Jul 20, 2026 · 58m · big-technology
Kimi K3 & AI’s Price War, What’s Happening To Google?, OpenAI’s Partner Trouble
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this episode of the Big Technology Podcast, Alex Kantrowitz and Ranjan Roy analyze how the launch of Moonshot AI’s Kimi K3 model is driving industry-wide AI commoditization, while evaluating Google’s internal Gemini delays and Apple’s trade secret lawsuit against OpenAI.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Alex holds 57% of the talking time here. How this is scored →
Alex as informed peer 7.3 Guest teaching 1.7 Guest disagreement 1.9 Alex pushing back 2.9
speaking balance: gold is Alex, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 37:37 Guest pressing host on regulatory definitions
Ranjan directly pushes back on Alex's framing, forcing him to separate purely political interference from genuine AI safety rules.
Hardest push from Alex ▶ 39:30 Host rejecting political model bansAlex firmly distinguishes between disciplined staged model rollouts and unjustified governmental bans enacted without demonstrable cyber risks.
Biggest teaching moment ▶ 27:00 Guest explaining partner risk from model verticalizationRanjan explains how foundation model creators building competing apps like Claude Design and Claude Code will inevitably cause clients like Figma and Cursor to flee to open-weight models.
Alex holds their own ▶ 24:51 Host citing Dario Amodei and Boris Cherny interviewsAlex demonstrates authoritative expertise by citing his own consecutive exclusive interviews with Anthropic executives to prove their operational pivot from API reliance to native products.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Alex as informed peer | Guest teaching | Guest disagreement | Alex pushing back | Why |
|---|---|---|---|---|---|---|
| Moonshot's Kimi K3 Launch and the AI Model Pricing Battle | 7 | 2 | 1 | 2 | Alex anchors the segment with reporting on Kimi K3, comparing parameter counts, benchmark standings against Fable and Opus 4.8, and specific pricing from competitors like Grok 4.5 and Meta MuseSpark. Ranjan agrees with the commoditization thesis and adds per-token pricing details. | |
| The Gavin Baker Thesis and Collapsing Inference Margins | 8 | 1 | 2 | 3 | Alex introduces Gavin Baker's inference margin analysis and combines it with Ryan Greenblatt's open-weights timeline and past interview insights from Meta's Nick Clegg. Ranjan collaboratively explores the declining belief in closed frontier AGI moats. | |
| Vertical Product Strategy and Partner Data Protection Risks | 8 | 2 | 2 | 4 | Alex leverages his direct reporting and interviews with Dario Amodei and Boris Cherny to demonstrate Anthropic's tenfold revenue growth via native products like Claude Code rather than raw APIs. Ranjan argues that replicating partners like Cursor and Figma will eventually alienate enterprise customers. | |
| Satya Nadella's Information Paradox and Cloud AI Economics | 7 | 2 | 2 | 2 | Alex reads and contextualizes Satya Nadella's reverse information paradox post before formulating a hypothetical Andy Jassy scenario showing how hyperscalers profit regardless of lab insolvencies. Ranjan praises the economic framing and adds observations about synthetic data training. | |
| AI Regulation, Geopolitical Policy, and the David Sacks Debate | 7 | 2 | 3 | 4 | Alex points out the contradiction in David Sacks advocating permissionless innovation after his administration's interventionist ban on Fable. Ranjan presses Alex to define proper regulation versus political overreach, prompting Alex to defend phased glasswing rollouts. | |
| Google's Gemini 3.5 Delays and Organizational Bottlenecks | 7 | 2 | 2 | 3 | Alex diagnoses Google's Gemini 3.5 delay by tracing its over-reliance on small Flash models post-DeepSeek and questioning why leadership allowed internal teams to run out of compute while licensing it to third parties. Ranjan highlights organizational friction among DeepMind, Cloud, and Android teams. | |
| Apple's Espionage Lawsuit Against OpenAI and Strained Alliances | 7 | 1 | 1 | 2 | Alex breaks down Apple's corporate espionage suit against OpenAI and synthesizes a broader thesis about OpenAI systematically turning essential partners into adversaries. The conversation finishes in a collaborative, comedic riff on Apple hardware chief John Ternus. |
Statements from this episode (14)
Kantrowitz: China's open-source AI is at most four to five months behind
“It seems to show that China's open source movement is not the 10 or 15 months behind US AI, but maybe four or five months at the very, at the most, maybe even less than that.”
Roy: Frontier AI models are unnecessary for most agentified tasks
“Is our frontier models required for the majority of tasks that are going to be done and become agentified? And I don't think they are.”
Roy: Kimi K3 launch will destroy the frontier model moat narrative
“I think this is actually going to be a moment where even more so the idea that Frontier models are a massive moat really goes away.”
Roy: Kimi K3 is 40% cheaper than GPT-5.6 and 70% cheaper than Fable
“Three dollars per million token input, 15 dollars output. It's 40% cheaper than GPT, 5.6, 70% cheaper than Fable.”
Roy: OpenAI gets hit harder than anyone by Kimi K3
“So I think OpenAI's story and their pivot gets hit harder than anyone else with this announcement. Because suddenly, why OpenAI becomes a much more salient question than it was just a week ago.”
Kantrowitz: AI frontier intelligence will commoditize into a price war
“If you have five companies that have this level of intelligence, it's a price war. It commoditizes. So I think we should assume, and we've talked about this on the show, that intelligence is gonna commoditize.”
Kantrowitz: Model commoditization forces OpenAI and Anthropic to win on product
“Now if intelligence is abundant and available to be accessed through multiple providers, it will come down to who builds the best product. And so, it goes from a two-person race, right, or a two-company or a three-company race, OpenAI, Anthropic, maybe Google,…”
Roy: Partners may avoid Anthropic models over fears of product replication
“Now with that option, if you have any fear that they're going to take everything that you're giving them because you're using their models and then recreate your business, you now have an option. And I think that's gonna, that can slow things. Again, the speed…”
Roy: Non-enterprise training data is a key flywheel for OpenAI and Anthropic
“I mean, I think in Anthropic and opening, I are very open that All non-enterprise data, at least in theory, is not, is used for training, and that is one of their biggest advantages, and it's kind of that flywheel”
Kantrowitz: David Sacks pushing permissionless AI is ironic after Fable ban
“I appreciate a lot of what David Sachs has to say, and I've invited him on this show but this is, like, hilarious, and so rich, and ironic, that he being part of the Trump administration, which did, like, the most interventionalist policy to ban fable, is now …”
Kantrowitz: Fable lacked cybersecurity capabilities beyond Anthropic's Claude Opus
“It wasn't like Fable had cyber capabilities beyond, that weren't available, you know, from like Opus.”
Roy: Google remains far ahead of OpenAI in multimodal AI capabilities
“I'll say, I'll give them credit that still on like multimodality image and video, they still kind of own it right now. And everyone kind of puts them far in a way like I do. I really never hear that much about chat GBT image too, or, Any of these others, like …”
Kantrowitz: Apple sent document preservation letters to dozens of OpenAI employees
“So the latest is that Apple has now sent dozens of OpenAI employees legal letters asking them to preserve their documents, and Apple has also said that this, they only found the tip of the iceberg.”
Kantrowitz: OpenAI alienating tech allies will eventually cause severe consequences
“OpenAI's loss of these friends over time is going to add up to something really bad. Who knows what that exactly is, but there's gonna come a time where OpenAI's gonna need Satya, or it will need John Ternus, or it will need Elon. And they won't be there for t…”