Technology
Google’s Gemini assistant is a fantastic and frustrating glimpse of the AI future
It’s not surprise that Google is rapidly fast-following all the GenAI innovation that is cropping up. They have billions of reasons to protect their near monopoly in search advertising engines. I’ve been playing with the paid version of Gemini. Overall, it’s roughly on par with competitive products from OpenAI, MSFT, and Perplexity. The real benefit of the Gemini product is its integration with the Google Workspace suite of products. “Check my email for the last stay at Hilton and let me know how much the room cost”. That returned a very satisfying result.
Technology
This Paper Reveals The Surprising Influence of Irrelevant Data on Retrieval-Augmented Generation RAG Systems’ Accuracy and Future Directions in AI Information Retrieval
That’s the actual title of the article. Who writes the titles for these articles? Probably someone who does search engine optimization. Or AI. RAG (Retrieval Augmented Generation) is the newest (even though it’s not particularly new) twist on making large language models more trustworthy. Some proprietary data I saw recently demonstrated that RAG improved results, just not enough to be totally hallucination free. Research published in January indicates “that including irrelevant documents can unexpectedly enhance performance by more than 30% inaccuracy, contradicting our initial assumption of diminished quality.” Techniques for improving the impact of RAG on generative language are going to continue to make Language Models more trustworthy partners in knowledge discover and information retrieval.
Marketing
3 ways search marketers can prepare for the big cookie crumble
Unless you’ve gone on a long sabbatical to the mountains of Tibet, you’re aware of the Cookiepocalypse. This article is a quick reminder of some of the key actions that you should be taking to innoculate yourself from the impact of these changes. Ask your CMO how they’re doing.
Sales
AI is saving sales professionals more than two hours of work each day
I spoke with a sales exec earlier today. We were discussing the “20% of rep’s time focused on bullshit” and how that could be improved. This article supports his hypothesis that AI’s biggest impact on sellers will be productivity so they can spend less time on the bullshit and more time on connecting with prospects and progressing deals.
Career
AI Isn’t Taking Your Job — It’s Setting You Up For a Better One (Here Are 12)
I spoke with a good friend the other day. She’s been focused on prompt engineering. AI didn’t come for her job. AI created a new career opportunity for her. I’ve been playing with Python—yes, I’m taking a Coursera course—but only because I’ve been using ChatGPT (and other GenAI platforms) to write code and I’m wondering if I’m missing someting. I’m just playing right now, but I think it may be possible for me to create an MVP software product just with my tinkering. It’s scary (for all you real engineers out there) that I could fake being a software engineer. Here are a few better jobs.