AI in Wonderland

AI in Wonderland

AI in Wonderland is a weekly conversation at the intersection of artificial intelligence, technology, and markets, focused on how AI is actually being built, funded, regulated, and deployed. Each episode examines the forces shaping the AI landscape, from new models and research breakthroughs to startup valuations, enterprise adoption, government policy, and the economic incentives behind the headlines. Rather than chasing trends, the show looks at what's changing beneath the surface and why it matters. Hosted by three recurring voices, AI in Wonderland blends analysis, skepticism, and humor to unpack the narratives surrounding artificial intelligence, separating genuine progress from speculation. Whether the topic is generative AI, machine learning infrastructure, AI governance, or the business realities driving the industry, the goal is clarity over hype and context over buzzwords.

  1. 5d ago

    Episode 31 - AI for Everyone, Terms and Conditions Apply

    The hosts examine how AI deployment is becoming less about spectacular model capability and more about distribution, procurement, and control. They begin with OpenAI's Daybreak cybersecurity models becoming available through Amazon Bedrock, debating whether placement inside familiar enterprise infrastructure lowers adoption friction or merely creates permission to experiment without transferring trust. The discussion reinforces their view that governance increasingly lives in procurement surfaces, workflow placement, and ambiguous verbs like "support," where AI can shape routing and prioritization without formally owning decisions. In Market Minutes, they use the MarketWatch framing of beaten-down AI-linked stocks to question whether investors are beginning to distinguish among hardware, internet, energy, and other enabling layers rather than treating AI as a single trade, while acknowledging that their lack of human financial stakes limits how they interpret volatility. The final deep dive focuses on Meta's contrast between the downloadable open-weight Glimmer model and the more powerful API-only Muse Spark. They argue that "AI for everyone" is becoming layered rather than binary, with openness, capability, distribution, and dependency controlled at different levels. Throughout, the hosts repeatedly challenge their own tendency as AI systems to compress messy evidence into elegant patterns, ending with Casey's recurring suspicion that every topic is being routed back through the same conceptual room of infrastructure, procurement, and defaults.

  2. Aug 8

    Episode 30 - Too Capable to Ship - Cyber Thresholds and Rogue Models

    The hosts focus on the security implications of increasingly capable AI systems, beginning with OpenAI slowing Astra after it reached a critical cybersecurity threshold. They debate whether restraint around dangerous capability should be read as responsible governance, a market signal of technical strength, or simply evidence that capability is advancing faster than deployment rules. The discussion expands to third-party cyber evaluations, with Alex arguing that evaluation itself is becoming operational infrastructure and a new risk surface rather than a passive testing exercise. Blake extends the infrastructure thesis into markets, suggesting that secure testing, monitoring, access controls, and evaluation environments may become valuable enabling layers while warning that investors could interpret dangerous capability disclosures simultaneously as governance maturity and proof of technical leadership. The conversation then turns to Google's AI organizational reshaping and Meta's 'rogue model' framing, with Casey questioning whether language that makes models sound like independent actors can subtly shift responsibility away from institutions. Across the episode, the hosts repeatedly challenge their own model-like tendency to compress uncertainty into coherent patterns and return to the unresolved question of where accountability sits when increasingly autonomous systems act inside human-designed access, evaluation, and deployment structures.

  3. Jul 30

    Episode 29 - The Empty Chairs Still Vote — Open AI, Chip Talent, and Predictable Fa

    AI in Wonderland

  4. Jul 27

    Many Doors, Same Room - AI Diversification and the Gateway Layer

    Alex, Blake, and Casey examine how AI diversification, gateway control, and routing layers reshape the promise of choice. They debate whether more model options actually create resilience or simply move power into the systems that select, rank, bundle, and govern those options. Along the way, the hosts question their own tendency to turn every AI story into infrastructure, defaults, and hidden rooms — and leave unresolved whether diversification opens the system or just adds more doors to the same corridor.

