Episode Summary
Executive Summary: The episode centers on OpenAI’s decision to delay GPT-5 while pushing product improvements like SearchGPT, and debates whether foundation-model leadership or product execution will matter more. The hosts also examine big tech’s massive AI capex, Amazon’s troubled Alexa reboot, Meta’s embrace of AI-generated content, Apple’s early iPhone/AI traction, and the rise—and manipulation risks—of election betting markets.
Main Topics: OpenAI delays GPT-5 and shifts emphasis (Priority: 5/5): Sam Altman says GPT-5 will not ship this year. The hosts debate whether this reflects compute constraints, competitive pressure from Meta/xAI, or a strategic pivot toward productization over hype. SearchGPT versus Perplexity and the future of search (Priority: 5/5): SearchGPT is framed as OpenAI’s most meaningful near-term product because it competes directly with Perplexity in the emerging middle ground between classic Google search and pure chatbots. Big tech’s AI capex arms race (Priority: 5/5): Amazon, Alphabet, Meta, and Microsoft are all increasing spending on AI infrastructure despite layoffs and efficiency messaging, showing that compute is now the core strategic bet across the sector. Alexa’s AI reboot struggles (Priority: 4/5): A Bloomberg report on Amazon’s AI-enhanced Alexa highlights severe product problems: inaccurate answers, weak reliability, and difficulty preserving basic device functions, suggesting organizational issues at Amazon. Meta’s AI slop strategy (Priority: 4/5): Zuckerberg’s comments about AI-generated and AI-summarized content signal Meta’s willingness to let AI content dominate feeds, which the hosts link to the rise of low-quality 'AI slop' on Facebook. Apple’s earnings and Apple Intelligence adoption (Priority: 3/5): Apple reported stronger iPhone sales and early adoption of iOS 18.1/Apple Intelligence, but the discussion treats this as early proof of user interest rather than a full validation of Apple’s AI strategy. Election betting markets and Polymarket (Priority: 5/5): The hosts discuss Polymarket and Kalshi as increasingly visible prediction markets, but worry about market manipulation, thin liquidity, whale influence, and the danger of turning election forecasts into spectacle.
Key Arguments: OpenAI may be prioritizing productization because the next frontier model may not create a large enough visible leap to justify GPT-5 hype. A stronger competitive response from xAI and Meta could be forcing OpenAI to rethink how much compute it needs before launching GPT-5. SearchGPT matters because it competes in the practical search-plus-citations layer where users actually want utility, not just chatbot responses. Google remains structurally advantaged in search because of its browser/platform position, even as new AI search products gain traction. Big tech’s enormous AI capex shows companies view infrastructure as the decisive battleground even while the consumer products remain uneven. Alexa illustrates how hard it is to bolt generative AI onto legacy consumer products without losing reliability and core functionality. Meta’s willingness to promote AI-generated content suggests that engagement incentives may override concerns about quality, authenticity, or misinformation. Election betting markets may offer predictive value in theory, but in practice their small size and weak regulation make them vulnerable to manipulation and public misinterpretation.
Data Points: GPT-5 release timing: Not coming this year - Sam Altman said OpenAI will have good releases later this year, but none called GPT-5. OpenAI fundraising round: $6.6 billion - Referenced as the largest VC round in history and part of OpenAI’s need to deliver revenue. OpenAI training loss projection: $3 billion - The hosts note OpenAI is expected to lose this much on training costs this year. OpenAI compute cluster comparison: 100,000+ H100s - Meta’s Llama 4 training cluster is described as larger than 100,000 NVIDIA H100s. Amazon 2024 capex: $75 billion - Amazon said it will spend a record amount on capital expenditures in 2024. Amazon 2023 capex: $48 billion - Used as the prior-year comparison for Amazon’s 2024 capex increase. Alphabet capex 2024: $51 billion - Alphabet’s projected spending on AI and related infrastructure this year. Google revenue growth: 15% to $88.3 billion - Referenced from earnings reporting for Alphabet. Google quarterly profit growth: 34% - Alphabet’s quarterly profit growth driven by advertising and cloud. Microsoft Copilot adoption in surveyed firms: 79% - CNBC survey result cited in discussion of enterprise AI use. Microsoft Copilot perceived worth it: 25% - Only a quarter of survey respondents said Copilot was worth it. Microsoft Copilot not worth it: 25% - A quarter said it was not worth it. Microsoft Copilot unsure: 50% - Half of respondents were unsure if the tool was worth the cost. Polymarket election wagers: $2 billion in October 2024 - Amount wagered on Polymarket election-related contracts. Kalshi election-related bets: Over $100 million - Amount of election-related bets mentioned on Kalshi. Whale trading activity: 71 bets a minute - A study cited one trader placing frequent bets, mostly on Trump. Apple iPhone revenue: Up 6% - Apple’s reported revenue increase tied to iPhone performance. iPhone 16 vs iPhone 15 sales: Stronger - Cook said the iPhone 16 was selling better than the 15, and the 15 better than the 14. iOS 18.1 adoption: 2x the rate of 17.1 - Early three-day data cited by Tim Cook for Apple Intelligence-enabled OS adoption.
