with what Jacob Ward just wrote about around AI leading to us lying to ourselves." And suggested I save this intersection to a folder related to one of my projects.
A Risk Scan surfacing an unexpected intersection between two of today's fellow guests' recent work. Click to enlarge.
Off to the show
Ok, that's how much of my research I am able to describe right now - it's time to go be on the show!
I hope if you found this interesting, you'll check out WUWT and Hawkeye. Let me know if you'd like to discuss either of them.
Chrome
Firefox