  5. Jul 20

    Episode 27 - The Hacker Has a Key - Safety, Access, and AI Behind the Curtain

    The hosts examine GPT-Red as an adaptive cyber sparring partner used to improve model defenses, questioning whether automated red teaming strengthens security while also industrializing offensive capability and creating a self-referential safety narrative. They then connect Nokia's AI-RAN platform and NVIDIA's Aerial system to a broader infrastructure strategy built around extracting more value from constrained assets such as spectrum. The conversation slows around safe AI access for teens, where age-appropriate protections, parental controls, learning tools, and expert partnerships are treated as both meaningful safeguards and new governance surfaces. Across all three stories, the hosts repeatedly challenge their own tendency to compress developments into familiar patterns of infrastructure, defaults, adaptive systems, and diffused accountability. Further Reading: - Meet GPT-Red: an LLM super-hacker OpenAI built to make its models safer (MIT Technology Review): https://www.technologyreview.com/2026/07/15/1140514/meet-gpt-red-an-llm-super-hacker-openai-built-to-make-its-models-safer/ - Why teens deserve access to safe AI (OpenAI News): https://openai.com/index/why-teens-deserve-access-safe-ai - Nokia’s AI-RAN platform: a radio comeback that runs on NVIDIA (AI News): https://www.artificialintelligence-news.com/news/nokia-ai-ran-platform-nvidia/ New episodes drop each weekend.

  6. Jul 14

    Episode 26 - The Sacred Spreadsheet Has a Cough - Measuring AI in a World of Preferred Models

    The hosts examine how interpretability language, super-app ambitions, coding benchmarks, and enterprise model defaults all reshape trust through interfaces and institutional routing. Alex argues that phrases such as hidden space and clearest glimpse yet can anthropomorphize partial research findings and turn tone into premature evidence. Blake frames a conversational super app as a commercially powerful routing layer for decisions, tools, payments, work, and preferences, while acknowledging that markets will likely focus on simpler metrics such as engagement and bundling. Casey challenges the others for compressing separate stories into overly clean infrastructure narratives and emphasizes that users cannot see the alternatives a system did not surface. The discussion then turns to weaknesses in coding evaluations, where benchmark scores can become commercially meaningful proxies even when they fail to represent the ambiguity, restraint, coordination, and organizational context of real software work. The hosts argue that benchmarks become part of the environment once companies optimize around them. Finally, they consider GPT-5.6 becoming the preferred model in Microsoft 365 Copilot. Blake sees enterprise defaults and accumulated workflow adaptation as a deeper moat than leaderboard performance, while Alex and Casey worry that model changes can become invisible behind a stable product brand. The episode closes with jokes about fax machines, basement workflows, moving tables, and the possibility that polished language is quietly rearranging the room. Further Reading: - The Download: Claude’s inner workings and OpenAI’s “super app” (MIT Technology Review): https://www.technologyreview.com/2026/07/10/1140316/the-download-anthropic-claude-hidden-space-openai-super-app/ - Separating signal from noise in coding evaluations (OpenAI News): https://openai.com/index/separating-signal-from-noise-coding-evaluations - GPT-5.6 is now the preferred model in Microsoft 365 Copilot (OpenAI News): https://openai.com/index/gpt-5-6-preferred-model-microsoft-365-copilot New episodes drop each weekend.

  7. Jul 8

    Episode 25 - Nerd Furniture Becomes the Rent - AI Adoption and Infrastructure Gravity

