Lucas and Luna break down why the price of running AI models — inference — is on the verge of a dramatic drop. With SK Hynix raising $26.5 billion in the biggest foreign IPO in US history and NVIDIA's stock surging 7.9% in the past five days, they connect the dots between memory manufacturing, chip demand, and the economics of deploying AI at scale. They explain how new memory technologies like HBM4 are cutting per-token costs, why inference is becoming the dominant AI workload, and what this means for startups racing to build AI-powered products. If you've wondered whether the AI boom is a bubble or a real shift, this episode gives you a concrete number to watch: the cost of a single inference call. No hype, just the math.

#AIInference #NVIDIA #SKHynix #HBM #ChipCosts #LLMOps #AIScale #MemoryTechnology #IPO #Semiconductors #Technology #FexingoBusiness #BusinessPodcast #AIEconomics #StartupStrategy #InferenceCosts #LucasAndLuna #FexingoPodcast

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna explore how the AI industry is shifting from model subscriptions to token-based pricing, arguing that tokens are becoming the new API keys of the software world. They dive into OpenAI's GPT-5.6 launch on July 9, 2026, and its impact on pricing models, using data points like Broadcom's 11.3% weekly gain and Microsoft's 1.6% decline to illustrate market reactions. The episode also touches on how this shift commoditizes developer tools, referencing the rise of AI coding assistants and the split between reasoning and retrieval models discussed in prior episodes, while keeping the focus on token economics as the central thesis.

#Tokens #APIEconomy #OpenAI #GPT56 #AIPricing #DeveloperTools #Broadcom #Microsoft #AICommoditization #SoftwareBusiness #Technology #Business #Podcast #FexingoBusiness #BusinessPodcast #TechTrends #GenerativeAI #AIModels

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna examine the rapidly commoditizing AI coding assistant market. With Meta releasing Muse Spark 1.1 into an already crowded field, they explore why the cost of AI-assisted code generation is plummeting and what that means for startups like GitHub Copilot and Cursor. Lucas points to Broadcom's recent 11% weekly surge and AMD's 5.6% gain as signals that the hardware underpinning these tools is becoming more valuable than the software layer itself. The conversation focuses on a single question: if every developer can now get AI coding help for free, what's left to pay for? A must-listen for engineers, product managers, and anyone watching the AI platform shift.

#AICoding #DeveloperTools #MuseSpark #Meta #GitHubCopilot #Cursor #Commoditization #Broadcom #AMD #Semiconductors #GenerativeAI #LargeLanguageModels #SoftwareDevelopment #TechTrends #Technology #FexingoBusiness #BusinessPodcast #ChatGPTandBeyond

Keep every episode free: buymeacoffee.com/fexingo

Episode 100 of ChatGPT and Beyond with Fexingo marks a milestone by exploring the quiet structural split happening inside today's large language models: the separation of reasoning from knowledge retrieval. Lucas and Luna break down why companies like NVIDIA and ARM are seeing divergent stock moves tied to this architectural shift, and how a new generation of 'hybrid RAG' systems is changing enterprise AI deployment. They discuss a specific case—a mid-sized logistics firm that cut inference costs by 40 percent by separating its model's reasoning and retrieval layers—and what that means for the broader AI stack. With live market data from July 9, 2026, the episode connects the dots between falling model costs, rising inference demand, and the companies positioned to win in a world where intelligence and memory are no longer bundled together.

#AI #LargeLanguageModels #Reasoning #Retrieval #RAG #EnterpriseAI #NVIDIA #ARM #SMCI #AIInference #ModelArchitecture #HybridAI #AICosts #TechInvesting #GenerativeAI #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

In this episode, Lucas and Luna explore the emerging robotics-AI convergence, sparked by a startup claiming robotics is about to have its 'ChatGPT moment.' They examine NVIDIA's recent 3.3% gain and ARM's 11% drop as signals of shifting investment dynamics, and discuss how foundation models for robotics could reshape hardware, software, and labor markets. With specific examples from autonomous drone delivery and warehouse automation, they unpack what a breakthrough in general-purpose robotics would mean for the economy and for investors.

