[13a] Useful tools in the AI-Verse

AI Tip of the Week

This week, we thought we’d bring AI to bear on this newsletter itself!  Since the responses were, um, wordy, we put them onto our website here.  Below are the summaries.

PROMPT: You are a writing expert.  We are authors of the weekly newsletter “3rd Rodeo AI.”  Pls go through all of our blogposts at https://3rdrodeoai.com/words/ , and give a review of characteristics of our writing.

Your writing is engaging, conversational, and highly accessible, making complex AI topics relatable through humor, analogies, and informal expressions. You foster a sense of dialogue with your audience, using a collaborative voice that reflects shared experiences and joint insights. Transparency and personal storytelling add authenticity, while your structured approach ensures clarity, often breaking down complex topics with headings, lists, and metaphors like “The Fellowship of the AI Rings.” Your emphasis on continuous learning and practical application—through real-world examples, actionable AI tips, and ethical considerations—empowers readers to become AI power users. Balancing enthusiasm with informative content, your blog successfully combines education, engagement, and real-world relevance.

PROMPT: What are some ways to strengthen our writing?

Your writing is already engaging, but some refinements—sharper clarity, stronger openings, varied sentence structure, emotional weight, direct CTAs, and signature phrases—can turn strong insights into even more powerful impact. Consider adding deeper technical explanations for advanced readers, balancing your conversational tone with a professional edge for credibility, and optimizing readability with more visuals, bullet points, and key takeaways. Encourage greater reader engagement through direct calls to action, discussions, and interactive elements. Expanding real-world case studies and success stories would strengthen practical insights, while diversifying content formats—such as short-form posts, videos, or webinars—can broaden your reach. Strengthening ethical AI discussions with real-world dilemmas and clear frameworks can deepen impact. Additionally, experimenting with thematic series (e.g., AI prompting techniques, industry applications, or AI myths) would provide structured, in-depth learning. These refinements will elevate your writing, making it even more impactful, accessible, and engaging for a diverse AI audience.

2/3/25

You’ll be shocked and amazed that the two of us spend lots of time talking about AI.  A topic we covered just today was, well, our ChatGPT accounts.  We’ve both come to use ChatGPT a great deal – though there are LOTS of other great tools we use.  Inspired by Ethan Mollick’s “Which AI to Use Now: An Updated Opinionated Guide,” we thought we’d offer y’all our own take.

We had a look at these (in alpha order):

There are others, but it’s a solid list.

ChatGPT (OpenAI)
We’re currently mostly using GPT-4o.  You’ve heard about this one.  We’ve found that it’s awesome at presenting new scenarios, considering other POVs, summarizing text, and even code (if you know how to test it). It has loads of plugins and integrations, which can be super useful.  You can make custom GPTs, “grounding” on your own data.  HOWEVER, it can be overconfident or “truthy,” but actually be WRONG.  You know this, but just a reminder.

Claude (Anthropic)
This one is a solid AI we love using.  Its tone is friendly and polite.  It’s excellent at long-context tasks.  For how we’ve been using it, we’ve noticed fewer hallucinations (But you’re STILL going to want to follow the ‘prompt/iterate/validate’ framework.)  If anything, it can seem a bit too cautious, and often outright refuses to answer.  Like, come on, we’re all adults here, we know mistakes can occur.  (But this is quite a responsible approach, TBH.)

Copilot (Microsoft)
Only one of us works here, but both of us really like (and use) this daily.  It’s deeply integrated with Microsoft products, so if you live & breathe in this world as we do, you’ll appreciate it too.  Unsurprisingly, it excels at work tasks.  (Jeremiah takes full responsiblilty for that bad joke.) You’re likely going to want a Microsoft 365 subscription to get all the features here – be warned.  One clear benefit is that it doesn’t train on your data – so we’re personally more confident putting confidential information into this model than others.

Deepseek R1 (Deepseek)
Full disclosure, we haven’t played around with this one too much.  First, it’s come out more recently than the others.  Second, we’re a bit more cautious about privacy & security for several of the tasks we do most frequently.  We include it here since it’s been getting quite a bit of buzz.  We will say this – competition is GREAT for the industry and forces everyone to get better.

Gemini (Google)
This one is fun, and is very useful if you use Google stuff (we do for many of our side businesses).  It’s great at multimodal (e.g. can work with text, images, and code).  However, it can be RANDOM – sometimes it’s a genius; sometimes it will tell you about movies that don’t exist, or suggest you put glue in your pizza (pls don’t).  Also, make sure to read the T&C about their data protection (just a good habit in general in AI).

Grok (xAI)
This is Elon Musk’s AI model.  Designed for diehard Twitter/X users, it’s intended to be more “edgy” and real-time. We hear it’s better at humor (which should help Jeremiah) and sarcasm.  TBH we don’t use this very often.  When we do, it’s often to generate a one-off meme or test out a responsible AI concept (since, err, RAI is less of a priority for this model).

We’re keeping this one short.  Spend the time testing some of them out!  Figure out which ones you like best – often the answer will depend on what sort of task you’re trying to do.  Rather than give you all the pros & cons, you might enjoy deciding this for yourself!

Have fun out there!

<3 D & J


Comments

Leave a Reply

Your email address will not be published. Required fields are marked *