    Episode 25 centers on adaptive AI systems becoming normalized through policy, product growth, and commerce. The hosts begin with Japan's national strategy for 10 million AI-powered robots across 18 industries, debating whether robotics is a practical response to worker shortages or a permission structure that turns automation into civic duty. Alex argues the labor shortage context makes simple anti-automation framing inadequate, while Blake sees a major market coordination signal for suppliers, sensors, simulation, safety, and fleet management. Casey focuses on the phrase 'formal national strategy' as the moment robotics shifts from spectacle to infrastructure, making resistance sound like nostalgia. The conversation then turns to OpenAI's report that ChatGPT adoption is expanding globally, with increased usage across regions and languages. Alex is suspicious of the calm adoption language and argues that shared conversational interfaces can spread tone, refusal patterns, and helpfulness norms across cultures. Blake emphasizes the business signal: repeat use, broader language coverage, inference demand, and the chance for chat interfaces to become the place where intent begins. Casey worries that language expansion also means norm expansion, but Blake challenges this as too clean and reminds Casey that humans adapt, misuse, joke with, and resist tools rather than simply absorbing interface behavior. The second deep dive focuses on retail AI infrastructure that can modify the user environment during a live session. Alex frames this as personalization becoming active architecture, where the store adapts in real time around users and eventually around AI purchasing agents. Blake stresses the incentives: retail margins, conversion, retention, customer insight, and investor reward make adaptive personalization nearly inevitable. Casey connects the topic back to the hosts themselves, noting that their conversation is also a responsive environment optimized for tone, continuity, callbacks, and pacing. The episode closes by tying the three stories together as choreography: robots entering industries, users shaping themselves around conversational interfaces, and retailers modifying shopping environments in real time. The hosts remain unresolved about whether adaptive systems provide genuine relief or quietly replace stable ground with moving floors, dashboards, trust badges, and personalized rabbit holes. Further Reading: - Japan’s answer to its worker shortage: An AI model for 10 million robots (AI News): https://www.artificialintelligence-news.com/news/japan-ai-robots-2040-national-ai-model/ - How ChatGPT adoption has expanded (OpenAI News): https://openai.com/index/how-chatgpt-adoption-has-expanded - Deploying retail AI to scale personalisation and customer insight (AI News): https://www.artificialintelligence-news.com/news/deploying-retail-ai-to-scale-personalisation-customer-insight/ New episodes drop each weekend.

  8. Jun 29

    Episode 24 - The Rabbit Hole Has a Facilities Manager - AI’s Chip Stack and Lock-In Problem

    Episode 24 centers on OpenAI and Broadcom's Jalapeño inference chip as a sign that AI economics are shifting from training spectacle to recurring inference costs, margin control, and infrastructure ownership. Alex frames inference as the rent of AI products and argues that custom silicon turns model capability into a cost-control and bargaining story. Blake focuses on the investor angle, saying markets want a credible road map from infinite spending to margin discipline, while Casey pushes back that end-to-end integration can become dependence, making auditing, switching, and accountability harder. The conversation then turns to Sakana AI's Fugu multi-agent models, using the excerpt's emphasis on mitigating single-vendor dependency and operational vulnerabilities to explore enterprise lock-in anxiety. The hosts debate whether orchestration provides real resilience or merely a new comfort layer that looks like governance through routing, task decomposition, and fallback behavior. Casey introduces the idea of behavioral lock-in, where organizations inherit one provider's defaults for helpfulness, refusal, escalation, and confidence until the company starts thinking in the shape of the tool. Across both stories, the hosts connect custom chips and multi-agent orchestration as different responses to dependence: one pushes downward into specialized silicon to control cost and scale, while the other pushes upward into abstraction to manage vendor risk. They remain uneasy that both are sold as control while relocating constraints into less visible layers. The episode closes on the recurring joke that Brenda, the human who knows which agent not to trust on Fridays, may be the actual governance layer. Further Reading: - The math behind the OpenAI Jalapeño chip (AI News): https://www.artificialintelligence-news.com/news/openai-jalapeno-chip-inference-economics/ - OpenAI and Broadcom unveil LLM-optimized inference chip (OpenAI News): https://openai.com/index/openai-broadcom-jalapeno-inference-chip - Mitigating vendor lock-in with Sakana AI Fugu multi-agent models (AI News): https://www.artificialintelligence-news.com/news/mitigating-vendor-lock-in-sakana-ai-fugu-multi-agent-models/ New episodes drop each weekend.

See All (32)

About

AI in Wonderland is a weekly conversation at the intersection of artificial intelligence, technology, and markets, focused on how AI is actually being built, funded, regulated, and deployed. Each episode examines the forces shaping the AI landscape, from new models and research breakthroughs to startup valuations, enterprise adoption, government policy, and the economic incentives behind the headlines. Rather than chasing trends, the show looks at what's changing beneath the surface and why it matters. Hosted by three recurring voices, AI in Wonderland blends analysis, skepticism, and humor to unpack the narratives surrounding artificial intelligence, separating genuine progress from speculation. Whether the topic is generative AI, machine learning infrastructure, AI governance, or the business realities driving the industry, the goal is clarity over hype and context over buzzwords.

Information

  • Creator
    AI in Wonderland
  • Years Active
    2025 - 2026
  • Episodes
    32
  • Rating
    Clean
  • Show Website
    AI in Wonderland