Strategic Leadership in the Age of Generative AI: A Guide for Long-Term Success

Nov 09, 2024

Can a new book finally help executives grasp the full potential of AI? In this issue of “The AI Economy,” we look at “Winning with Generational AI” and chat with one of its authors, Charlene Li, to understand if this 90-day blueprint can offer highly sought-after clarity.

You have probably heard about NotebookLM, but what is it, and is it any good? We examine this AI-powered notebook app to see if it has any potential.

And right before press time, Meta shocks with its video-generation AI models to take on OpenAI, Google, and Runway.

All of this and this week’s roundup of AI news you may have missed! 🙏

The 90-Day Gen AI Transformation

We constantly hear company executives struggling to figure out how to apply artificial intelligence in the workplace to generate real value. However, a new book from analyst Charlene Li and Harvard Business School’s Chief Digital Officer Katia Walsh could provide the blueprint organizations need to properly and effectively integrate generative AI into their business strategy — all in just 90 days.

In “Winning with Generative AI,” Li continues her pattern of turning executives into digital transformation leaders. Unlike in the era of Social Business, companies cannot delegate responsibility to different departments. With AI, management needs to be involved and should be required to demonstrate their knowledge of the technology. With this leadership vacuum, frontline workers are left anxious and unclear about how far they can use AI and its impact on their jobs.

It’s time for company leaders to lead.

If there’s any confusion about how to craft a responsible AI policy, Li reminds me that organizations need only remember their foundational values. These are the North Star for any rule or order established by management. Any decision to bake generative AI into a long-term business strategy must build on these values and the company’s purpose and mission.

This is a critical step because of how transformative AI can be for a business. Li calls it paradigm-shifting, and having the proper guardrails and rules for how workers can and cannot use AI will be important to ensure success.

“Winning with Generative AI” is less of a book about technology. Instead, it’s a guide designed for leaders. There are a dozen steps for them to complete to succeed, each customizable. The courses cover everything from “how to get your team ready to foundational values, how to write a generative AI strategy, pull out your business strategy and how to think about it to create value, how do you start looking at the feasibility of it and prioritizing that once we put in the various plan.” And that’s only one-half of the process.

Each chapter outlines what will be accomplished, the lessons learned, and outcomes with clear objectives and key results.

Can this upcoming book help leaders get up and running with AI in just 90 days? Check out my interview with Li to learn more. Plus, I asked her if the lessons learned in her inaugural book, “Groundswell,” about technology adoption inside organizations, still hold up nearly two decades later. Find out what she said ⤵️

▶️ Read more about “Winning with Generative AI” (My Two Cents)

Why Are We Obsessed with NotebookLM?

Google first released NotebookLM in December, a notebook app powered by the company’s Gemini Pro model. Initially, it used AI on notes and documents to create summaries, outlines, study guides, or other forms of content. In September, Google added audio support, letting users turn their documents into audio discussions—it even generates a podcast with two artificial hosts from user-provided content. Days later, another update let it provide summaries of YouTube videos and audio files and generate shareable audio discussions.

NotebookLM has since grown in popularity and set the internet ablaze. It shouldn’t surprise us that AI makes content creation easier; Google’s app is another example. In a test, I shared my blog post about Charlene Li’s upcoming book, and NotebookLM spit out a seven-minute podcast with two AI-generated hosts discussing Li’s work based on my writing. It inserts footnotes to attribute claims to added sources to lend credibility to the results. I could also generate an FAQ, briefing guide, and more from my content.

Like all AI tools, NotebookLM is no stranger to hallucinations, so it’s important to double-check the information before sharing it publicly. The “podcast” the app generated for me has all the right points, but it suggests Li’s book has already been published—it’s not on store shelves yet—and also mispronounces her last name.

Consider each “notebook” as a project supporting up to 50 sources. Once content has been added, the experience is similar to using Google Gemini, in which you can prompt it to generate a summary or craft other content. In another example, I added 10 YouTube links from “The Created Economy” podcast I co-hosted with Gregarious Narain. I asked NotebookLM to generate a case study based on the transcripts from the videos, and the result could have been better.

For fun, I created a notebook and added news articles about the app before prompting NotebookLM to tell me what makes it so popular. This is what it generated in response:

NotebookLM provides a new way to interact with and process information. Users have quickly adapted to AI tools like NotebookLM, moving beyond initial reactions to focus on how to use the tool effectively. This suggests that NotebookLM is meeting a need for tools that can help people make sense of complex information.
NotebookLM offers features that appeal to a wide range of users. Students use it to accelerate their learning, while professionals use it for research and knowledge sharing.
NotebookLM has a user-friendly interface and offers a variety of features. Users appreciate the auto-generated summaries and suggested follow-up questions that help them understand difficult text and synthesize connections between multiple documents. The “Noteboard” feature allows users to save responses, excerpts, and notes, and then use them to create summaries, outlines, study guides, and even email newsletters and scripts.
NotebookLM is constantly being updated with new features. The team behind NotebookLM is committed to listening to user feedback and adding features that meet their needs. For example, based on feedback about the “Audio Overviews” feature, the team is adding more controls, such as the ability to select different AI host “personas” and adjust the length of podcast episodes.
NotebookLM is free for Google Account users. This makes it accessible to a wide range of people, which likely contributes to its popularity.

However, I do see promise in using an app like NotebookLM. It’s a blend of Microsoft OneNote and Otter.ai, all powered by Gemini. You could upload an audio recording of an executive interview and have NotebookLM generate a summary to share in a presentation. Or, you could create a notebook with multiple speeches your CEO made to identify key talking points or themes you’d like to build a content program around. In addition, you could create a notebook while learning a new skill and then prompt NotebookLM to generate study guides to keep the information fresh in your mind.

