Welcome to The Upgrade

Welcome to my weekly newsletter, which focuses on the intersection of AI, media, and storytelling. A special welcome to my new readers from AugXLabs, Thomson Reuters, Merchbots, and many other top organizations—you’re in good company!

In today’s issue:

• The Week’s Top AI Stories 📰

• 🎓 20% off the MindStudio Academy: Learn AI Today! ⚡️

• 🎧 Podcast: Confessions of an AI Explorer with Synthminds.Ai Founder Wes Shields

• 🎙️The Big Interview: Noah Kravitz, Host of the NVIDIA AI Podcast

The Week’s Top AI Stories

Top AI Headlines

• OpenAI Insiders Warn of a ‘Reckless’ Race for Dominance — The New York Times

• Adobe Promises That It Hasn’t Gone Full Big Brother — Slate

• Meta's AI chatbot says it was trained on millions of YouTube videos — Business Insider

• How Apple Fell Behind in the AI Arms Race — The Wall Street Journal

• Microsoft Lays Off 1,500 Workers, Blames "AI Wave" — Yahoo Finance

Ethics, Society, & Safety

• We’re unprepared for the threat GenAI on Instagram, Facebook, and Whatsapp poses to kids — Fast Company

• Google's AI Overviews Will Always Be Broken. That's How AI Works — WIRED

• OpenAI insiders are demanding a “right to warn” the public — Vox

Legal, Policy & Copyright

• Why the few big AI players worry US antitrust regulators — Reuters

• Washington is waking up to AI’s risks about three years too late — CNN

• The Senate’s failure on AI policy leaves legislation up to the states — The Hill

AI in the Workplace

• AI Is Your Coworker Now. Can You Trust It? — WIRED

• Writers accept lower pay when they use AI to help with their work — New Scientist

• No AI skills on your resume? Expect lower salaries and fewer roles in the future — The Hill

• Deepfakes, Fraudsters and Hackers Are Coming for Cybersecurity Jobs — The Wall Street Journal

• Generative AI Can Write Computer Code. Will We Still Need Software Developers? — Forbes

AI Tools

• Humane warns AI Pin owners to ‘immediately’ stop using its charging case — CNET

• AI agents are having a 'ChatGPT moment' as investors look for what's next after chatbots — CNBC

• Will Apple's AI Finally Make Siri Smart? — PCMag

• How to Break Out of AI-Fueled Analysis Paralysis — Entrepreneur

🎓Learn AI with MindStudio Academy! 💻

Ready to learn the fastest way to build no-code AI-powered apps and automations? The Upgrade is partnering with MindStudio to lead the MindStudio Academy! ⚡️

The next cohort starts next Saturday, June 15th at 10 am ET / 7 am PT. Only a few spots left, so don’t wait!

SAVE 20% with code: THEUPGRADE20

The Big Interview: Noah Kravitz, Host of the NVIDIA Podcast 🎙️

Noah Kravitz is the host of NVIDIA's AI Podcast, owner of Resonant Digital, and a longtime digital storyteller.

Peter: Tell me about your background as a storyteller. I know that you’ve been at the intersection of types of storytelling for a long time in your career. You've also been a freelancer and a business owner.

Noah: Yeah, the straight-and-narrow career path just didn't seem to be for me. Writing has always come naturally, and I've always been interested in radio, film, video making, and all that kind of stuff. I started writing professionally in high school as a stringer for a local sports newspaper. It went from there. I tried my hand at a traditional publishing career path with an internship at a magazine, but it just wasn't for me. I realized I wanted to write; I didn't want to be an editor or publisher. So then, I got into writing for some tech-focused publications, which snowballed from there.

Peter: How did you land the coveted role of host of the Nvidia podcast?

Noah: I've been hosting for nearly seven years now. To give the short version, I moved to California in the early 2000s. I was teaching computers in Oakland for a bit. Towards the end of that, I saw an ad looking for cell phone reviewers. I wasn't sure what to do next, so I answered the ad, borrowed a phone, and wrote a review. They published it and asked if I wanted to do more. It was with a site called PhoneDog.

I ended up working with them for a number of years as a reviewer and then editor-in-chief, helping build out their editorial. That got me involved with different tech companies. Through that, some folks from Nvidia PR reached out. We built a relationship, and I started doing some writing projects for them.