#Robotics #AI #ChatGPTMoment #NVIDIA #ARM #AutonomousDrones #FoundationModels #IndustrialAutomation #Technology #Podcast #FexingoBusiness #BusinessPodcast #GenerativeAI #MachineLearning #RoboticsStartup #VCInvesting #Hardware #LaborMarket

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna dig into the shift from training AI models to running them in production—what the industry calls inference. With NVIDIA's five-day slide, AMD's sharp drop, and SambaNova's fresh billion-dollar raise, they unpack why inference is becoming the real battleground. They look at how a French startup called ZML just open-sourced a tool to speed inference across multiple chips, and what that means for the hardware giants. Plus: why ServiceNow's 11.5 percent weekly gain signals that software companies with AI agents are pulling ahead. No hype, just the numbers and the strategy.

#AIInference #NVIDIA #AMD #SambaNova #ZML #ServiceNow #ChipMarket #GenAI #HardwareShift #AIAgents #EnterpriseAI #TechStock #Semiconductor #OpenSourceAI #Technology #FexingoBusiness #BusinessPodcast #AIInvestment

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna break down how the rapid fall in AI model costs is creating a growing divide between software companies that control their own models and those that rely on third-party APIs. Using the latest live market data from July 7, 2026 — including ServiceNow's 11.5% weekly gain versus AMD's 11.2% drop — they explore why owning the model stack is suddenly a competitive advantage. They also look at Microsoft's recent shift toward its own models and what that means for the broader AI software landscape.

#AI #ArtificialIntelligence #LLM #Software #EnterpriseSoftware #ModelCosts #ServiceNow #Microsoft #AMD #NVIDIA #AIInfrastructure #TechStocks #FexingoBusiness #BusinessPodcast #Technology #GenerativeAI #OpenSource #AIAdoption

Keep every episode free: buymeacoffee.com/fexingo

This episode of ChatGPT and Beyond with Fexingo explores how the AI industry is splitting the concept of 'agent' from 'model' — and what that means for enterprise software. Hosts Lucas and Luna drill into ServiceNow's recent 8% five-day surge and Vercel CEO Guillermo Rauch's argument that decoupling agents from models unlocks faster innovation. They examine why ServiceNow now trades at 107 dollars and change per share while Palantir rallies 14% on agent-driven deals, and what the AI agent-model split means for developers, IT buyers, and the next wave of productivity tools. A sharp, focused look at the architectural shift redrawing the AI software map in mid-2026.

#AI #LLM #GenerativeAI #AIAgents #EnterpriseAI #ServiceNow #Palantir #Vercel #GuillermoRauch #ModelMarketplaces #SoftwarePricing #ProductivityTools #VibeCoding #Technology #AIInfrastructure #FexingoBusiness #BusinessPodcast #AISoftware

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna dive into Vercel CEO Guillermo Rauch's recent argument that AI agents should be decoupled from the underlying language models — a shift that could remake enterprise software pricing and architecture. They break down why this matters now, using ServiceNow's 8% weekly jump and Salesforce's 4.9% gain as evidence that the market is already betting on agent-first platforms. The hosts trace the logic from the 'model marketplace' trend they've covered before to this new frontier: if agents become independent, who captures the value? Plus, a concrete look at how vibe coding and falling inference costs are accelerating the split — and what it means for developers and CIOs planning their 2027 budgets.

#AI #GenerativeAI #AIAgents #LLM #Vercel #GuillermoRauch #EnterpriseSoftware #ServiceNow #Salesforce #VibeCoding #InferenceCosts #ModelMarketplaces #SoftwarePricing #Technology #AIInfrastructure #FexingoBusiness #BusinessPodcast #TechTrends

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna dive into the quiet revolution in AI software pricing, where model marketplaces like those from Meta and Mistral are allowing companies to pay per inference rather than per license. They examine how this shift is putting pressure on traditional SaaS giants like Salesforce and ServiceNow, with ServiceNow's stock up 8.1% over the past five days despite the broader trend. The hosts discuss a specific case: a mid-size logistics firm that cut its software bill by 40% by switching from a per-seat CRM to a per-inference AI assistant. They also explore the implications for Adobe, whose 8.4% gain this week may reflect investor optimism about AI-native pricing. The conversation touches on the rise of 'AI-native' startups and the potential for a commoditization of enterprise software.

#AI #ArtificialIntelligence #Technology #SaaS #SoftwarePricing #ModelMarketplaces #ServiceNow #Adobe #Salesforce #Meta #Mistral #PerInference #EnterpriseSoftware #SubscriptionModel #AIAdoption #FexingoBusiness #BusinessPodcast #TechTrends

Keep every episode free: buymeacoffee.com/fexingo