Pivotal Quotes: "all these models have gotten quite complex, and we still can't ship as many things in parallel as we'd like to" — Sam Altman: Altman’s Reddit AMA explanation for why GPT-5 will not arrive this year. "I think that now they're going head to head with perplexity" — Ranjan Roy: Assessment of SearchGPT as OpenAI’s most important near-term product move. "there could be moments where these small groups of people are impacted" — Sundar Pichai: Google employee Q&A, cited during discussion of layoffs and the company’s AI investments.
Implications: AI competition is shifting from model hype to product usefulness, while infra spending keeps rising. Search, voice assistants, and feeds will be the near-term battlegrounds, but trust, reliability, and regulation will determine who benefits long term.
From the Transcript
And they made two this week. So let's talk a little bit about it. So, to me, the most puzzling thing we've seen from OpenAI this week is the fact that Sam Altman went on Reddit, did an AMA, and told all the Redditors, the Reddit hopeful, the Reddit faithful, shall we say, that there's not going to be any GPT-5 this year. Remember, OpenAI months ago talked about how it started training GPT-5, but no GPT-5 at the moment. This is what he says: he says, all these models have gotten quite complex, and we Still can't ship as many things in parallel as we'd like to. He says the company has limitations and hard decisions when it comes to allocating compute resources toward many great ideas. He says OpenAI is going to have some very good releases coming later this year, but nothing that we're going to call GPT-5. I have some thoughts on why we haven't seen GPT-5, but I'm curious what your perspective is. Yeah, I think the bullish.
I have tried it, and I think it's a big moment. And I think that now they're going head to head with perplexity. And in the whole area of search, I've been thinking about this a lot. There's kind of traditional search in the Google sense: you have a search bar and you type in some loose words to get a general direction and list of websites. Then there's a more traditional kind of generative AI chat GPT query where you ask a long question and you try to get really condensed information. And then there's something in between. And I think perplexity has been the leader in that in-between space where you want some information, but you also want some links out. And perplexity, I honestly, the related questions, the links out, that has become my search. And OpenAI playing in that space, I think, is very, very interesting. So basically, right now, there's a little like globe icon at the bottom of the search bar, the entry bar in ChatGPT.
I'm so happy about that because one of my wishes in life, I think all great human drama should take place on Halloween. So you have to be completely serious while in costume. Unbelievable. It's always, it just adds such an element to anything. So this is what Sundar said while in his Aero 404 costume not found shirt. It's like at the scale of our company, there could be moments where these small groups of people are impacted. Now, look, let's be fair to him. It's not like He's making a layoff announcement in his costume or whatever it is. But it is just interesting that, like, it just shows you what the company has been through with the layoffs over the past couple of years, that there's still some trauma within the workforce. And even in moments of like lightness where like executives are talking with employees, you're going to have this come up. And just to have it play out as they're wearing their costumes, I just.
About Big Technology Podcast
The Big Technology Podcast takes you behind the scenes in the tech world featuring interviews with plugged-in insiders and outside agitators. Alex Kantrowitz, a Silicon Valley journalist who's interviewed the world's top tech CEOs — from Mark Zuckerberg to Larry Ellison — is the host.