Then, one day, they said the host of their AI podcast had left and asked if I was interested. I said of course, but I wasn't sure if I knew enough about AI. They said don't worry. We'll give you stuff to study and do a trial run. We taped one that was too "chummy" and never aired. The second one went better, and we went from there. So, it was a case of personal relationships putting me in the right place at the right time when the gig opened up. The podcast has been one of the most fun things I've done and I'm happy to still be doing it!

Peter: I'm a big fan. How many episodes have you recorded now?

Noah: Thanks! I’ve lost track, but the podcast has over 200 episodes, as they had some other hosts early on and during COVID-19. We're creeping up on 4 million listens. Obviously, in the past two years, the inbound requests from people wanting to get on have increased quite a bit.

Peter: I can only imagine. I'm really curious to ask—as someone who hosted one of the quintessential AI podcasts before it was cool and buzzy—how have you seen the AI industry and space evolve over the past seven years?

Noah: Before 2023, a lot of what was happening with AI was stuff that everyday folk wouldn't know about but might be affecting their lives — like machine learning for analytics, computer vision in smart city applications, and researchers experimenting with deep learning models.

Over time, particularly once transformers were invented and generative AI emerged, these AI capabilities started coming more into mainstream consciousness. The evolution has been nonlinear but steady, with some big step-function leaps when there were major breakthroughs like transformer architecture and modern generative AI.

Peter: Let's dig in there a bit more. Most people in my network really started paying attention to AI in November 2022 when ChatGPT's model came out, and everyone could interact with a large language model, getting a sense of the state of the art. The last 18-24 months have been insane: AI model iterations, new entrants, big tech catching up, Nvidia's stock soaring, and pressure for more chips and data centers. As someone who has been courtside this whole time, I'd love to hear how it's impacted you and the podcast and how you're viewing this AI moment midway through 2024.

Noah: Wild times indeed. One thing we talk about on the podcast is the "democratization of tools." Early on, we had "bedroom hobbyists" who realized they had a GPU in their gaming PC and could start experimenting with AI models. There was a period where the software was catching up to the hardware to enable easier access.

Then it flipped, with GPU shortages during COVID-19 as crypto and AI demand spiked. Now, with the explosion of ChatGPT, Midjourney, Dall-E, Stable Diffusion, etc., anyone with a laptop and internet access can start using these tools. So, on the one hand, there is massive democratization of access to AI tools.

On the other hand, our listenership has increased dramatically, inbound guest requests have exploded, and my inbox is overflowing. Personally, as a freelance writer, in early 2023, I dove into trying to get AI to do as much of my work as possible. I wanted to see if I could automate my workflow or get the output of GPT close to something I could use. The answer was no—it wasn't quite there yet, at least with my ability to wrangle the models.

But I knew from my Nvidia work that this wasn't just hype. The tech has been advancing rapidly and isn't going away, even if we hit a hype bubble that pops. By mid-2024, we're in a phase where people and companies are trying to figure out how it can actually make an impact. The frontier is really society and organizations absorbing this leap in technology, not the tech itself, which will keep progressing one way or another.

Peter: Where do you see some of the biggest misconceptions about AI, especially among those in media?

Noah: One misconception is that AI will replace humans. Right now, there absolutely needs to be human oversight and review. Having domain expertise as a journalist, marketer, or filmmaker definitely still gives you a leg up on AI systems. The tech is augmenting what humans can do, not replacing us.

If AI is getting into more creative pursuits like music, video, and art, we're going to see new styles and genres emerge that explore what the tech can uniquely do. A lot of times, when new creative tech comes out, the first thing people do is try to replicate familiar things. But generative AI is going to open doors to imagine and create things we haven't seen before.

Peter: How do you recommend storytellers and creatives position themselves for success in an AI world?

Noah: First, get hands-on with AI tools to see firsthand what they can and can't do to assist in your work. I've found them really helpful for summarizing long texts, generating first-draft ideas, and brainstorming. Have an open mind but maintain a critical eye.

Second, develop deep domain expertise in your field, whether that's journalism, filmmaking, art, etc. Learn the foundations inside and out. That knowledge will always be an advantage, even as AI gets more sophisticated.

Finally, stay informed on the rapid developments happening in generative AI. Subscribe to newsletters and podcasts to keep a pulse on the industry. Go down rabbit holes related to your specific line of work to see how others are applying AI. Knowledge is power in navigating this terrain.

Don’t be shy—hit reply if you have thoughts or feedback. I’d love to connect with you!

Until next week,

Peter

Psst… Did someone forward this to you? Subscribe here!