As mentioned earlier, this tool isn’t without concern. It could be used for malicious purposes or even generate false information. Google says that more controls will be introduced soon.

What do you think about NotebookLM? Leave a comment below!

Meta Announces Movie Gen, an AI Model Set to Rival OpenAI and Google

A screenshot taken of a video created using Meta’s Movie Gen AI models. Image credit: Meta

There is a new entrant in the AI movie wars: Meta. Facebook’s parent company announced the release of Movie Gen, a new generative AI model suite that creates realistic-looking images, videos, and audio media. In a clip posted to his Instagram account, Chief Executive Mark Zuckerberg revealed Movie Gen would be coming to the service next year.

Movie Gen has four capabilities: video generation, personalized video generation, precise video editing, and audio generation.

Video Generation

With 30 billion parameters, Movie Gen can generate videos of up to 16 seconds at 16 frames per second. By comparison, Google claims its counterpart Veo can create “high-quality 1080p clips that can go beyond 60 seconds,” and OpenAI’s Sora supports up to 60 seconds.

Personalized Videos

Meta will input someone’s image and combine it with a text prompt to generate a video containing that person along with the visual details derived by the query.

Precise Video Editing

You can use this model to specify elements within a video you wish to modify, such as objects or colors, using a text prompt, locally or throughout the entire film.

Audio Generation

With a 13 billion parameter model, Movie Gen can ingest a video and text prompt to spin out a high-quality and high-fidelity audio file of up to 45 seconds in length, including “ambient sound, sound effects (Foley), and instrumental background music—all synced to the video content.” Meta is also introducing an audio extension technique capable of generating coherent audio for videos based on “arbitrary lengths.”

Chart showing how human creators responded to using Meta’s Movie Gen compared to other industry models. Image credit: Meta

The AI video generation space is becoming increasingly crowded as Meta joins two major AI players and smaller competitors such as Runway, Synthesia, D-ID, and more.

Today’s Visual Snapshot

How much are Silicon Valley companies paying AI and machine learning specialists? A recent study shared by Menlo Ventures reveals the equity granted to the top 1 percent in these fields. Unsurprisingly, they’re rewarded generously, especially research scientists. The tech industry continues to pay premiums for these critical skills.

Quote This

I come from academic AI and have been educated in the more rigorous and evidence-based methods, so I don’t really know what all these words mean…I frankly don’t even know what AGI means. Like people say you know it when you see it, I guess I haven’t seen it. The truth is, I don’t spend much time thinking about these words because I think there’s so many more important things to do…

— Dr. Fei-Fei Li, renowned computer scientist and “Godmother of AI,” responding to a question about what she thought about an “AI singularity.”

This Week’s AI News

🏭 Industry Insights

🤖 General AI and Machine Learning

🔬 Science and Breakthroughs

💼 Business and Marketing

OpenAI’s Sam Altman concentrates power on path to $157 billion valuation (Bloomberg)
Accenture to train 30,000 staff on Nvidia AI tech in blockbuster deal (CRN)
Before Mira Murati’s surprise exit from OpenAI, staff grumbled its o1 model had been released prematurely (Fortune)
Anthropic hires OpenAI co-founder Durk Kingma (TechCrunch)
Early generative AI startup leader Tome lays off 31 percent of staff (The Information)
Pinterest launches Performance+, a suite of AI tools for advertisers (My Two Cents)
Google brings ads to AI Overviews as it expands AI’s role in search (TechCrunch)
Nvidia acquires OctoAI, a startup helping companies run AI models (GeekWire)
Y Combinator faces criticism after it backed an AI startup that admits it cloned another AI startup (TechCrunch)

📺 Media and Entertainment

💰 Funding

OpenAI raises $6.6 billion in largest VC round ever (Axios)
AI chipmaker Cerebras files for IPO to take on Nvidia (CNBC)
ElevenLabs reportedly could be fundraising at a $3 billion valuation (TechCrunch)
Eye on AI: AI’s torrid pace of funding continues (Crunchbase News)
Crescendo, an AI startup providing AI-enhanced contact center services, raises $50 million at $500 million valuation (Bloomberg)
Numa raises $32 million to bring AI and automation to car dealerships (TechCrunch)
Lorikeet, an AI customer service platform, comes out of stealth with $5 million in seed funding (Fortune)
11x.ai, a developer of AI sales reps, reportedly raises $50 million (TechCrunch)
Qodo raises $40 million Series A to bring quality-first code generation and testing to the enterprise (TechCrunch)
Retail supply chain AI startup Ameba raises $7 million seed round (TechCrunch)

⚖️ Copyright and Regulatory Issues

Federal judge blocks California’s new AI law in case over Kamala Harris deepfake (TechCrunch)
Many companies won’t say if they’ll comply with California’s AI training transparency law (TechCrunch)
European Commission probes TikTok, YouTube and Snapchat’s video recommendation algorithm (Euronews)
Can AI regulation survive the First Amendment? (Platformer)

💥 Disruption and Misinformation

End Output

Thanks for reading. Be sure to subscribe so you don’t miss any future issues of this newsletter.

Did you miss any AI articles this week? Fret not; I’m curating the big stories in my Flipboard Magazine, “The AI Economy.”

Connect with me on LinkedIn and check out my blog to read more insights and thoughts on business and technology.

Do you have a story you think would be a great fit for “The AI Economy”? Awesome! Shoot me a message – I’m all ears!

Until next time, stay curious!