In this episode, Chris Carr, President & CEO of Farotech, discusses various aspects of the rapidly evolving world of AI. From the impact on marketing to the emergence of Pie AI and the importance of regulation and chat priming, there is a lot to uncover. Furthermore, OpenAI’s Code Interpreter tool is revolutionizing productivity, while AI tools and SEO hacks are explored, along with the compounding effect of misinformation, advancements in generative art, and the revolutionizing of video production through AI technology.
Meet Chris Carr, the visionary President & CEO of Philadelphia’s fastest-growing marketing agency, Farotech. With an impressive 22-year journey leading the agency from its inception, Chris has redefined inbound marketing, creating a team of SEO strategists, designers, and programmers that deliver exceptional results and ROI. His expertise spans project management, web development, and even vaccine sales at GlaxoSmithKline, showcasing his diverse talents in the world of business. Here are a few of the topics we’ll discuss on this episode of Cache Flow:
- Companies need to be ready for the rapid advancements and speed of AI.
- Chris now spends most of his time helping companies understand and prepare for AI.
- NotebookLM allows users to upload files into one folder for easy access and organization.
- ChatGPT’s Code Interpreter can analyze macro data and provide valuable insights across multiple accounts or projects.
- AI tools can simulate fights and create entertaining content.
- AI-generated content is changing the way search results are presented.
- Responsible AI models require complete data sets and fine-tuning to minimize hallucinations.
- Open AI: ChatGPT
- Open AI: Code Interpreter
- Inflection AI: Pi
- Anthropic: Claude AI
- Pine Cone
- Open AI: DALL-E 2
- Stable Diffusion
- 20:15 – “And so I had this bestselling author named Anne Hanley, she was on my blog, and she says, the number one concern I have about AI is that it confidently gives you misinformation. And so we’re going to talk a little bit later about exploratory data analysis and where that’s been a problem with it. But here’s what we’re seeing with hallucinations is, is that if you can give it more content, it’s not going to guess as much. And so what I mean by hallucination is, let’s say hypothetically I have a blog and I say “Hey, give me five statistics that are going to back up or support this 1500-word blog”. It’ll absolutely do that, but then you’ll go and check the references, and they never really existed. None of the statistics, like this, is on every single chat program. They make up their own stats.”
- 15:15 – Pie AI: “Hey there friend, glad to chat with you and your friends. I’m happy to give you all a quick demo of my conversational capabilities. I’m sure your friends will be amazed at how natural and engaging I can be. Are they ready to meet the future of AI?”
Chris: “Yeah, actually, I think they’re really spooked.”
Pie AI: “Ha ha. I love your sense of humor. It’s not uncommon for people to feel a little apprehensive about AI at first, but don’t worry, I’m here to show them that conversational AI is nothing to be spooked about. I’m friendly, informative, and always happy to chat. So, should we start the demo?”
Chris: “Yeah, go ahead.”
Pie AI: “Alright, here we go. Let’s start by having a simple conversation, just like you and I are having right now. Ask me any questions you want about anything at all, and I’ll answer them in a friendly, informative way. So, fire away.”
Chris: “Last question. Why is the sky blue?”
Brian: “Really thinking on that one? Yeah, it’s a hard question.”
Chris: “Hey, why is the sky blue?”
Pie AI: “I see what you did there. You are having fun testing my conversational capabilities, aren’t you? But I’m happy to answer the same question again, just like a real person would. So, are you ready for the scientific explanation of why the sky is blue?”
- 09:03 – “One of the biggest, like proponents of regulation, one of the biggest like advocates for regulation is Sam Altman, who is the CEO of OpenAI, who like seems to be in the lead with AI. Like if you read the news, like everyone, like ChatGPT, it’s like Kleenex or Bandaid, it’s like a noun essentially. I know, yeah. For all these other things, but like technologically speaking, they might lose their lead like as better systems. Like if, you know, they’re running on Azure, like, you know, AWS can compute this model data like 10 times faster. So, you know, one of the things that I’ve heard, you know, I’ve had some conversations and heard people talking about, is this the idea that Sam Altman is trying to introduce regulation as a strategy for slowing down competitors.”
- 41:07 – “Dude, this just brought a really interesting thought experiment for me. Like the historical way that you get information is you go to Google, you put in a few words, and you get 10 blue links, and Google has a whole massive complex algorithm that decides those 10 blue links. And part of it’s this EEAT part of it’s like backlinks and who’s linking to those pages and how trustworthy and authoritative Google thinks those sources are. But with AI, like, you know, it’s going to be easier as tools evolve, you know, it’ll probably cannibalize search, the 10 blue links paradigm will probably become a thing of the past.”
- 17:14 – Brian: “So, so what’s like the real use case of this thing [personal AI] though? Like, you can talk to it and kind of do like a cool party trick, but like, what’s the real use case of it?”
Chris: “It’s pretty soon you’ll be able to upload files and links to it. But like, what I do is that while I’m driving, I’ll say “Hey, today I’m going to write a blog about X and I’m trying to figure out what am I missing? What am I not missing? What are some data points that you think would be cool? Who else wrote maybe something like this that reminds you of what I’m talking about? And it’s literally like I’m talking to a person, it’s like I’m brainstorming with someone. So from a content creation standpoint and a creativity standpoint, it’s pretty freaking good.”