This Day in AI Podcast: Are We In An AI Bubble? In Defense of Sam Altman & AI in The Enterprise | EP99.24

Hosts: Michael Sharkey, Chris Sharkey
Date: November 7, 2025
Episode Duration: ~1 hour

Episode Overview

Michael and Chris Sharkey tackle the question of whether the AI industry is caught in a speculative bubble, dissect the latest drama and criticism around OpenAI and Sam Altman, and discuss the evolving landscape of AI adoption in the enterprise. They offer their own “proudly average” takes, highlight recent data on market share, analyze enterprise challenges, and review the new Kimi K2 model. The tone is witty and self-deprecating, with plenty of sibling banter, but the analysis is well-informed and practical.

Key Discussion Points & Insights

1. Housekeeping, Carrot Merch, and Show Banter

• [00:24] Chris addresses complaints about his bookshelf and sports “I dig carrots” merch from Bolthouse Fresh – not a podcast sponsor, just eager for free stuff.
  • Chris: “Imagine that. Funny that like we're an AI podcast and the first sponsored thing we've ever done is carrots... It's just like really me being desperate for free clothes.” [02:24]

2. Sam Altman & OpenAI Boardroom Saga

• Recent “pile-on” against Sam Altman:
  • A tense exchange on the B2B Podcast between Sam and investor Brad Gerstner gets sampled and discussed.
    • Mike (playing Altman): “First of all, we're doing well. More revenue than that.” [03:04]
    • Chris: “That is a slap in the face ... what a bold response. I love it.” [03:15]
• Helen Toner’s criticism:
  • Former OpenAI board member Helen Toner accused Altman of deceiving the board ([03:44]).
    • Helen Toner: “Sam had made it really difficult for the board to actually do that job by, you know, withholding information, misrepresenting things ... in some cases outright lying.”
    • Chris is unequivocally dismissive: “This woman is a disgrace ... Like, you should get behind the company you're a director of.” [04:06]
• Podcast hosts' take:
  • View the OpenAI turmoil as typical high-stakes “gossipy internal politics and jealousy and backstabbing,” sparked by spectacular success more than any moral failings ([05:00]–[06:00]).
  • Chris draws parallels to Bezos/Amazon and Musk/Tesla:
    • “Everyone is jealous of the success this guy has created ... would you really bet against OpenAI at this point?” [06:02]
  • Mike (summarizing): “He stood by OpenAI after Elon Musk ... He had the foresight to say, yeah, just put ChatGPT out there ... That really created the AI explosion.” [07:32]
  • Chris: “You can’t on one hand say we're for business ... but also we've made it so you can make anime porn to share with your friends.” [19:49]
• Critique of OpenAI’s current direction:
  • Hosts note OpenAI’s perceived lack of focus and mounting competition—specifically, enterprise market share loss to Anthropic and Google.
    • Mike: “OpenAI fell from 50% in late 2023 to 25% by mid 2024 ... Anthropic now leads with 32%.” [15:57]
  • Chris: “I don’t even think they have the best model ... they lack vision, they lack cohesion.” [17:01]

3. AI Enterprise Market Realities

• Market share trends:
  • The hosts examine OpenAI's shift from consumer to enterprise focus and the importance of strategic market segmentation ([18:29]–[19:01]).
  • Mike: “Really, you should just see these models as tools ... not lock into a singular model.” [18:43]
• Product/brand confusion:
  • The podcast highlights OpenAI’s "mixed signaling" and product sprawl as business risks.
• Anthropic’s rise:
  • Mike: “Anthropic’s positioning themselves as the enterprise one.” [18:34]

4. Are We in an AI Bubble? The Financial Context

• [21:09] AI-driven market stats:
  • Nvidia is now worth $4.5 trillion—more than double Australia’s economy, ~16% of the entire US GDP.
  • $1.2 trillion in AI-related corporate debt in the system.
  • Chris: “None of this is real, is it? Like, money doesn’t exist. This is just all just numbers on a GPU somewhere.” [21:46]
  • Average market trades at 30x earnings; S&P 500 long-term average is 17x ([22:23]).
• Circular financing ([23:37]):
  • Describes the loop of massive investments pinging between Nvidia, OpenAI, and Microsoft, inflating earnings and valuations.
    • Mike: “Nvidia invests $100B in OpenAI, who then spend $100B buying Nvidia chips ... it’s this loop.” [24:36]
• Are these unsustainable shenanigans or actual productivity drivers?
  • Chris: “I don’t [think it’s a bubble], for the reason I said earlier, because companies have budgets for this stuff, all of them ... the demand's there.” [24:55]
• Host perspective:
  • Despite speculation and circular financing, the hosts believe real, lasting productivity gains are being unlocked across the global economy—from law firms to local governments ([27:07]).
  • Mike: “51% of SMBs that adopted generative AI reported revenue increases of 10% or more.” [28:20]
  • Chris: “It’s bringing smart people alive ... they suddenly have capabilities they didn’t have before.” [29:04]

