Earlier this week, I led an Avasant panel discussion on Generative AI and the New Data-Driven Productivity Paradigm. You can watch the entire video here.
I started with a brief introduction to set the stage, comparing today's generative AI (GenAI) services with earlier forms of AI, which date back several decades. The panel then discussed a number of important elements of generative AI:
Why GenAI has gotten so much attention over the past year.
Where we see GenAI delivering productivity gains as well as top line revenue growth.
Data quality as a prerequisite for realizing GenAI benefits as well as issues around confidentiality and data privacy.
The regulatory landscape around GenAI, even today with GDPR as well as in the future.
The enterprise risks for enterprises to consider in implementing GenAI systems.
We ended with a lightning round about practical steps for organizations to take in getting started with GenAI.
Don’t listen to a vendor about AI, says John Willis, a well-known technologist and author in the latest episode of The New Stack Makers.
“They’re going to tell you to buy the one size fits all,” Willis said. It’s like going back 30 to 40 years ago and saying, ‘Oh, don’t learn how to code Java, you’re not going to need it — here, buy this product.'”
Willis said that DevOps provides an example of how human capital solves problems, not products. The C-level crowd needs to learn how to manage the AI beast and then decide what to buy and not buy. They need a DevOps redo.
One of the pioneers of the DevOps movement, Willis said now is a time for a “DevOps redo.” It’s time to experiment and collaborate as companies did at the beginning of the DevOps movement.
“If you look at the patterns of DevOps, like the ones who invested early, some of the phenomenal banks that came out unbelievably successful by using a DevOps methodology,” Willis said. “They invested very early in the human capital. They said let’s get everybody on the same page, let’s run internal DevOps days.”
Just don’t let it sort of happen on its own and start buying products, Willis said. The minute you start buying products is the minute you enter a minefield of startups that will be gone soon enough or will get bought up by large companies.
Instead, people will need to learn how to manage their data using techniques such as retrieval augmentation, which provides ways to fine-tune a larger language model, for example, with a vector database.
It’s a cleansing process, Willis said. Organizations will need cleansing to create robust data pipelines that keep the LLMs from hallucinating or giving up code or data that a company would never want to let an LLM provide to someone. We’re talking about the danger of giving away code that makes a bank billions in revenues or the contract for a superstar athlete.
Getting it right means adding some governance to the retrieval augmentation model. “You know, some structuring, ‘can you do content moderation?'” Are you red-teaming the data? So these are the things I think will get really interesting that you’re not going to hear vendors tell you about necessarily; vendors are going to say, ‘We’ll just pop your product in our vector database.'”
This is a quick post to highlight my recent appearance on the Data Radicals podcast (Apple, Spotify), hosted by Alation founder and CEO, Satyen Sangani. I’ve worked with Alation for a long time in varied capacities — e.g., as an angel investor, advisor, director, interim executive, skit writer, and probably a few other ways I can’t remember. This is a company I know well. They’re in a space I’m passionate about — and one that I might argue is a logical second generation of the semantic-layer-based BI market where I spent nearly ten years as CMO of Business Objects.
Satyen is a founder for whom I have a ton of respect, not only because of what he’s created, but because of the emphasis on culture and values reflected in how did it. Satyen also appreciates a good intellectual sparring match when making big decisions — something many founders pretend to enjoy, few actually do, and fewer still seek out.
This is an episode like no other I’ve done because of that history and because of the selection of topics that Satyen chose to cover as a result. This is not your standard Kellblog “do CAC on a cash basis,” “use pipeline expected value as a triangulation forecast,” or “align marketing with sales” podcast episode. Make no mistake, I love those too — but this is just noteably different content from most of my other appearances.
Here, we talk about:
The history and evolution of the database and tools market
The modern data stack
Intelligent operational applications vs. analytic applications
Why I feel that data can often end up an abstraction contest (and what to do about that)
Why I think in confusing makets that the best mapmaker wins
Who benefits from confusion in markets — and who doesn’t
Frameworks, simplification, and reductionism
Strategy and distilling the essence of a problem
Layering marketing messaging using ternary trees
The people who most influenced my thinking and career
The evolution of the data intelligence category and its roots in data governance and data catalogs
How tech markets are like boxing matches — you win a round and your prize is to earn the chance to fight in the next one
Data culture as an ultimate benefit and data intelligence as a software category
I hope you can listen to the episode, also available on Apple podcasts and Spotify. Thanks to Satyen for having me and I wish Alation continuing fair winds and following seas.
Just a quick post to share the recording of the webinar we did yesterday, where my Balderton Capital colleague, Michael Lavner, and I discussed the SaaS Metrics That Matter. You can find the slides here.
Paris Marx is joined by Emily M. Bender to discuss what it means to say that ChatGPT is a “stochastic parrot,” why Elon Musk is calling to pause AI development, and how the tech industry uses language to trick us into buying its narratives about technology.
Emily M. Bender is a professor in the Department of Linguistics at the University of Washington and the Faculty Director of the Computational Linguistics Master’s Program. She’s also the director of the Computational Linguistics Laboratory. Follow Emily on Twitter at @emilymbender or on Mastodon at @emilymbender@dair-community.social.
Tech Won’t Save Us offers a critical perspective on tech, its worldview, and wider society with the goal of inspiring people to demand better tech and a better world. Follow the podcast (@techwontsaveus) and host Paris Marx (@parismarx) on Twitter, and support the show on Patreon.
The Future of Life Institute put out the “Pause Giant AI Experiments” letter and the authors of the “Stochastic Parrots” paper responded through DAIR Institute.
Books mentioned: Weapons of Math Destruction by Cathy O'Neil, Algorithms of Oppression by Safiya Noble, The Age of Surveillance Capitalism by Shoshana Zuboff, Race After Technology by Ruha Benjamin, Ghost Work by Mary L Gray & Siddharth Suri, Artificial Unintelligence by Meredith Broussard, Design Justice by Sasha Costanza-Chock, Data Conscience: Algorithmic S1ege on our Hum4n1ty by Brandeis Marshall.