5. The “Bubble” vs. Lived Experience

• Mike: “MIT said that 95% of organizations are seeing zero measurable return.” [30:14]
• Chris: “Yeah but ... when the Internet came out, newspapers were like, ‘Oh, people still prefer to get the newspaper delivered’ ... you keep believing that, bro.” [30:20]
• Hype cycles and adoption realities:
  • Hosts worry that AGI hype sets the industry up for disappointment.
    • Mike: “Anything they release is seen through the lens of those [AGI] statements ... you can never meet or live up to those expectations.” [33:20]
  • But see real value and success stories emerging; the “trough of disillusionment” will weed out laggards ([35:25]).
  • Chris: “The real gatekeepers to the future of AI: the IT dudes ... Become really good friends with the IT department ... because you’re going to need their help.” [43:25]

6. Technical Deep-Dive: AI Agents, Tool Use, and Model Tuning

• Context bloat and agent efficiency:
  • Mike details Anthropic’s post about more efficient agent architectures—i.e., not sending every tool to the model in every context window ([45:10]).
  • Chris isn’t convinced about code-execution approaches for tool selection:
    • “I don’t agree that writing code with the model and then executing is the solution ... you want them to have a holistic picture of everything they're able to do ... that's where advancements should go.” [45:10–49:11]
  • Discuss trade-offs in using skills, agents, or direct user control (for now), rather than full “magical AGI black box” approaches ([51:29]).
• Specialist assistants and agentic workflows:
  • Practical advice: Turn on/off MCPs (tools) based on your use—let the user “switch hats” as needed ([50:41]–[51:29]).
  • Chris: “The next step will be ... you have an agentic model where you are setting goals and then it is able to look at the resources it has available and maybe create those roles or those sets of MCPs ...” [51:38]
  • Both expect rapid advances here but see current optimization as mostly premature.

7. Model Review: Kimi K2

• Initial impressions:
  • Kimi K2 (“Thinking”) boasts 256k context window, 16k outputs, strong on “agentic reasoning” and code ([54:52]–[55:08]).
  • Mike: “It seems like a great coding model ... in terms of using MCPs and agentic workflows ... it’s absolutely, stupidly, shockingly bad.” [55:08]
  • Chris: “Why would I use this over Haiku? ... Maybe coincidentally it’s better at horse racing ... but only by total fluke.” [57:07]
• **Kimi K2’s marketing claims (up to 300 sequential tool calls) are met with deep skepticism; real-world usage does not match blog post hype ([59:08]).
• Song about the “bubble”:
  • Mike closes with a self-written song “In the Middle of the Bubble,” humorously recapping financial exuberance with catchy, AI-themed lyrics ([61:01]).

Notable Quotes & Memorable Moments

On OpenAI drama:
Chris: “This woman is a disgrace... there is nothing worse than a snively little person who waits until there's a pile-on and then reveals this information...” [04:06]

On the speculative bubble:
Chris: “None of this is real, is it? Like, money doesn’t exist. This is just all just numbers in a book. Not even a book. Numbers on a GPU somewhere.” [21:46]

On productivity gains:
Mike: “51% of SMBs that adopted generative AI reported revenue increases of 10% or more.” [28:20]
Chris: “It’s bringing smart people alive. They suddenly have capabilities they didn’t have before...” [29:04]

On AI agents and tool use:
Chris: “You want to use [AI models] for their intelligence. You want them to have a holistic picture of everything they're able to do... I think the goal [of AI] should be, ‘what can I produce with this?’” [45:10–46:50]

On enterprise adoption:
Mike: “A lot of these early adopters have got absolutely spanked by it. They don’t know how to measure the results, they’re not spending as much, they’re disappointed, the pilots aren’t going ahead because they’ve failed.” [33:20]

On model glut:
Mike: “I think some of these models, the claims versus the reality, it's getting a bit exhausting to even have to test them.” [57:18]

Advice to listeners:
Chris: “Here’s my thoughts before you knock something. Try it. Get out there and use this stuff. It’s exciting and fun and good and it’ll help you. The big bubble markets, it doesn’t matter. Just use it.” [60:10]

Timestamps for Major Segments

• [00:24]–[02:35] — Show banter and carrot merch
• [02:35]–[14:32] — OpenAI/Sam Altman board drama & host perspectives
• [15:57]–[20:02] — Market share shifts and OpenAI’s business focus
• [21:09]–[27:07] — The “AI bubble”: finance, hype, real-world adoption, and statistics
• [27:07]–[33:20] — Enterprise AI transformation & barriers to value
• [35:25]–[43:25] — Hype cycles, failed pilots, the role of IT in enterprise AI adoption
• [45:10]–[52:43] — Technical discussion: AI agents, context, tools, and optimization
• [54:52]–[59:08] — Kimi K2 model review & skepticism
• [61:01] — “In the Middle of the Bubble” (podcast song)

Final Thoughts

• Despite market exuberance ("bubble?" maybe), fundamental productivity gains and transformative potential in AI look real to the Sharkey brothers, especially among smaller and nimble organizations.
• OpenAI’s spectacular drama and strategic choices are less important than the practical adoption of AI in daily work and business.
• Enterprises and individuals should focus on pragmatic use, not hype or market panics; find what works and use it.
• Model development is hot as ever, but be skeptical of blog post claims—plug-and-play usability trumps benchmarks.

Recommended Segments:

• OpenAI drama & Altman defense: [02:35]–[14:32]
• AI bubble discussion: [21:09]–[33:20]
• Model review/test drive: [54:52]–[59:08]

The show remains an entertaining, highly approachable listen for anyone keeping up with the fast-evolving (and sometimes chaotic) world of AI.